Band 51, Heft 1 Boss-Verlag, Kleve
E D I T O R I A L
T h e articles contained in this issue go back to the fifth Dutch-German Seminar on Economic Geography which was jointly organised by E G B E R T W E V E R , Utrecht, and H A N S H . B L O T E V O G E L , Duisburg, and which took place in M ü l h e i m / R u h r in M a y 1996.
The first idea to organise a Dutch-German Seminar on Economic Geography was conceived in the mid 1980s when a small group of Dutch and (West) G e r m a n economic geographers decided to improve mutual con- tacts and to establish a platform for presenting and discussing both theoretical issues as well as empirical findings. The result was a first seminar held in Aachen in 1988, H E L M U T H B R E U E R being the local organiser.
Since then four more meetings followed: in Nijmegen (1990), Reinhausen/Göttingen (1992), Garderen/
Apeldoorn (1994) and M ü l h e i m / R u h r (1996). T h e seminar has by now become a tradition, and it is intended to continue this series of bilateral conferences. In fact, interest in participating has grown steadily. At the fifth meeting, in Mülheim, as many as 50 persons attended. T h e vivid exchange of ideas and the personal networks that have been established have already been beneficial for many participants.
T h e output of each of the first four meetings was a volume of collected contributions, presenting revised versions of most of the papers given during the conferences:
DE S M I T , M . and W E V E R , E. (Eds.) ( 1 9 8 9 ) : Regional and local economic policies and technology. Utrecht/
Amsterdam. Nederlandse Geografische Studies, vol. 99.
DE S M I T , M . and W E V E R , E. (Eds.) ( 1 9 9 1 ) : Complexes, formations and networks. Utrecht/Amsterdam.
Nederlandse Geografische Studies, vol. 132.
G A E B E , W. and S C H A M P , E. W. (Eds.) ( 1 9 9 4 ) : Gateways to the European market: case studies from the Netherlands and Germany. Münster. Wirtschaftsgeographie, vol. 4.
VAN DER K N A A P , G . A . and W E V E R , E. (Eds.) (1996): Industrial organization: the firm and its labour market.
Utrecht. Nederlandse Geografische Studies, vol. 207.
After the Mülheim seminar it was decided not to produce another book but to launch a special issue in one of our major journals instead. T h e main reason was not a lack of good contributions compared to the former conferences, but a reflection of the recent discussion among German geographers on the inflation of publications and especially of books containing heterogenous conference proceedings and the subsequent drying out of the established journals. As one of the organisers of the Mülheim seminar is a m e m b e r of the editorial board of Erdkunde the present issue was arranged.
For this publication we put together six of the papers presented at the Mülheim seminar. This small selection does by no means imply that all other contributions are not worth being published. Either they are or will be published elsewhere or the speakers declared themselves unable to produce a written version within a few months, and in the end we just had to select from several nearly equivalent contributions due to the given space limit. We surely regret that our decision in favour of a journal issue prevents an inclusion of all good and stimulating contributions.
This selection of six articles represents a cross-section of the whole programme of the Mülheim seminar.
According to our intention of emphasising theoretical and methodological questions three contributions mainly deal with theoretical concepts: R O B E R T H A S S I N K (Dortmund) on 'clusters', 'districts' and 'milieus',
P Ä I V I O I N A S (Helsinki/Rotterdam) on 'regional embeddedness', as well as R O N B O S C H M A (Enschede) on the concept of 'windows of locational opportunity'. T h e other three papers are more empirical, but nevertheless with a clear (and corresponding) theoretical orientation: B A R B A R A L E N Z and K L A U S K U L I N A T (Stuttgart) deal with contact-networks and regional milieus in Württemberg, co-operation between small and medium sized firms in the Neckar-Alb-Region is studied by R E I N H O L D G R O T Z and B O R I S B R A U N (Bonn) whereas S T E P H A N I E
L O W E Y (Würzburg) is investigating a similar topic in Lower and Middle Franconia.
T h e mere listing of headlines shows the key notions around which the papers and discussions of the Mülheim seminar were centered and which form the focus of the present issue as well. We hope that the reader will get at least some impression of the stimulating atmosphere of the Mülheim conference and appreciate the selected constributions which mark a major front of current research in economic geography.
H A N S H . B L O T E V O G E L - Duisburg E G B E R T W E V E R - Utrecht
W H A T D I S T I N G U I S H E S ' G O O D ' F R O M ' B A D ' I N D U S T R I A L A G G L O M E R A T I O N S ?
R O B E R T H A S S I N K
Zusammenfassung: Was unterscheidet ,gute' von schlechten' Industriedistrikten?
Moderne regionalwirtschaftliche Theorie-Konzepte zielen darauf ab, den wirtschaftlichen Erfolg einer kleinen G r u p p e von Regionen zu erklären. Einige von diesen Konzepten sind von Wirtschaftsgeographen u n d Sozialwissenschaftlern ent- wickelt worden, die von der Frage ,warum sind manche Regionen wirtschaftlich erfolgreich?' ausgehen, während andere Konzepte von Wirtschaftswissenschaftlern oder Regionalökonomen entwickelt worden sind, die die Frage ,warum kon- zentrieren sich international erfolgreiche Branchen in ein p a a r Ländern oder Regionen?' als Ausgangspunkt n e h m e n . Diese Konzepte stimmen bezüglich folgender Aspekte überein: sie versuchen alle, das Entstehen und Entwickeln von Innovationen zu erklären, sie betonen die wichtige Rolle der wirtschaftlichen Organisationsstrukturen für regionale Innovationsprozesse und sie konzentrieren sich auf Erfolgsregionen. Das größte Defizit dieser Konzepte ist jedoch, daß sie nicht in der Lage sind ,gute' Industriedistrikte, wie das Dritte Italien und Silicon Valley, von s c h l e c h t e n ' , wie dem Ruhrgebiet und der Route 128 bei Boston, zu unterscheiden.
Hauptziel dieses Aufsatzes ist es herauszufinden, was ,gute' Industriedistrikte von .schlechten' unterscheidet u n d wann sich die Koordination von Aktivitäten zwischen Akteuren in Industriedistrikten von einer Stärke in eine Schwäche wan- delt. Dazu werden erst die modernen Theorie-Konzepte zur regionalwirtschaftlichen Entwicklung kurz präsentiert.
Danach wird näher auf Literatur eingegangen, die betont, daß die Kombination von Technologie und räumlicher Ballung wirtschaftlicher Aktivitäten allein keine positiv wirkenden wechselseitigen Abhängigkeitsverhältnisse schafft, sondern daß diese Verhältnisse eher das Resultat von wirtschaftlichen Organisationsstrukturen sind. Diese Literatur unterscheidet zwischen regionalen netzwerkartigen Industriesystemen als Grundlage für ,gute' Industriedistrikte u n d unabhängigen großfirmenorientierten Industriesystemen als Grundlage für .schlechte' Industriedistrikte.
Summary: Modern theoretical concepts on regional economic development try to explain the economic success of a small group of regions. Some of these concepts are developed by economic geographers and sociologists who take the question 'why are some regions economically successful?' as a starting-point, whereas others are developed by economists who focus on the question 'why do internationally successful industries tend to concentrate in a few nations or regions?' These concepts share the attempt to explain the origin and development of innovation, stress the significance of industrial organisation for regional innovation processes and focus on success stories. T h e problem with most of these concepts, however, is that their central explanatory aspects cannot distinguish between 'good' industrial agglomerations, such as the T h i r d Italy and Silicon Valley, and ' b a d ' ones, such as the R u h r Area and Route 128 near Boston.
