(note : To read parts of this text you must download, free of charge,
Acrobat reader).

ABSTRACT : In France, the end of the 1980s witnessed the rise of an unexpected orthodoxy concerning regional planning. According to it, the growth of a megapolis around Paris (in the region Île-de-France) would be the only card in the hands of French productivist planners in order to resist the marginalization of France within Europe. This orthodoxy expressed a quasi-consensus reflected in the debates about the new Schema Directeur Regional de l’Île de France (SDRIF), the regional plan for the Paris region over the next quarter of a century. Yet the orthodoxy was criticized from two sides: the theoretical disadvantages of a megapolis, and the growing opposition of concerned citizens, first the mayors of suburban cities (from any side of the political chessboard), and above all the emergence of ecologists (18% at regional elections in Île de France, 1992). Unexpectedly, in 1993, the new (right-wing) Minister of the Interior, Charles Pasqua, decided to put an end to the growth of the Parisian megapolis. He reflected thus a general turn in mindsets. Yet this does not mean that everybody agrees with the strategy for avoiding megapolis, and in fact it could mean a new form of megapolization.

 1. Introduction

In this text, I will first present a theorization of the reasons why megapoles sometimes develop in the world, and sometimes not, and I will use for this the regulation approach (Lipietz, 1987). Then I will introduce the ’Île de France debate’, as viewed from inside the Regional Council. Since that debate was not supported by professional regional planners, but by standard politicians, readers may be surprised by the lack of academic references. On the other hand, this reportage on a life size experiment could be useful for later theorizations in the field of planning and political science.

 2. Employment and City Growth

Most models of city growth, while assuming inter urban mobility, are based on employment growth. For instance, the mainstream scenario for the growth of the Paris megapolis, at the end of the 1980s, projected 5.6 million jobs and 12-13 million inhabitants by 2015. Yet this standard analysis had nothing to say about the types of job that would exist in France at the turn of the century. Clearly, the type of jobs created, or more precisely the mode of creation of jobs, will have an influence on the type of urbanization.

Should we expect full time or part time-jobs? In the ’primary’ (stable) or ’secondary’ (Precarious) labour market? Sociology and economics may be able to cast some light on this question. But we need a more specific urbanistic connexion: in what sense could changes in labour status, wage contract and labour time influence the concentration of urban activity, type of cities, way of life?

We shall try to follow this sequence: ’nature of jobs at the dawn of the twenty first century, urbanization’. Clearly, capital labour relations are important in shaping urban forms but are not necessarily to be privileged. Other considerations are relevant, either economic (such as industrial organization) or not (such as physical geography, or historical urban culture). But capital-labour relations are part of a broader ’societal paradigm’ encompassing many relevant causal factors of urban development.

2.1 Nature of Jobs and Model of Development

The ’nature of jobs’ implies the social norms of the capital-labour contract, their stability, the skill and involvement required from workers, the importance of a job in everyone’s life, the wage accruing to it; in short, determinant elements of a ’model of development’( [1]).

The former idea of a ’normal job’ expressed the dominant model of development of advanced capitalist countries, often labelled as Fordist’. We could reduce it to its two main characteristics:

(1) A hierarchical organization of labour, inspired by Taylorism and its division of the manual intellectual aspects of labour: the engineers, designers, organizers of labour on the one hand, the unskilled, manual, routinized operators on the other hand. In France, this divide had a clear urbanistic consequence, with the polarization between a central megapolis with a high composition of service and command jobs, and a decentralization of unskilled and manual jobs in middle size and small cities (Lipietz, 1977).

(2) ’Rigid’ industrial relations, i.e. strong social legislation, precise and enforcing collective agreements, heavy welfare state. This rigidity was obtained by unions after World War II, in exchange for their subordination to Taylorism at the work place. At the same time, it offered to capitalist development steady outlets for mass production.

This compromise was a real success up to the 1970s. Yet the first characteristic (hierarchical control of labour) appeared as a weakness in the 1970s, by comparison to a more efficient mobilization of human resources. On the other hand, the second characteristic (rigidity) appeared as a weakness in relation to the fall of profit rates, the fiscal crisis of the welfare state, and increasing competition from lower wage Third World countries.

