realizing potential: building regional organizing capacity in polycentric urban regions

http://eur.sagepub.com/European Urban and Regional Studies

http://eur.sagepub.com/content/10/2/173The online version of this article can be found at:

DOI: 10.1177/0969776403010002005

2003 10: 173European Urban and Regional StudiesEvert Meijers and Arie Romein

Realizing Potential: Building Regional Organizing Capacity in Polycentric Urban Regions

Published by:

http://www.sagepublications.com

can be found at:European Urban and Regional StudiesAdditional services and information for

http://eur.sagepub.com/cgi/alertsEmail Alerts:

http://eur.sagepub.com/subscriptionsSubscriptions:

http://www.sagepub.com/journalsReprints.navReprints:

http://www.sagepub.com/journalsPermissions.navPermissions:

http://eur.sagepub.com/content/10/2/173.refs.htmlCitations:

What is This?

- Apr 1, 2003Version of Record >>

at GEORGIAN COURT UNIV on December 4, 2014eur.sagepub.comDownloaded from at GEORGIAN COURT UNIV on December 4, 2014eur.sagepub.comDownloaded from

http://eur.sagepub.com/

http://eur.sagepub.com/content/10/2/173

http://www.sagepublications.com

http://eur.sagepub.com/cgi/alerts

http://eur.sagepub.com/subscriptions

http://www.sagepub.com/journalsReprints.nav

http://www.sagepub.com/journalsPermissions.nav

http://eur.sagepub.com/content/10/2/173.refs.html

http://eur.sagepub.com/content/10/2/173.full.pdf

http://online.sagepub.com/site/sphelp/vorhelp.xhtml



REALIZING POTENTIAL: BUILDING REGIONAL ORGANIZINGCAPACITY IN POLYCENTRIC URBAN REGIONS

Evert Meijers and Arie RomeinDelft University of Technology, The Netherlands

Polycentricity and regional development

The concept of polycentricity is becomingincreasingly popular in spatial policies on a varietyof spatial scales, ranging from the European to thelocal. Although the meaning of the concept, as wellas the purposes of the policies, differ between thesevarious scales, polycentricity in general is chieflyconsidered a means to achieve both a more balancedspatial pattern of development and a higher level ofinternational territorial competitiveness by the areaat stake. This article deals with the application ofthis concept to the regional scale, the so-called‘polycentric urban regions’.

At the regional level, polycentricity is the resultof a rather general contemporary tendency in theurban geography of advanced post-industrial

societies: the emergence of polycentric cities. InEurope, a transition can be observed since the1960s1 from urban patterns that were dominated bythe self-contained functional entities of a centralcity with its immediate hinterland, towardsnetworking urban systems of multiple centres ofresidence, employment and services. Due topredominant tendencies of spatial de-concentrationand scaling-up, most urban functions, i.e. residence,manufacturing, office-based sectors, retail,wholesale, warehousing and leisure services, haveextended over increasing territories to new suburbancentres or to places that are strategically locatedfrom a transport point of view. Generally speaking,the traditional functional hierarchy and dualitybetween the city centres and the multitude ofsuburban places is eroding in many city-regions.

Abstract

★

European Urban and Regional Studies 10(2): 173–186 Copyright © 2003 SAGE Publications0969-7764[200304]10:2; 173–186;035032 London, Thousand Oaks, CA and New Delhi, www.sagepublications.com

Regional planning for and in polycentric urban regionsmay entail certain competitive potentialities over astand-alone development of their individual cities orcity-regions.These potentialities relate to the poolingof resources, complementarities and spatial diversity.It seems that planners are increasingly aware of thesepotentialities as in several European countriesattempts are made to identify such polycentricregional systems of formerly independent and distinctcities that are located close to each other, oftenbuilding on increasing functional relationshipsbetween them. This article argues, however, that inorder actually to exploit the theoretical potentialplanning for polycentric urban regions has, one needsto do more than just identify a polycentric system onthe map. Rather, an active building of regionalorganizing capacity is needed – that is, the ability toregionally co-ordinate developments through a moreor less institutionalized framework of co-operation,

debate, negotiation and decision-making in pursuit ofinterests at the regional scale – to shape a polycentricurban region’s competitive advantages. This need forregional organizing capacity may sound obvious, but inpractice successful examples of proclaimedpolycentric urban regions developing networks forregional co-ordination and action are rather thin onthe ground. Basing our argument on evidence fromfour polycentric urban regions in North West Europe,it was found that the building of regional organizingcapacity is conditioned by a number of spatial-functional, political-institutional and cultural factors.Major constraints in the examined regions includeinstitutional fragmentation, an internal orientation ofkey persons and the lack of identification with theregion at large.