T h e central aim of the article is to find out what distinguishes 'good' from ' b a d ' industrial agglomerations and at what point co-ordination of activities among the actors in an industry cluster turns from an advantage into a disadvantage.
First, it will review modern theoretical concepts on regional economic development on this issue. Secondly, it will go deeper into work that stresses that the combination of technology and spatial clustering alone does not create mutually beneficial interdependencies, but that they are rather the result of organisational structures. This work distinguishes between regional network-based industrial systems as the basis for 'good' agglomerations and independent firm-based industrial systems as the basis for ' b a d ' ones.
1 Introduction
I n e c o n o m i c g e o g r a p h y a n d r e g i o n a l e c o n o m i c s i n t e r e s t i n g t h o u g h t s w e r e d e v e l o p e d in t h e s e c o n d h a l f of t h e 1980s a n d t h e b e g i n n i n g of t h e 1990s. B u t since 1992 w e a r e w i t n e s s i n g s o m e t h i n g like a d e a d l o c k . S c h o l a r s e i t h e r criticise r e c e n t l y d e v e l o p e d i n d i v i d u a l c o n c e p t s , s u c h as t h e i n n o v a t i v e m i l i e u a n d i n d u s t r i a l d i s t r i c t a p p r o a c h e s , o r t r y to d e v e l o p a n o v e r v i e w i n t h e glut of c o n c e p t s ( S T O R P E R 1995; L A G E N D I J K 1996).
I n this article t h e s e e x e r c i s e s will n o t b e r e p e a t e d . I n s t e a d , a s e r i o u s s h o r t c o m i n g in m o d e r n c o n c e p t s will b e s t r e s s e d a n d it will b e i n d i c a t e d in w h i c h d i r e c -
t i o n r e s e a r c h h a s t o d e v e l o p t o o v e r c o m e t h i s s h o r t - c o m i n g .
M o s t m o d e r n t h e o r e t i c a l c o n c e p t s o n r e g i o n a l eco- n o m i c d e v e l o p m e n t r e v o l v e a r o u n d t w o q u e s t i o n s , ' w h y a r e s o m e r e g i o n s e c o n o m i c a l l y s u c c e s s f u l ? ' a n d ' w h y d o i n t e r n a t i o n a l l y successful i n d u s t r i e s t e n d to c o n c e n t r a t e in a few n a t i o n s o r r e g i o n s ? ' T h e s e c o n - c e p t s , s u c h as i n d u s t r i a l d i s t r i c t s , i n n o v a t i v e m i l i e u x , r e g i o n a l p r o d u c t i o n c l u s t e r s , h a v e c e r t a i n c o m m o n f e a t u r e s : t h e y all a t t e m p t to e x p l a i n t h e o r i g i n a n d d e v e l o p m e n t of i n n o v a t i o n , s t r e s s t h e s i g n i f i c a n c e of i n d u s t r i a l o r g a n i s a t i o n for r e g i o n a l i n n o v a t i o n p r o - cesses a n d focus o n success s t o r i e s . T h e p r o b l e m w i t h
these concepts, however, is that their central explana- tory aspects cannot distinguish 'good' agglomera- tions, such as the Third Italy and Silicon Valley, from ' b a d ' ones, such as the R u h r Area and Route 128 near Boston.
T h e central aim of this article is to find out what distinguishes 'good' from ' b a d ' agglomerations and at what point co-ordination of activities among the actors in an industry cluster turns from an advantage to a disadvantage. Before this main question will be dealt with in Section 4, the two following questions shall be explored first:
- why are some regions economically successful? and why do internationally successful industries tend to concentrate in a few nations or regions? (Section 2), - why may regional production clusters, industrial
districts, milieux and the like fail? (Section 3).
2 Why are some regions economically successful? and why do internationally successful industries tend to concentrate in a few nations or regions?
In the 1980s concepts were developed by economic geographers and sociologists who asked themselves the classical question in economic geography: why are some regions economically successful? T h e main change in orientation generated by these new con- cepts is that scholars no longer just stress the inno- vativeness of firms and industries in a region (pro- duction structure) as explanatory factors of regional economic inequalities. They found out that the corre- lation between the innovativeness of the regional pro- duction structure and regional economic develop- ment is not watertight ( L A P P L E 1994 and 1996; R O N - NEBERGER 1995; M A L E C K I 1991; S T E I N E R 1985). In order to explain regional economic inequalities it is not so much important what is produced in a region (the production structure), but how and under which conditions ( L A P P L E 1996; F R O M H O L D - E I S E B I T H 1995;
K R Â T K E 1996). These conditions might be the modes of inter-firm co-operation, the functional division of labour, the position of firms in the supply pyramid, the qualification of the workforce, the institutional fabric, social and technical infrastructures, economic history and cultural traditions in the region. T h e indi- vidual firm is no longer seen as an isolated actor, but the dependence of the firm on its direct regional environment is stressed ( K I L P E R a. L A T N I A K 1996).
T h e above-mentioned conditions are important for the diffusion of new technologies. In combination with fundamental changes in production organisa- tion, these conditions have been taken u p by many
scholars in the 1980s to explain the rise of high-tech regions and industrial districts in the U S A and Western Europe. They have come u p with flexible specialisation, industrial districts and innovative milieu approaches.
T h e flexible specialisation approach, indicating the shift from mass production to a new organisation of production, re-discovered some industrial districts, mainly districts in the Third Italy and Baden-Wiirt- temberg, in which it found the clearest example of the new organisation form in a regionally-clustered form
( P I O R E a . S A B E L 1 9 8 4 ; P Y K E a . S E N G E N B E R G E R 1 9 9 2 ; S A B E L 1 9 8 9 ; S A B E L e t a l . 1 9 8 7 ; S C O T T a . S T O R P E R 1 9 8 7 ; S C O T T a . S T O R P E R 1 9 9 2 ; S T O R P E R a . S C O T T
1995; L A P P L E 1996). Important elements of these districts are networks between flexible SMEs and a wide range of institutions.
T h e industrial district scholars use three arguments to support their hypothesis of regional clustering.
First, the increased out-sourcing by firms to suppliers and other partners should take place in close prox- imity to enable face-to-face interaction and thus to avoid high transaction costs ( S A X E N I A N 1 9 9 4 and Economist 1996 a with regard to Silicon Valley).
Secondly, new logistic strategies, mainly just-in-time, which aim at reducing stocks, compel suppliers to locate not too far from their customers ( O E C D 1 9 9 4 ) .
Thirdly, multinational corporations increasingly tend to decentralise boundary-spanning functions previ- ously done by headquarters to divisional offices and production units, so that sub-units become more anchored in the region than in earlier times ( D I C K E N e t a l . 1 9 9 4 ; H I R S T a . Z E I T L I N 1 9 9 1 ; S A B E L 1 9 8 9 ) .
Despite the popularity of the industrial district approach during the second half of the 1980s, several authors distrust the relevance of this approach to ex- plain present regional economic inequalities ( M A L M - BERG 1990). They point at three broad weaknesses.