Certainly, any new model combining ’flexibility’ and ’Involvement’ of waged workers would be a blessing for capitalist accumulation. Yet, the myth of a ’flexible specialization’ seems to be inconsistent ( [2]). In fact, a single worker cannot feel himself or herself involved in work while being connected to a job by a precarious link. In reality, in the 1980s most firms in advanced capitalist countries had to choose between two routes out of the Fordist crisis; nowadays, this choice appears as a characteristic of regions or even nations. Of course, concrete nations combine in various proportions of ’flexible’ and ’involved’ segments of the labour force, using gender and ethnic differences. But, with all these nuances, two broad families of models capture the concrete national trajectories of countries.

(1) A first model could be labelled ’flexible/hierarchical’, or ’polarizing’, or ’neo Taylorian’. It privileges a Taylorist division in labour and suppresses the rigidity of collective bargaining and the welfare state, This model triumphed in the US, the UK, Spain ... and partly in France.

(2) The second model privileges ’negotiated involvement’ of human resources, which imply workers’ participation in the struggle for quality and productivity in exchange for some kind of ’rigidity’ of labour contracts. The negotiation may be carried on at firm level Japan), sector level (Germany) or society level (Sweden), with of course contrasting social consequences.

According to the experience of the 1980s, the second family of models dominates the first one from the point of view of capitalist competitivity. But it is not likely to become hegemonic, for flexibility could dominate involvement in routinized industries. Europe, for instance, is likely to experience the coexistence of a ’negotiated involvement’ core and various more or less flexible peripheral regions, according to a geography which may differ from national boundaries (Leborgne & Lipietz, 1991, 1992). Moreover, the negotiation at the level of firmes or sectors may allow for a dualism of labour markets in the same territory.

2.2 From Production to Cities

At the social level, it is clear that the two families of model have different consequences.

’Flexible’ models mean an end to the classical Fordist way of life based on a large middle class encompassing a large and stable fraction of blue collar and white collar employees. A flexible and hierarchical society implies a division of the society into four tiers:

(1) Involved and over worked yuppies, executives and entrepreneurs.

(2) A relatively stable segment of skilled wage earners (the residual of the former Fordist compromise).

(3) A large segment of flexible workers, with irregular labour time in their life cycle.

(4) A larger or smaller segment of marginalized workers, having a marginal experience of formal wage labour.

In sum, an ’hour glass society’, a ’Brazilianization’ of some advanced capitalist countries, the proportion between segment (3) and (4) being the main difference between a neo-Taylorist North and ’peripheral Fordism’ in the South.( [3]).

By contrast, all the societies based on some form of negotiated involvement preserve the important Fordist characteristic of ’middle classes’. This is true for Sweden, Germany (despite ’flexible sectors’ occupied by Gästarbeiter)and even Japan (despite the generally flexible situation of women at work). This gives a ’barrel shaped’ income distribution.

This contrasting structure of household incomes (hour glass versus barrel) reflects into the structure of the social division of urban space. In the ’flexible’ models, polarization in incomes will be reflected in sharp contrasts between neighbourhoods, while ’negotiated involvement’ models will imply more homogeneity.

This first result is well known, obvious for geographers (compare Sao Paulo, Los Angeles and Frankfurt) and theoretically trivial. Less trivial is the following: it could be argued that not only the quality, but the very size of the agglomerations windepend. on the model of development. The argument could be summarized as follows.

– In ’negotiated involvement’ models, local societies are able consciously to mobilize their human resources and to reproduce them in a planned manner. School, university and professional training equilibriate the supply of and demand for a skilled labour force. Active local policy provides employment through partnership between labour and business unions. Stable ’principal agent’ relations characterized by ’loyalty’ involve not only capital labour relations but also capital capital relations: principal firms subcontracting networks, local bank local industries. As a result, local productive systems develop into small industrial districts d la Third Italy, a middle size metropolis such as Frankfurt (0.8 million inhabitants) or, at most, Munich or Milan (around 4 million), or else, an ’articulated’ megapolis like Randstad, the Rotterdam Amsterdam area (see Benko & Lipietz, 1992).

In the ’flexible’ models, a random job search process leads job searchers to central places where the labour market is known as the most active, since they cannot anticipate the creation of jobs in smaller places. Firms, in search of a large supply of diversely skilled labour, choose the same location, thus justifying the expectations of job searchers and relieving themselves from organizing the skilling of labour supply. This process of ’stochastic megapolization’ was typical of late industrialization in newly industrializing countries, such as Sao Paulo, Mexico or Seoul. Now it applies entirely to some northern megapoles (Scott 1992; Storper & Walker, 1989). There are in fact only four megapoles in the North: New York New Jersey, Los Angeles, London and Paris. All of them are located in ’Brazilianizing’ developed countries. Let us scrutinize the case of the last one.