First, the most fundamental criticism concerns the trend of re-regionalisation of production systems
( A M I N a . R O B I N S 1 9 9 0 ; K I L P E R a . L A T N I A K 1 9 9 6 ; G R O T Z a. B R A U N 1993; L A G E N D I J K 1996). As multi- nationals with their global networks have by far more impact on the world economy than locally embedded firms, flexibility is more a matter of industrial organi- sation on a global rather than on a local or regional scale ( R O N N E B E R G E R 1995; O E C D 1994). Secondly, since only a few success regions are analysed in an anecdotal way, there is a lack of evidence to speak about a theory which has general validity ( S T A B E R 1 9 9 6 ; K R A T K E 1 9 9 6 ; T Ó D T L I N G 1 9 9 4 ; R O N N E B E R G E R
1995; L A G E N D I J K 1996). Even the highlighted success regions Emilia-Romagna, Baden-Württemberg and
Silicon Valley differ concerning a whole range of fun- damental aspects ( H E I D E N R E I C H 1 9 9 6 ; D I G I O V A N N A 1 9 9 6 ) . Thirdly, more detailed, the trend of subcon- tracting to suppliers in close proximity is doubted
( M A L E C K I 1 9 9 1 ; M A I R 1 9 9 2 ; B E R T R A M 1 9 9 2 ; M O R G A N 1 9 9 6 ) .
Despite the criticism, some points of the flexible specialisation approach remain unchallenged ( S T O R - PER 1995; D Ó H L a. S A U E R 1995). Flexibility and spe- cialisation can be regarded as fundamental alter- natives to mass production. There is also broad agree- ment on both the fact that dynamic forces in contem- porary capitalist development are both localised and territorially specific and on the increasing importance of institutionalised networks. Therefore, regional co-makership (especially with regard to R&D) and global-sourcing strategies (mainly low-value produc- tion) generate at the same time an internationalising and a re-agglomeration of modern economies ( M A L M - BERG 1990; S A X E N I A N 1994; M O R G A N 1995).
Compared with the industrial district approach, advocates of the innovative milieu approach place greater stress on socio-cultural factors and inter- personal relationships as the basis for inter-firm rela- tionships, collective learning processes and thus for regional economic growth ( A Y D A L O T a. K E E B L E 1 9 8 8 ) .
In contrast to industrial district scholars, they do not strictly define the spatial borders of the milieu and even stress that having some actors with outside con- tacts is an important prerequisite for long-term inno- vativeness of the milieu ( F R O M H O L D - E I S E B I T H 1 9 9 5 ) .
They also stress the encouraging role of proximity for collective learning ( C A S T E L L S a. H A L L 1 9 9 4 ) . Advo- cates of the approach believe that firms are the pro- duct of their environment so that the innovativeness of this environment determines the innovativeness of the firms ( A Y D A L O T a. K E E B L E 1 9 8 8 ; T Ó D T L I N G 1 9 9 0 ) .
The milieu approach, however, has been criticised for its lack of capacity to operationalise and lack of empirical evidence ( S T E R N B E R G 1 9 9 5 a, 5 8 ) . More- over, neither the term 'milieu' nor its spatial implica- tions have been clearly defined ("innovation occurs because of a milieu, and a milieu is what exists in regions where there is innovation" S T O R P E R 1 9 9 5 , 2 0 3 ; see also S T E R N B E R G 1 9 9 5 b; T Ó D T L I N G 1 9 9 0 ; L A G E N D I J K 1 9 9 6 ) .
Recently some well-known economists have been dealing with the question why internationally success- ful industries tend to concentrate in a few nations or regions. P O R T E R (1990), E N R I G H T (1995b) and K R U G -
MAN (1991), contribute in their theoretical concepts around networking and clustering much attention to agglomeration effects and regional economic develop-
ment. They stress two striking features. First, they state that internationally successful industries are exceptional in the way firms within these industries are related with each other, namely through network relations. T h e content of these relations cannot be confined to prices and quantities alone: it is richer and encompasses trust, experience and history ( L A M B O O Y
1994). Compared with markets (where trust is at a minimum) and hierarchies (where knowledge is power), networking is more efficient in realising the potential for creating and diffusing economically ex- ploitable knowledge as the key to competitive success
( C O O K E 1995). Secondly, they focus on geographic clustering of industries. K R U G M A N (1991) points to the fact that the historical process of industrialisation in the U S A and Europe is marked by stories of small accidents leading to the establishment of one or two persistent centres of production (see also E N R I G H T
1994 and O E C D 1994). Thereafter cumulative pro- cesses can generate a geographical structure of pro- duction which may be stable for long periods of time.
E N R I G H T (1994) emphasises spin-off firms within regional clusters as an ability of such clusters to foster new companies and to enhance innovation (see also O E C D 1994).
An interesting question is, of course, why indus- tries tend to cluster geographically. K R U G M A N (1991, 98) and other economists stress transaction costs as a general explanation for geographic clustering: "Be- cause of the costs of transacting across distance, the preferred locations for each individual producer are those where demand is large or supply of inputs is particularly convenient - which in general are the locations chosen by other producers." O r as S A X E N I A N
(1994, 173) puts it: "producers benefit from sharing the costs of common external resources such as infra- structure and services, skilled labor pools, specialized suppliers, and a common knowledge base . . . When these factors of production are geographically concen- trated, firms gain the additional benefits of spatial proximity, or 'economies of agglomeration'."
In addition to these general explanations, more specific factors explain why geographical concentra- tion is good for competitiveness ( K R U G M A N 1991;
P O R T E R 1990; E N R I G H T 1995a). First, internationally successful industries tend to concentrate because (informal) information flows locally more easily than over greater distances, and industrial centres gen- erate technological spillovers (Economist 1996 b).
Therefore, geographical concentration can stimulate a fast diffusion of new technologies. E N R I G H T (1994, 4):
"Even in the age of rapid communication and ad- vanced information systems, it appears that impor-
tant forms of information are still best transmitted when the parties are in close geographic proximity."
Secondly, geographical clustering can help firms to achieve vertical and horizontal disintegration
( E N R I G H T 1 9 9 5 a; L A M B O O Y 1 9 9 4 ) . In fact, firms within a geographic cluster often exhibit lower levels of vertical integration than their dispersed counter- parts, since transaction costs tend to be lower in close proximity. S O K O L O F F ( 1 9 9 5 ) , however, points to the fact that proximity is only one of many factors that influence the degree of vertical integration. Thirdly, geographical concentration stimulates competition between rivals. T h e regional press continually com- pares firms, people know each other and want to enjoy high prestige ( E N R I G H T 1 9 9 5 a). Competition, on its turn, combined with intensive R&D-co-operation with customers, helps to upgrade suppliers technol- ogically. Fourthly, geographical concentration stimu- lates firms both to fund local training and research centres and to put pressure on political support in creating specialised factors, such as specific training and research centres.
Paradoxically, regional clusters entail both greater co-operation and greater competition among direct competitors than geographically dispersed industries
( E N R I G H T 1 9 9 5 b). Co-operation in vertical relation- ships (buyer-suppliers), however, is easier to under- stand than co-operation in horizontal relationships (those with direct competitors). In many successful clusters, there is co-operation on a horizontal level in some activities, such as lobbying, while other activ- ities, such as company-specific marketing, tend to be done in a competitive manner ( E N R I G H T 1 9 9 4 ) .
Authors of the regulation school have criticised most of the above-mentioned approaches as being technologically and economically deterministic and have tried to put the explanations for regional eco- nomic development in a broader social, cultural and political context, often at a national level ( K R A T K E 1 9 9 6 ; M A R S H A L L 1 9 8 7 ; M O U L A E R T a . S W Y N G E D O U W 1 9 9 1 ) .
In a recent review article on the above-mentioned concepts, S T O R P E R ( 1 9 9 5 ) tries to bring together eco- nomic geography with the neo-Schumpeterian or evo- lutionary school of technological change, an attempt that has also been undertaken by M O R G A N ( 1 9 9 5 ) and
M A S K E L L and M A L M B E R G ( 1 9 9 5 ) . This school regards the mutual relations between innovations, firms and the political and socio-institutional forces as con- ditions for an optimal diffusion process and thus for economic growth ( D A V E L A A R 1 9 8 9 ) . It does not con- sider the innovation process as a linear model, but as an interactive process in which interactive learning
and feedback loops take place on a continuous basis
( M A L E C K I 1991; A S H E I M 1996). Furthermore, the school stresses that innovation is shaped by a variety of institutional routines and social conventions (tacit knowledge; trust, which cannot be bought, but has to be earned through repeated transactions; social capital). Key part of S T O R P E R ' S (1995) explanation of regional economic growth is the association between organisational and technological learning and agglomeration, which has two roots, namely localised input-output relations of traded interdependencies and, more important, untraded interdependencies (labour market, regional conventions, norms and values, institutions, knowledge systems, tacit knowledge).
These untraded interdependencies attach to the pro- cess of economic and organisational learning and co- ordination and where they are localised the region is a key, necessary element in the 'supply architecture' for learning and innovation ( S T O R P E R 1995).
3 Why may industrial districts, production clusters, inno- vative milieux and the like fail?
In a recent article the Economist (1996b, 66) stresses only one side of the clustering coin as it con- cludes an article on this issue as follows: "it seems that, just like everything else, success tends to cluster".
S A X E N I A N (1994, 161) considers this topic more care- fully where she states: "spatial clustering alone does not create mutually beneficial interdependencies.
An industrial system may be geographically agglom- erated and yet have limited capacity for adaption.
This is overwhelmingly a function of organizational structure, not of technology or firm size."
Since 1992 the explanatory power of many of the above-mentioned modern theoretical concepts that are based on the economic success of some industrial districts is put into question since some of these districts are faced with severe economic problems
( B R A C Z Y K e t a l . 1 9 9 6 ; S T A B E R 1 9 9 6 ; G R O T Z a . B R A U N
1993, 160 with regard to Baden-Württemberg).
B R A C Z Y K et al. (1996) question whether these concepts have overlooked important points or whether they only had explanatory power in the 1980s or whether both aspects count. They and also S T A B E R (1996) con- vincingly unravel the theoretical myth of Baden- Württemberg as being an industrial district and its economy as being characterised by flexible specialisa- tion.
Although most authors hail clustering in one way or another as an explanation for national or regional competitive advantage, there are also many scholars who stress that the same phenomenon, clustering,
may be responsible for the loss of national or regional competitive advantage ( E N R I G H T 1 9 9 5 b; G R A B H E R 1 9 9 3 a n d 1 9 9 4 ; G L A S M E I E R 1 9 9 4 ; B U T Z I N 1 9 9 1 ; R E H - F E L D 1 9 9 4 ; P O R T E R 1 9 9 0 ; B R A C Z Y K etal. 1 9 9 6 ; A S H E I M
1 9 9 6 ) . Geographically concentrated clusters can be- come insular, inward-looking systems, as m a n y old industrial areas have shown ( H A L L et al. 1 9 8 7 ; G L A S - MEIER 1 9 9 4 ; B U T Z I N 1 9 9 1 ) .
So what are the reasons then for the failure of some regional production clusters ? G R A B H E R ( 1 9 9 0 , 3 ) gives us a first broad explanation for the problem, as he states: "the initial strengths of the industrial districts of the past, the industrial atmosphere, the highly developed and specialised infrastructure, the intense inter-firm linkages, and the strong political support by regional institutions turned into heavy obstacles to innovation (the 'rigid specialization' trap)".
There are some more specific failure mechanisms of regional clusters, as well.
First, many authors point to the decrease of com- petition and domestic rivalry through ossification that might occur in regional production clusters ( P O R T E R
1990; E N R I G H T 1995b). If co-operation in regional production clusters is going too far and co-ordination is allowed to insulate firms from competitive pres- sures, incentives can become skewed, and the local- ised industry can lose its vitality ( E N R I G H T 1995 a).
Mergers and concentration can be responsible for this dwindling of competition. Close relationships between firms may eliminate the need for firms to develop certain functional specialties, such as mar- keting, that are carried out through personal rela- tionships within clusters. T h e same geographically impacted information may prevent firms from react- ing quickly and effectively to stimulus from outside the cluster. G R A B H E R (1993) has defined these kinds of failures as functional lock-in (inter-firm relationships) and cognitive lock-in (a common world-view that might confuse secular trends with cyclical down- turns). These lock-ins were not only observed in old industrial areas such as the R u h r Area ( G R A B H E R
1993), but they are also partly responsible for the inflexibility of modern industrial districts in Baden- Wiirttemberg ( B R A C Z Y K et al. 1996; M O R G A N 1996;
H E R R I G E L 1996). Although many regional production clusters might fall apart because of a decrease in competition within the cluster, others might decline through competition from other clusters (Sheffield's cutlery industry, for instance, was overtaken by Solingen, whereas Solingen is now challenged by the 'Japanese Solingen', Seki) ( E N R I G H T 1995b, 15).
Secondly, closely related to the decreasing compe- tition and dynamism is the possibility of a political
lock-in or institutional sclerosis that might come up in a regional production cluster ( H A M M a. W I E N E R T 1 9 8 9 ; G R A B H E R 1 9 9 4 ; S T O R P E R 1 9 9 5 ; K L E M M E R 1 9 8 8 ; S T O R P E R a . S C O T T 1 9 9 5 ; L Á P P L E 1 9 9 4 a n d 1 9 9 6 ) . T h e
particular and dense regional institutional tissue to- gether with the firms and workers can form a so-called self-sustaining coalition or "preperestroika consensus culture" ( G R A B H E R 1 9 9 0 , 1 1 ; H U D S O N 1 9 9 4 ; K U N Z - MANN 1 9 9 6 ) . This coalition aims at preserving existing structures and causes a loss of creativity and indige- nous development potential. Large companies do not want to give u p sites for the attraction of inward investment, as they are afraid to lose qualified em- ployees to competitors. Local authorities do not see the point in attracting inward investment or in pro- moting restructuring in another way, as large tax incomes are paid by traditional industries. T h e self- sustaining coalition also lobbies for sectoral interven- tions, which h a m p e r the restructuring process more than they support it, as they remove the incentives to take initiatives for entrepreneurs and thus paralyse competition and tranquillise large industries ( H A M M a . W I E N E R T 1 9 8 9 ) .
Thirdly, in some regional production clusters the spirit of the Schumpeterian entrepreneur might dwindle due to an increasing industrial concentration and the domination of large companies. T h e n u m b e r of management functions will decrease, social struc- tures will homogenise and the gap between workers and management will widen ( H U D S O N 1 9 9 4 ; M A I L L A T 1 9 8 8 ) . Since the workers are used to be cared for from the cradle to the grave, they lack entrepreneurial spirit ( K U N Z M A N N 1 9 9 6 ) .