 3. The Battle of Île de France

Following the previous section, a theorem could be suggested:

– The more polarizing (flexible hierarchical) a society, the deeper the opposition between its metropolis and its hinterland, and the deeper the social division of space within the megapolis.
– The more organized and the more negotiated the development of a society from its own human resources, the more likely the formation of a balanced network of middle-size metropoles, and the smoother the social division of space within the cities.

The growth of the Paris metropolis is a good illustration of this theorem.

3.1 Childhood of a Megapolis

In the pre Fordist period, France experienced one century of restricted capitalist accumulation, due to active protection of the petty bourgeoisie, mainly the peasantry. In 1945, the choice for modernization was made, and in 20 years one third of the French population moved from the rural sector to construction, manufacturing and tertiary jobs. At first, this unprecedented break entailed a rush to the Paris metropolis, with a destruction of rural networks but also a breakdown of traditional industrial districts (Ganne, 1992). Yet the opportunities offered to branch plants in the countryside, or in small middle size cities, within a Fordist division of labour, complemented an active policy of state management of French territory (DATAR: Délégation à l’Aménagement du Territoire et à l’Action Régionale), in the form of limited megapolization (Lipietz, 1977).

In the first regional plan (Schéma Directeur d’Aménagement Urbain de la Région Parisienne, SDAU) in 1965, the Paris region was supposed to reach the size of 14 million inhabitants in the next 25 years. In fact, by 1968 the stream was already reversed; Paris had become the biggest metropolis on the European continent, but with 8 million inhabitants it was not a real megapolis, and an important movement of active population from Paris to other regions was developing. This movement continued until the end of the Fordist era (around 1982); the Paris region (now Île de France) had become a region with a net natural (birth/death) growth and a net migration decrease. The revision of the SDAU in 1976 took into consideration this deceleration in growth and accepted the idea that the Paris region would have ’only’ 12 million inhabitants by the year 2000. Once the results of the census of 1982 were published, the forecast for 2000 was downgraded to 10.5 million for the whole region. In fact, Île de France, an administrative region of some 50 km radius, is larger than the real urban agglomeration, but the rest of the region is largely occupied by forests and wide open fields with, accordingly, a small population.

Yet, after the failure of a last attempt to restore Fordism in France, the French socialist government gave up any attempt at planning (notably through DATAR), and accepted the neo liberal, flexible outcome of the crisis of Fordism. As a result, the 1980s experienced a U-turn by comparison to the 30 previous years ( [4]). Society underwent a new polarization of wages and other incomes, of skills, and even of status (with a dramatic growth of marginalized and extremely poor people: segment (4) of the typology outlined earlier), and, at the same time, a new polarization of jobs and population streams towards Paris. Following London or Los Angeles on the path to ’Brazilianization’, Île de France became the unique growth pole in France, with most of the positive and negative consequences. Sociological, historical, organizational and administrative factors were converging towards a ’recentralization’ (Veltz, 1992). During the cyclical business upturn of the late 1980s, the majority of jobs created in France were in Île de France, but at the same time the Paris megapolis (and the Lyon metropolis) experienced urban and ethnic riots.

3.2 The Megapolist Offensive

At the end of the 1980s, the debate on the renewal of the 1976 SDAU started with an offensive in favour of megapolization. A SDAU, or more recently a Schéma Directeur Régional (SDR), is a document decided by the state after discussion with local authorities, which frames the more precise urban plans of municipalities. As such, an SDR has to define the total amount of developable land (and necessary networks) in relation to forecast growth in the next 25 years. In turn, this forecast is connected with a vision of the future hierarchy of cities.

The first document of the Direction Régionale de l’Equipement de I’lle de France (DREIF, a subdivision of the central state) clearly expressed confidence in the former tendency of restricted growth of the megapolis( [5]). In fact, at the beginning of 1989, few had noticed the U-turn in the locational structure of French growth. On the contrary, the ’insufficient’ growth of the 1970s and early 1980s was perceived as a weakness by comparison to ’competitors’. Research by a group of modernist geographers of Montpellier, GIP-RECLUS, was ordered by DATAR to investigate the ranking of French cities within Europe. On the basis of purely productivist and competitive criteria, the study outlined the opposition between a central ’blue banana’ from London to Milan via the Rhine Valley, and a marginalized periphery of the European Community( [6]). Only Paris (rank 2) could compete with London (rank 1), thus appearing as ’the only French opportunity’. Implicitly, but consistently with the choice of flexibility, the reference was the growth pole megapolis a laScott (1992) and Storper and Walker (1989).