Fourthly, falling demand for a cluster's product might turn a structurally strong regional economy into a weak one. This might particularly be the case in strongly defence-dependent regions. T h e decline of the minicomputer and lower defence spending, for instance, has put the Massachusetts economy or the military aerospace cluster of South California in dif- ficulties ( E N R I G H T 1995b, 15).
4 What distinguishes 'good'from 'bad' agglomerations?
Why one regional production cluster may flourish 'for ever' and another may suffer from decline is a problem modern theoretical concepts in economic geography have not sufficiently dealt with. T h a t is not to say that there are no people who have tried to ex- plain the decline of old industrial clusters (see for instance S T E I N E R 1 9 8 5 ; H Á U S S E R M A N N 1 9 9 2 ; H A M M a . W I E N E R T 1 9 8 9 ; H U D S O N 1 9 9 4 ; K L E M M E R 1 9 8 8 ;
T I C H Y 1 9 8 5 ) . In economic geography either popular concepts take 'good' agglomerations as their starting- point, such as the concepts presented in Section 2, or scholars work on explaining the decline of old industrial areas. Few, however, have done work on distinguishing 'good' from ' b a d ' agglomerations
( S T A B E R 1 9 9 6 , 3 1 2 ; S A X E N I A N 1 9 9 4 , 6 ; S T O R P E R 1 9 9 5 , 2 0 1 ; H E I D E N R E I C H 1 9 9 6 , 4 0 2 ) .
In the literature some authors distinguish 'good' from ' b a d ' agglomerations in a general way. Inter- estingly, L A P P L E ( 1 9 9 4 ) uses the milieu concept, which is based on innovative and successful regions, to explain the lack of innovativeness in declining old industrial areas. Milieux can have both a function as 'amplifier' of existing positive developments and as a 'filter' which keeps external information and new actors, which are both important for continuing renewal, outside. H e points at the thin line that exists between milieux that strengthen regional economic growth and sclerotic milieux causing regional eco- nomic decline. As milieux tend to change more slowly than industries, a sclerotic milieu can remain in a region even after the industrial structure to which it belonged already has disappeared. F R O M H O L D - E I S E - B I T H ( 1 9 9 5 ) distinguishes between 'creative milieux', in which innovative effects are gained from learning processes, and 'milieux' which are not necessarily positive in their effects. T h e main difference between creative milieux and sclerotic milieux is their open- ness or lack of openness towards the outward world.
F R O M H O L D - E I S E B I T H ( 1 9 9 5 , 3 5 ) also points to the dynamic character of milieux, which go through their own life cycle. Different kind of milieux can even be found in one region at the same time, all in their own life cycle stage. M A S K E L L and M A L M B E R G ( 1 9 9 5 )
distinguish 'good' from ' b a d ' agglomerations by pointing at their ability to ' un-learn'. T h e regions that are able to adjust their institutional endowment to meet contemporary demands of the firms require 'un-learning'. T h e process of 'un-learning' neces- sitates the disintegration and removal of formerly significant institutions which now act as a hindrance to further development. T h e r e appears a great varia- tion in the ability of regions to 'un-learn', "which makes it possible in some regions but not in others to inaugurate new and simultaneously dissolve imped- ing old institutions" ( M A S K E L L a. M A L M B E R G 1 9 9 5 , 2 5 ) . A S H E I M ( 1 9 9 6 , 4 0 ) , finally, is speaking about turning "traditional" industrial districts into "learn- ing regions" in order to avoid "lock-in" of develop- ment caused by localised path-dependency.
Some other studies present more in-depth com- parisons of 'good' and ' b a d ' agglomerations. They
focus on the conditions regions possess for the dif- fusion of technologies. These conditions, in turn, are often related to the functional specialisation of the division of labour and particularly to the economic history (crafts tradition) of a region ( H Ä U S S E R M A N N
1 9 9 2 ) .
By analysing regional economic history in regions in Germany, H E R R I G E L (1990) creates some very original and convincing explanations of regional eco- nomic growth and decline. H e distinguishes two industrial orders ( " t h e sum of practices, rules and institutions that constitute and shape the way that the production of goods and its administration takes place"), which appeared in different regions of West G e r m a n y to a varying extent.
First, the decentralised-region-based industrial order emerged in regions where pre-industrial crafts formed the basis for further economic development, such as Baden-Württemberg. A wide range of specialised S M E s supported by regional institutions provided a fruitful base for the current fluent diffusion process of new technologies, vertical disintegration and exter- nalisation (bottom-up networks) (see also S A B E L et al.
1 9 8 7 ; H E R R I G E L 1 9 9 0 ; P I O R E a . S A B E L 1 9 8 4 ) .
Secondly, the autarkic-firm-based industrial order, on the other hand, emerged in regions where no pre- industrial crafts existed, such as the R u h r Area. T h e business starters in this area had to generate their own infrastructure and supply inside the company. Conse- quently, some large companies or 'cathedrals in the desert' dominated the regional labour market, the infrastructure and institutes and did not leave space for the development of SMEs ( G R A B H E R 1 9 8 9 ; T I C H Y 1 9 8 5 ) . Of course, gradually large firms started to hive off functions to suppliers and some small firms were set up to supply large concerns. This emerged net- work economy, however, was dominated and ini- tiated by large concerns so that it can be characterised as a top-down network. This hierarchical organisa- tion structure inhibits the diffusion of knowledge and the innovativeness of firms outside the complex (see also L Ä P P L E 1 9 9 4 , 4 3 ) . It causes small firms to supply large firms by blueprint production. This makes in- house marketing and extensive R & D unnecessary.
Hence, these suppliers are restrained from shifting to more promising markets ( G R A B H E R 1 9 9 0 ) .
Similar to H E R R I G E L ' s (1990) explanations for regional economic growth differences in Germany are
S A X E N I A N ' s (1994) explanations for the differences in development between Silicon Valley and Route 128 in the USA. These leading centres of innovation in electronics in the 1970s shared two common origins:
university-based research and postwar defence spend-
ing. In the early 1980s, they both came into crisis.
In Silicon Valley a new surge of computer start-ups emerged alongside established companies, such as Hewlett-Packard, and the region regained its former vitality. Route 128, in contrast, did not recover from the crisis. In contrast to their counterparts in Silicon Valley, start-ups in Route 128 were isolated from sources of essential market information, technology and skill.
What were the reasons for these differences in development in the 1980s? According to S A X E N I A N ( 1 9 9 4 ) the answer can be found in the distinct indus- trial systems, which consist of local institutions (such as universities), culture, industrial structure (degree of vertical integration; extent and nature of links between firms) and corporate organisation. Silicon Valley has a regional network-based industrial system that promotes collective learning and flexible adjustment among specialist producers of a complex of related technologies. Dense social networks and open labour markets encourage entrepreneurship ( " . . . founda- tions of a decentralised industrial system that blurred the boundaries between social life and work, between firms, between firms and local institutions, and between managers and workers" S A X E N I A N 1 9 9 4 , 5 6 ) .
Route 128, in contrast, is dominated by a small num- ber of relatively integrated corporations (vertical inte- gration of a wide range of activities; centralised and hierarchical organisation; lack of social or commer- cial interdependencies; the main example is Digital), which form an independent firm-based industrial system.