This idea immediately gathered a consensus among elites in France, from the left (the socialist government of France, and its vision of the Paris region expressed by DREIF - despite the early reluctance of some ’greenish’ technocrats) to the right (the government of the city of Paris, its mayor, Jacques Chirac, its Atelier Parisien d’Urbanisme, APUR, and the government of Île de France, with its Institut d’Aménagement Urbain de la Région he de France, IAURIF). Finally, the three ’think tanks’ - DREIF, APUR and IAURIF - published a Livre Blanc de I’lle de France in January 1990, with a clear reference to the RECLUS criteria and an open ambition: to bypass London in the competition for the title of ’the’ megapolis of Europe ( [7]). This ambition implied a concentration of the main crossroads of Europe in the region, for motorways, high speed trains (TGV), optical fibres, and an airport hub of 80 million passengers a year, plus a preferential national support for high-tech industries and some tax exemptions. The idea was to attract the ’floating’ headquarters of multinational companies, with the justification that such a location would induce (by trickledown) some manufacturing jobs in other French regions.

3.3 The Anti megapolis Reaction

In July 1990, a Commission du Développement Economique et Social du Schéma Directeur was nominated by the central state to organize the discussion around the new SDR. Immediately, a vigorous opposition formed against the prospect of a megapolis. The elected elites (of any tendency) from Paris suburban municipalities expressed their rejection of this London model. Life in Paris, the urban ecology (water networks, transportation, multi-ethnicity, housing, lack of green spaces, etc.), was seen as problematic, so the prospect of a new growth wave was unanimously rejected.

Moreover, the publication of the results of the 1990 census began to emerge. While the central hypothesis was forecasting a population of between 12 and 13 million in 2015, with 5.6 million jobs, the census revealed that Île de France had already reached 10.6 million inhabitants, 10 years in advance of the projections of the 1980s. In fact, the ’natural’ demography proved more positive than had been thought (more and more women having an unexpected third child after the age of 40), and the polarization of French society during the 1980s caused a reverse of the migration flows ’in favour’ of Paris. In a note to the Commission du Développement Economique et Social, the figure of 13 million inhabitants in 2015 is presented as an unavoidable threshold, thus 15 million may be thought a reasonable consequence of a megapolist policy.

At the spring 1992 regional elections, the ecologists obtained 18%, and a blocking minority ( [8]). This expressed a fall in the popularity of the Socialists, but also a growing awareness by the population of green arguments: against the megapolis, against cars and motorways, in favour of collective transportation and improvement of the environment. As a result, a new consensus appeared against the growth of the population in Île de France.

But this does not imply a consensus about ’how’ to avoid it. For the ecologists, the only positive way is a suitable national development policy (involving agriculture, industrial relations, control on the location of job creation), plus the generalization of local employment and environment policies in other regions, in order to attract or at least retain population. But Île de France could help in avoiding any measure which would foster its own attractiveness: avoiding concentration of investment, either in infrastructure or office buildings, delocalizing ’mobile’ jobs to other regions. Thus, the only possibilities to fight unemployment in the Paris region would be the reduction of labour time and the endogenous creation of services to serve the local population. That is of course unacceptable for other political groups, who are still willing to attract jobs from other regions (in order to fight unemployment in the Paris region) but not their population.

In the successive versions of the SDR proposed by the representatives of the central tate and its socialist government, the population forecasts for 2015 were progressively reduced to 12.2 million. Yet, the final project was eventually rejected, by the ecologists for consistency reasons, by the other non-socialist groups for political reasons.

In March 1993, the right won the national elections. The new Minister of the Interior and Territorial Management was to be Charles Pasqua, ’boss’ of the Hauts-de-Seine, the richest and most expansionist department of the Paris region. To general amazement, Charles Pasqua chose to enlarge his national popularity by a dramatic discourse against the weight of the Paris region, and launched a great national debate on territorial management. Symbolically, he recruited to his staff Jacques Voisard, a long time prophet of decentralization. At the Interministerial Council on Territorial Management, held in the tiny provincial city of Mende, he proclaimed a new downsizing of the prospective population of Île de France to 11.8 million in 2015.

These battles about statistics, one quarter of a century in advance, may seem ridiculous when we have in mind 30 years of blunders in economic and demographic forecasts. But the symbolism is clear: the acceptable growth of Île de France is proclaimed to be less than its natural growth. This has concrete consequences for the SDR, in reducing the land for urbanization and office building. Thus, some accelerated development, somewhere, has to be organized. But where, and how?

3.4 The ’Grand Bassin Strategy. A False Window?

At the end of 1993, the IAURIF made explicit what seemed to be a possible compromise:

Clearly, the uncontrolled growth of Île de France would aggravate hazards [social and ecological]. Thus, today, there is no more real debate on the objective of a limitation to this growth. The question is only ’how’. Most of the answer lies probably in a change of scale of regional management, by relocating it in the frame of the Paris Bassin ( [9]).

This was a clear reference to a former proposition of DATAR. In April 1992, this state agency had proposed three contrasting scenarios encompassing the whole of the eight administrative regions composing the physical Parisian Bassin, a broad plain of some 250 km in radius, from the river Loire to the liver Somme, between Brittany and Lorraine( [10]). The first scenario was clearly the megapolis. The Île de France population would reach 13 million (in 2015) and its immediate outskirts (in the other regions of the Bassin) 2.3 million. To the contrary, the third one was based on the development of historical metropoles some 200 km from Paris. Then, the population of Île de France would be 12 million, the outskirts would reach 4 million, but, for instance, an area such as Tours Blois Orl6ans (the Jardin de la France’ along the river Loire) would reach 1.5 million inhabitants.

Immediately, the presidencies of the other regions of the ’Grand Bassin’ accepted some form of the third scenario, and the president of the region Île de France, Michel Giraud, in line with his objective of a new alliance with the ecologists, accepted cooperation within this frame. The IAURIF report matches this ’Grand Bassin’ compromise. But, a critical view demands that things are to be examined more closely.

First, the IAURIF seems to accept the analyses of the Laboratoire d’Observation de l’Economie et des Institutions Locales (L’ŒIL, 1992), a laboratory of Creteil University supporting the aficionadosof a Paris megapolis. The L’ŒIL (and the IAURIF report) insists that, far from ’pumping’ incomes from other regions, Île de France was in fact the source of a net flow of state expenditures to the other regions ( [11]). " Actually, the ’rate’ of local taxes in the Paris region is always the smaller in France, and only the importance of its fiscal basis compensates for the weakness of the rates. In the same way, it is true that the national welfare state organizes a net transfer from Île de France incomes to non-Parisian regional payments, but this is only the expression of current life cycles: many non Parisian French people come to work in the Paris region for a part of their life cycle and then retire to ’their’ region. Geographically, the IAURIF report offers two very significant maps, reproduced here. Figure 1 shows the cantons (small administrative units) which have lost population to the Paris region between 1982 and 1990. The whole active population of France is represented here, the whole East and North of France, all the major cities except Toulouse: from Rennes to Marseille, from Bordeaux to Strasbourg, from Grenoble to Lille.{{}}The megapolistic growth of the Paris region is actually ’pumping’ its blood from smaller growth poles.

Figure 2, on the contrary, represents the cantons which gain population from the Paris region. Two very distinct groups appear. The first one is exactly the map of holiday and retirement regions: all the Atlantic coast, and still more the Mediterranean coast, the mountains and the deep countryside. Compared with Figure 1, this is the explanation of the fiscal result outlined by the L’ŒIL investigations. People come to work in he de France and return to rest in their homeland and in the sun.

The second group of cantons in Figure 2is much moreworrying.Itis the ring around the administrative limits of Île de France, 50 to 80 km from Notre Dame. In fact, ’scenario (1)’ of DATAR is already in force. Though the limits of Île de France exceed the continuous urban area of Paris, urban settlements resulting from its demographic explosion are already developing just outside these administrative limits.

Thus, the late conversion of IAURIF is incomplete and ambiguous. Though still supporting the pro Île de France theorical lobby, such as scholars in L’ŒIL, it acknowledges dangerous trends (Figures 1 and 2). But, neither IAURIF nor any speaker for the anti-megapolist resistance (except for the ecologist) connected the size of the megapolis to its social polarization and to the ’Brazilianization’ of society. Mass exclusion and ethnic unrest are treated, in these texts, as some unavoidable consequence of progress and of the anonymity of huge human concentrations, to be dealt with by some cosmetic micro local policies.