T h e latter system is a typical mass production, Fordist type of production organisation; hierarchical struc- tures limit the ability to adapt quickly as conditions change and risk-avoidance becomes self-reinforcing as there are only a handful of successful role models to inspire potential entrepreneurs. T h e geography of the regions reinforced these divergent industrial systems.
Technology companies in Massachusetts were scat- tered widely along the Route 128 corridor, whereas in Silicon Valley, due to the valley, firms clustered in close proximity to one another ( S A X E N I A N 1994, 6 0 ) .
In contrast to the German situation explained by
H E R R I G E L ( 1 9 9 0 ) , where Baden Württemberg's net- work-based industrial system emerged in a region with pre-industrial crafts and the R u h r ' s autarkic- firm-based industrial system emerged in a region with no preindustrial crafts, this relationship is just the other way around in the two U S high-tech regions explained by S A X E N I A N ( 1 9 9 4 ) . Silicon Valley's re- gional network-based industrial system emerged in an agricultural, rural area, " a n environment that lacked indigenous industrial traditions and experi-
enced m a n a g e r s , " in which "Silicon Valley's pio- neers explicitly sought to avoid the hierarchical struc- tures of East Coast companies" ( S A X E N I A N 1994, 50).
Route 128's independent firm-based industrial system, instead, emerged in a region with a long industrial (textile, armaments, machine tool industry and later on car industry and electrical manufacturing) and cultural tradition (New England society; conservative traditions; hierarchical and authoritarian ethic of Puritanism; identities shaped by family and class backgrounds).
Despite the fruitful work that has recently been done on distinguishing 'good' from ' b a d ' agglomera- tions, it should be realised, however, that there is no deterministic relationship between the organisational form of inter-firm and firm-institution relationships in a region, on the one hand, and regional economic development on the other h a n d1' . T h e idea of this deterministic relationship, although seductive to eco- nomic geographers because it makes advising policy- makers much easier, is wrong, since the conditions under which a particular organisational network form will be successful constantly change in a rapid pace.
In today's successful showpiece Silicon Valley, for instance, a different kind of network relationships might be used than yesterday's.
5 Conclusions
M o d e r n theoretical concepts around networking and clustering share with each other the focus on the origin and development of innovation and the signifi- cance of industrial organisation and inter-firm link- ages for regional competitiveness and regional inno- vation processes ( S T E R N B E R G 1995 b; R E H F E L D 1994).
These concepts increasingly turned from 'economic' reasons for growth of new industrial agglomerations, such as product specialisation and vertical disintegra- tion of the division of labour, to 'social' and 'cultural' reasons such as intense levels of inter-firm collabora- tion, a strong sense of common industrial purpose, social consensus and extensive institutional support for innovation, skill formation and the circulation of ideas ( A M I N a. T H R I F T 1994).
S T E R N B E R G (1995a, 1995b) and T O D T L I N G (1992), however, empirically proved the limited general value of these concepts to explain regional economic development. After careful and thorough research,
S T E R N B E R G (1995 b) stated that no single modern
1 1 I owe thanks to RALF SPIELBERG for this eye-opening comment.
theoretical concept, that is product life cycle, long waves, flexible specialisation, milieux and networks, could explain the genesis and development of the main high-tech regions in the world to a satisfying extent. Also T Ô D T L I N G ( 1 9 9 2 ) had to apply different theoretical concepts of technological change to ex- plain the innovativenessof different Austrian regional economies.
T h e main shortcoming of these concepts is that they all try to explain the surge of some archetypes of industrial districts, and by doing so they have put few regions too much in the limelight so that general lessons can only be drawn to a limited extent. They depend too much on dense and historical institutional contexts to become useful as a general theory ( S T O R - P E R 1 9 9 5 ) . As a reaction, the globalisation argument has been put forward, and subsequently the globalisa- tion-régionalisation debate.
Another important shortcoming of these concepts, which formed the core topic of this article, is that they cannot distinguish between 'good' and ' b a d ' agglom- erations. In the literature work has been done on explaining either the growth of economically success- ful regions or the decline of old industrial areas, but little research has been done on distinguishing 'good' from ' b a d ' agglomerations. There are some promis- ing exceptions, mainly work recently done on the diverging development of regional production clusters
in G e r m a n y and the USA. These studies come u p with regional network-based industrial systems as the basis for 'good' agglomerations, such as Silicon Valley and the Third Italy, and independent firm- based industrial systems as the basis for ' b a d ' ones, such as Route 128 and the R u h r Area. Further re- search is needed in this direction, particularly on the origin of functional, cognitive and political lock-ins in ' b a d ' agglomerations and on how to avoid them.
Answers to these questions are needed in order to be able to warn regions of future problems, rather than to provide them with lessons learnt from current or even past success stories.
However, a whole series of recently published arti- cles on the crisis of the Baden-Württemberg industrial districts, not only show the limited explanatory power of theoretical concepts explaining regional economic change due to their static character ( H E I D E N R E I C H 1 9 9 6 ; H E R R I G E L 1 9 9 6 ; S T A B E R 1 9 9 6 ; B R A C Z Y K e t a l . 1 9 9 6 ) , but also show the thin line that exists between 'good' and ' b a d ' agglomerations, between creative milieux and sclerotic milieux. Moreover, they demonstrate how dependent the functioning of parti- cular forms of networks and milieux are on conditions that mainly lie outside the regions, such as global competition, and how careful one should be in cor- relating particular forms of networks and milieux with regional economic success or failure.
References
AMIN, A. a. ROBINS, K . (1990): T h e re-emergence of re- gional economies? T h e mythical geography of flexible accumulation. In: Environment and Planning D: Society and Space 8, 7-34.
A M I N , A . a . THRIFT, N . ( 1 9 9 4 ) : L i v i n g i n t h e G l o b a l . I n :
AMIN, A. a. THRIFT, N. (Ed.): Globalization, Institu- tions, and Regional Development in Europe. Oxford,
1 - 2 2 .
ASHEIM, B. T. (1996): " L e a r n i n g regions" in a globalised world economy: towards a new competitive advantage of industrial districts? Paper presented at the European U r b a n and Regional Studies Conference, 11-14 April, Exeter.
AYDALOT, P. a. KEEBLE, D. (Ed.) (1988): High technology industry and innovative environments: the European experience. London, New York.
BERTRAM, H . (1992): Industrieller Wandel und neue For- men der Kooperation; ein transaktionsanalytischer An- satz am Beispiel der Automobilindustrie. In: Geogra- phische Zeitschrift 80, 214-229.
BRACZYK, H . - J . ; SCHIENSTOCK, G . a . STEFFENSEN, B .
(1996): Die Regionalökonomie Baden-Württembergs - Ursachen und Grenzen des Erfolgs. In: BRACZYK, H . - J .
a. SCHIENSTOCK, G. (Ed.): Kurswechsel in der Industrie:
Lean Production in Baden-Württemberg. Stuttgart, Ber- lin, Köln, 24-51.
BUTZIN, B. (1991): Regional life cycles and problems of
r e v i t a l i s a t i o n i n t h e R u h r . I n : W I L D , T . a . JON ES, P .
(Ed.): De-industrialisation and new industrialisation in Britain and G e r m a n y . London, 186-198.
CASTELLS, M . a. HALL, P. (1994): Technopoles ofthe world:
the making of twenty-first-century industrial complexes.
London, New York.
COOKE, P. (1995): Introduction. In: COOKE, P. (Ed.): T h e Rise of the Rustbelt. London, 1-19.
DAVELAAR, E. J . (1989): Incubation and innovation: a spatial perspective. P h D thesis at the Free University of Amsterdam.
DICKEN, P . ; FORSGREN, M . a . MALMBERG, A . ( 1 9 9 4 ) : T h e
Local Embeddedness ofTransnational Corporations. In:
AMIN, A. a. THRIFT, N. (Ed.): Globalization, Institu- tions, and Regional Development in Europe. Oxford, 23-45.
DIGIOVANNA, S. (1996): Industrial Districts and Regional Economic Development: A Regulation Approach. In:
Regional Studies 30, 373-386.
D Ö H L , V . a . SAUER, D . ( 1 9 9 5 ) : N e u e U n t e r n e h m e n s s t r a t e -
gien und regionale Entwicklung. In: ISF, I N I F E S , IIS, S O F I (Ed.): J a h r b u c h sozialwissenschaftliche Technik- berichterstattung 1995; Schwerpunkt: Technik u n d Region. Berlin, 103-157.
Economist (1996a): Contract manufacturing; make it up.
T h e Economist J a n u a r y 27th 1996, 62-63.
- (1996b): Single market, single-minded. T h e Economist M a y 4th 1996, 65-66.
ENRIGHT, M . J . (1994): Regional Clusters and Firm Strat- egy. Harvard Business School Working Paper, Boston.
- (1995 a): Organization and Coordination in Geograph- ically Concentrated Industries. In: LAMOREUX, N. a.
RAFF, D. (Ed.): Coordination and Information: Histori- cal Perspectives on the Organization of Enterprise. Chi- cago, 103-142.
- (1995 b): Regional Clusters and Economic Development:
A Research Agenda. H a r v a r d Business School Working Paper, Boston.
FROMHOLD-EISEBITH, M . (1995): Das „kreative M i l i e u "
als M o t o r regionalwirtschaftlicher Entwicklung: For- schungstrends und Erfassungsmöglichkeiten. In: Geo- graphische Zeitschrift 83, 30-47.
G E U N S , R . C . VAN ( 1 9 9 0 ) : D e t r a n s f o r m a t i e v a n e e n o u d
industriegebied Wallonie, voorbeeld ofgeval apart? P h D thesis at the University of Amsterdam.
GLASMEIER, A. (1994): Flexible Districts, Flexible Regions?
T h e Institutional and Cultural Limits to Districts in an Era of Globalization and Technological Paradigm Shifts.
In: AMIN, A. a. THRIFT, N. (Ed.): Globalization, Institu- tions, and Regional Development in Europe. Oxford, 118-146.
GRABHER, G. (1989): Industrielle Innovation ohne institu- tionelle Innovation? Der U m b a u des Montankomplexes im Ruhrgebiet. Wissenschaftszentrum Berlin für Sozial- forschung.
- (1990): T h e Weakness of Strong Ties: T h e Ambivalent Role of Inter-Firm Cooperation in the Decline and Re- organization of the R u h r . Paper presented at the Work- shop " O n the Socio-Economics of Inter-Firm Coopera- t i o n " , Social Science Center Berlin (WZB), 11-13 J u n e . - (1993): T h e weakness of strong ties; the lock-in of regional development in the R u h r area. In: GRABHER, G. (Ed.):
T h e embedded firm; on the socioeconomics of industrial networks. London, New York, 255-277.
- (1994): Lob der Verschwendung: R e d u n d a n z in der Regionalentwicklung: ein sozioökonomisches Plädoyer.
Berlin.
G R O T Z , R . a . BRAUN, B . ( 1 9 9 3 ) : N e t w o r k s , m i l i e u x a n d
individual firm strategies: empirical evidence of an inno- vative S M E environment. In: Geografiska Annaler 75B,
1 4 9 - 1 6 2 .
H A L L , P . ; BREHENY, M . ; M C Q U A I D , R . a . HART, D . ( 1 9 8 7 ) :
Western Sunrise, the genesis and growth of Britain's m a j o r high tech corridor. London.
HAMM, R . a. WIENERT, H . (1989): Strukturelle Anpassung altindustrieller Regionen im internationalen Vergleich.
Rheinisch-Westfälisches Institut für Wirtschaftsforschung, Essen.
HÄUSSERMANN, H . (1992): Ö k o n o m i e und Politik in alten Industrieregionen. In: HÄUSSERMANN, H . (Ed.): Ökono- mie und Politik in alten Industrieregionen Europas: Pro- bleme der Stadt- und Regionalentwicklung in Deutsch- land, Frankreich, Großbritannien und Italien. Basel, Boston, Berlin, 10-34.
HEIDENREICH, M . (1996): Beyond Flexible Specialization:
T h e Rearrangement of Regional Production Orders in Emilia-Romagna and Baden-Württemberg. In: European Planning Studies 4, 401-419.
HERRIGEL, G. B. (1990): Industrial Organization and the Politics of Industry: Centralized and Decentralized Pro- duction in G e r m a n y . P h D thesis at Massachusetts Insti- tute of Technology, Department of Political Science.
- (1996): Crisis in G e r m a n decentralized production: un- expected rigidity and the challenge of an alternative form of flexible organization in Baden-Württemberg. In:
European U r b a n and Regional Studies 3, 33-52.
HIRST, P . a . ZEITLIN, J . ( 1 9 9 1 ) : F l e x i b l e s p e c i a l i z a t i o n
versus post-Fordism: theory, evidence and policy impli- cations. In: Economy and Society 20, 1-56.
HUDSON, R . (1994): Institutional Change, Cultural Trans- formation, and Economic Regeneration: Myths and Realities from E u r o p e ' s Old Industrial Areas. In: AMIN, A. a. THRIFT, N. (Ed.): Globalization, Institutions, and Regional Development in Europe. Oxford, 196-216.
K I L P E R , H . a . LATNIAK, E . ( 1 9 9 6 ) : E i n f l u ß f a k t o r e n b e t r i e b -
licher Innovationsprozesse - Z u r Rolle des regionalen
U m f e l d e s . I n : BRÖDNER, P . ; PEKRUHL, U . a . REHFELD, D .
(Ed.): Arbeitsteilung ohne Ende? Von den Schwierig- keiten inner- und überbetrieblicher Z u s a m m e n a r b e i t . M ü n c h e n , M e r i n g , 217-240.
KLEMMER, P. (1988): Adaption Problems of Old Industrial Areas: T h e R u h r Area as an Example. In: HESSE, J . J . (Ed.): Regional structural change and industrial policy in international perspective. Baden-Baden, 512-533.
KRÄTKE, S. (1996): Regulationstheoretische Perspektiven in der Wirtschaftsgeographie. In: Zeitschrift für Wirt- schaftsgeographie 40, 6-19.
KRUGMAN, P. (1991): Geography and T r a d e . C a m b r i d g e , Massachusetts.
KUNZMANN, K. R . (1996): Das Ruhrgebiet: alte Lasten u n d neue C h a n c e n . In: A R L ( E d . ) : Agglomerationsräume in Deutschland: Ansichten, Einsichten, Aussichten. H a n - nover, 112-153.