 4. Summary

The growth of the cantons on the outskirts of the Paris region is the crux of the matter. Does the strategy of ’Grand Bassin’ mean the organization of a constellation of growth poles in distant metropoles (that is, far from the outskirts)? This would imply an active policy of ’negotiated involvement’ of local human resources at the level of other regions in the ’Grand Bassin’, plus a corresponding strategy by the national state, plus a real conversion of Parisian elites against the privileges of their central city. Or does the ’Grand Bassin’ strategy mean the acknowledgement of a supemova, beyond the administrative limits of the central region? If so, it would be nothing more than the realization of the worst scenario of megapolis, with a fantastic concentration of ethnic tensions at its periphery (already illustrated by Dreux, the first important city governed by the Front National), and an unmanageable growth of road traffic, despite all the anti greenhouse effect promises signed by France at the Rio Conference.

The IAURIF leaves two options open. In Figure 3, two rings are organized within the ’Grand Bassin’. The first one, the ’Wheat Road’, is just organizing the overflow of Paris expansion through the administrative limits of Île de France: towards Chartres, Dreux, Evreux, Beauvais, Creil, Compi»gne. A second ring, the ’Road of Cathedrals’, organizes a network fostering the potentialities of Caen, Le Havre, Amiens, Reims, Orleans, Tours and Le Mans, far away from Paris.

Unfortunately, on 5 April 1994, the eight regions of the Parisian Bassin signed the Charte du Grand Bassin. The ’Wheat Road’ was preferred, and the President of Île de France admitted that it represented an opportunity to build the ’fourth ring motorway’ outside his region (while the ecologists contest the utility of the third). Three weeks later, on 28 April, without any new negotiation, a final version of the SDRIF{{}}was decided by the central government, in line with the decisions of Mende: Île de France will have 11.8 million inhabitants in 2015, no more, and the new areas dedicated to urbanization are downsized by 13%.

Yet this kind of wishful thinking has no significance in the absence of political choices. And, if the theorem at the beginning of this article is correct, these choices are wen beyond urbanistic options.

 References

BENKO, G. and LIPIETZ, A. (Eds) (1992) les régions qui gagnent. Districts et réseaux: les nouveaux paradigmes de la géographie économique.Paris: PUF.
– GANNE, B. (1992) Place et évolution des syst»mes productifs locaux en France: économie politique d’une transformation, in BENKO and LIPIETZ (1992).
– LEBORGNE, D. and LIPIETZ, A. (1991) Two social strategies in the production of new industrial spaces, in G. BENKO and M. DUNFORD (Eds) Industrial Change & Regional Development.London: Belhaven Press.
– LEBORGNE, D. and LIPIETZ, A. (1992) Conceptual fallacies and open questions on post Fordism, in M. STORPER and A. SCOTT (Eds) Pathways to Industrialization and Regional Development.London: Routledge.
– LIPIETZ, A. (1977) Le Capital et son espace.Paris: F. Maspéro.
– LIPIETZ, A. (1987) Mirages and Miracles. The Crises of Global Fordism.London: Verso.
– LIPIETZ, A. (1992) Towards a New Economic Order. Postfordism, Ecology and Democracy.Oxford: OUP; London, New York: Polity Press.
– LIPIETZ, A. (1995) Capital-labour relations at the dawn of XXIst century, in D. SHOR and H. You (Eds) Changing Production Relations: A Global Perspective.London: Edward Elgar.
– L’ŒIL (1992) Budgets publics et redistribution spatiale,IAURIF, November.
– PAGE, E. (1993) Gain or Drain? London in the UK: a review of regional taxation and expenditures. Paper presented at the Conference Métropoles et Aménagement du Territoire,Paris-Dauphine, 12-13 May.
– PIORE, M. and SABEL, R. (1984) The Second Industrial Divide: Possibilities for Prosperity.New York: Basic Books.
– SCOTT, A. (1992) L’économie métropolitaine: organisation industrielle et croissance urbaine, in BENKO and LIPIETZ (1992).
– STORPER, M. and WALKER, R. (1989) The Capitalist Imperative. Territory, Technology, and Industrial Growth.Oxford: Blackwell.
– VELTZ, P. (1992) Hierarchies et réseaux dans l’organisation de la production et du territoire, in BENKO and LIPIETZ (1992).