LAGENDIJK, A. (1996): Spatial clustering at the cross-roads of territorial and industrial development: a review. Paper presented at the E U N I T Seminar on the Territorial Dimension of Innovation, 21-23 M a y , D o r t m u n d . LAMBOOY, J . G . (1994): Networks and proximity: trans-
actions and knowledge. Paper presented at the 34th European Congress of the Regional Science Association, 23-26 August, Groningen.
LÄPPLE, D. (1994): Zwischen gestern und übermorgen: Das Ruhrgebiet - eine Industrieregion im U m b r u c h . In:
K R E I B I C H , R . ; SCHMID, A . S . ; SIEBEL, W . ; SIEVERTS, T .
a. ZLONICKY, P. (Ed.): Bauplatz Zukunft: Dispute über die Entwicklung von Industrieregionen. Essen,
3 7 - 5 1 .
- (1996): Städte im U m b r u c h : Zu den Auswirkungen des gegenwärtigen Strukturwandels auf die städtischen Öko- nomien - Das Beispiel H a m b u r g . In: A R L (Ed.): Agglo- merationsräume in Deutschland: Ansichten, Einsichten, Aussichten. H a n n o v e r , 191-217.
MAILLAT, D. (1988): T h e Role of Innovative Small and Medium-sized Enterprises and the Revival of Tradition- ally Industrial Regions. In: GIAOUTZI, M . ; NIJKAMP, P.
a. STOREY, D . J . (Ed.): Small and medium size enter- prises and regional development. London, 71-84.
MAIR, A. (1992): New Growth Poles? Just-in-Time M a n u - facturing and Local Economic Development Strategy.
Motor Industry Local Authority Network ( M I L A N ) Working Paper, Warwick.
MALECKI, E . J . (1991): Technology and economic develop- ment: the dynamics of local, regional, and national change. Harlow, Essex.
MALMBERG, A. (1990): Linkages, labour and location.
Local industrial change in an internationalized economy.
Paper presented at the Workshop " O n the Socio-Eco- nomics of Inter-Firm C o o p e r a t i o n " , Social Science Cen- ter Berlin (WZB), 11-13 J u n e .
MARSHALL, M . (1987): Long Waves of Regional Develop- ment. London.
MASKELL, P. a. MALMBERG A. (1995): Localised learning and industrial competitiveness. University of California, Berkeley Roundtable of International Economy Working
Paper No. 80, Berkeley.
MORGAN, K. (1995): T h e Learning Region: Institutions, Innovation and Regional Renewal. Paper in Planning Research No. 157, Cardiff.
- (1996): U m k e h r u n g der Z e r m ü r b u n g ? Das baden-würt- tembergische Automobil-Cluster. In: BRACZYK, H . - J . a.
SCHIENSTOCK, G. (Ed.): Kurswechsel in der Industrie:
Lean Production in Baden-Württemberg. Stuttgart, Ber- lin, Köln, 245-268.
MOULAERT, F . a . SWYNGEDOUW, E . ( 1 9 9 1 ) : R e g i o n a l D e v e l -
opment and the Geography of the Flexible Production System: Theoretical Arguments and Empirical Evidence.
In: HILPERT, U . (Ed.): Regional Innovation and De- centralization: High-tech industry and government policy. London, New York, 239-265.
O E C D (1994): Globalisation and Local & Regional C o m - petitiveness. Organisation for Economic Co-operation and Development, Paris.
P I O R E , M . J . a . SABEL, C . F . ( 1 9 8 4 ) : T h e s e c o n d i n d u s t r i a l
divide: possibilities for prosperity. New York.
PORTER, M . E. (1990): T h e Competitive Advantage of Nations. London, Basingstoke.
PYKE, F. a. SENGENBERGER, W. (Ed.) (1992): Industrial Districts and Local Economic Regeneration. Geneva.
R E H F E L D , D . ( 1 9 9 4 ) : P r o d u k t i o n s c l u s t e r u n d r ä u m l i c h e
Entwicklung. Beispiele und Konsequenzen. Institut Arbeit und Technik Working Paper, Gelsenkirchen.
RONNEBERGER, K. (1995): Von High-Tech-Regionen lernen?
In: ISF, I N I F E S , IfS, S O F I (Ed.): J a h r b u c h sozialwis- senschaftliche Technikberichterstattung 1995; Schwer- punkt: Technik und Region. Berlin, 19-78.
SABEL, C. F. (1989): T h e Reemergence of Regional Eco- nomics. Wissenschaftszentrum Berlin für Sozialfor- schung.
SABEL, C . F . ; HERRIGEL, G . B . ; D E E G , R . a . KAZIS, R .
(1987): Regional Prosperities C o m p a r e d : Massachusets and Baden-Württemberg in the 1980's. Wissenschafts- zentrum Berlin für Sozialforschung.
SAXENIAN, A. (1994): Regional advantage: culture and competition in Silicon Valley and Route 128. Cambridge, Massachusetts, London.
SCOTT A . J . a. STORPER, M . (1987): High technology in- dustry and regional developoment: a theoretical critique and reconstruction. In: International Social Science J o u r n a l 112, 215-232.
- (1992): Regional development reconsidered. In: ERNSTE, H . a. MEIER, V. (Ed.): Regional Development and Con- temporary Industrial Response: Extending Flexible Spe- cialisation. London, New York, 3 - 2 4 .
SOKOLOFF, K. L. (1995): C o m m e n t on Enright 1995a. In:
LAMOREUX, N. a. RAFF, D. (Ed.): Coordination and Information: Historical Perspectives on the Organiza- tion of Enterprise. Chicago: 142-146.
STABER, U . (1996): Accounting for Variations in the Per- formance of Industrial Districts: T h e Case of Baden- Württemberg. In: International J o u r n a l of U r b a n and Regional Research 20, 299-316.
STEINER, M . (1985): Old Industrial Areas: A Theoretical Approach. In: U r b a n Studies 22, 387-398.
STERNBERG, R . (1995a): Technologiepolitik und High-Tech Regionen - ein internationaler Vergleich. M ü n s t e r , H a m b u r g .
- (1995b): Wie entstehen High-Tech Regionen? Theo- retische Erklärungen und empirische Befunde aus fünf Industriestaaten. In: Geographische Zeitschrift 83, 48-63.
STORPER, M . (1995): T h e resurgence of regional econom- ies, ten years later: the region as a nexus of untraded interdependencies. In: European U r b a n and Regional Studies 2, 191-221.
STORPER, M . a. SCOTT, A . J . (1995): T h e wealth of regions:
market forces and policy imperatives in local and global context. In: Futures 27, 505-526.
TICHY, G. (1985): A sketch of a probabilistic modifica- tion of the product-cycle hypothesis to explain the problems of old industrial areas. In: United Nations Industrial Development Organization: International economic restructuring and the territorial community, 83-102.
TÖDTLING, F. (1990): Räumliche Differenzierung betrieb- licher Innovation: Erklärungsansätze und empirische Befunde für österreichische Regionen. Berlin.
- (1992): Technological change at the regional level: the role of location, firm structure, and strategy. In: Envi- ronment and Planning A 24, 1565-1584.
- (1994): T h e U n e v e n Landscape of Innovation Poles:
Local Embeddedness and Global Networks. In: AMIN, A.
a. THRIFT, N. (Ed.): Globalization, Institutions, and Regional Development in Europe. Oxford, 68-90.