KEY WORDS ★ governance ★ polycentricity ★ regional planning

at GEORGIAN COURT UNIV on December 4, 2014eur.sagepub.comDownloaded from

www.sagepublications.com


Polycentric urban regions deserve specialattention as they are the result of this spatialreorganizing process in the specific case of regionswhere historically distinct and both administrativelyand politically independent cities are located inmore or less close proximity – say withincommuting distance – and are well connectedthrough infrastructure. These cities have coalescedboth functionally and morphologically into largerand more dispersed regional urban systems (e.g.Dieleman and Faludi, 1998a; Ghent Urban StudiesTeam, 1999; Bontje, 2001; Champion, 2001). Theerstwhile dominant hierarchical patterns offunctional relationships and mobility betweencentral cities and their respective hinterlands havebeen supplemented with more horizontal patterns ofrelationships and flows between these formerlyindependent and separate city-regions. Often citedexamples of polycentric urban regions are theRandstad in The Netherlands, the RheinRuhr inGermany and the Flemish Diamond in Belgium.Some examples can also be found in other parts ofEurope, for example the Italian Padua-Treviso-Venice and Emilia Romagna regions and theSpanish Basque Country. Such regional polycentricsystems are referred to by various concepts that arelargely synonymous with the polycentric urbanregion-concept used here, for instance ‘networkcities’ (Batten, 1995), ‘city networks’ (Camagni andSalone, 1993), ‘polynucleated metropolitan regions’(Dieleman and Faludi, 1998b) or ‘city clusters’(CEC, 1999).

Literature on the polycentric urban region is stilllimited and rather unconsolidated (Bailey andTurok, 2001). Consequently, a diversity ofsometimes more or less implicit definitions,operationalizations and approaches to this type ofurban configuration is still in circulation(Kloosterman and Musterd, 2001). The subject has,however, been attracting growing attention for morethan a decade by professionals of variousbackgrounds, including (a) academics such asgeographers, economists and social and politicalscientists, and (b) planners and policy-makers. Theway academics and policy-makers deal withpolycentric urban regions differs to some extent. Ingeneral, the concept has both anempirical/analytical and a strategic/conceptualcomponent to it. So far, empirical-analyticalresearch has strongly focused on the tenability of

the notion of the polycentric urban region as afunctional spatial entity. Mostly, this question isbeing approached by research that focuses on spatialbehaviour and mobility patterns of individuals andhouseholds (Clark and Kuijpers-Linde, 1994; Vander Laan, 1998) or firms (Camagni and Salone,1993; Batten, 1995; Lambooy, 1998). In addition,some analyses focus on the ‘regional discourse’, i.e.the process of institutionalizing such polycentricsystems in both policy and society (de Boer, 1996;Blotevogel, 1998; Knapp, 1998).

Planners and policy-makers put more emphasison strategic motives and action: these professionalsconsider the region an ‘actor’ rather than just a‘space’ (cf. Keating, 2001). Against the backgroundof the growing belief that the region is becoming themost important spatial level of internationalterritorial competition, planners, policy-makers andother stakeholders view the strengthening of itscompetitiveness as the predominant purpose ofstrategic actions. They usually refer to ‘high road’competitiveness, i.e. the creation of an environmentthat lures investments in hi-tech production andhigh level services, highly qualified manpower, andvisitors with great purchasing power. Oftenreferring to the ‘global city regions’ debate (seeScott, 2001), some planners in major polycentricurban regions sometimes even voice the ambitionthat these should be able to compete successfullywith the highest level metropolitan agglomerationssuch as New York, London or Paris as it provideseconomies of scale without incurring the costs oragglomeration diseconomies that these largemetropolises entail. Bailey and Turok (2001: 698)state that ‘the PUR (Polycentric Urban Region)concept is alleged to offer a sound basis to promoteregional economic competitiveness . . . It promotesthe advantages of stronger interaction betweenneighbouring cities to develop specialized andcomplementary assets, while avoiding large-scaleurban sprawl and destructive territorialcompetition.’

In this article we will establish a bridge betweenboth the academic approaches to polycentric urbanregions and the strategic approach of such regionsadopted by many planners and policy-makers. Weargue that for the optimal development andexploitation of the theoretical potentialities that aregional planning approach in polycentric urbanregions holds over a stand-alone development

European Urban and Regional Studies 2003 10(2)

174 EUROPEAN URBAN AND REGIONAL STUDIES 10(2)



strategy of its individual ‘member’ cities, efficientco-operation and joint policy-making on a regionalscale, i.e. regional organizing capacity, is animportant condition. Although the benefits of a co-ordinated regional planning approach have beenacknowledged by stakeholders in several polycentricurban regions (see Ipenburg and Lambregts, 2001),examples of the development of frameworks forregional co-ordination and action are rather thin onthe ground. Perhaps, the development of regionalorganizing capacity or even the further applicationof the concept of polycentric urban regions is notvery appealing under certain circumstances. This

article’s main objective is to explore the questionwhy such examples are the exception rather than therule, in spite of the advantages of a co-ordinatedregional approach. This question will be answeredby analysing the institutional, political, cultural andspatial-functional contexts of polycentric urbanregions in relation to the development of regionalorganizing capacity. For this, we use empiricalevidence from four polycentric urban regions (theRandstad, the RheinRuhr, the Flemish Diamondand Central Scotland; see Figure 1) that werestudied in the research-project called EURBANET2

on which this article builds. Our main ‘benchmark’


MEIJERS & ROMEIN: BUILDING REGIONAL ORGANIZING CAPACITY IN POLYCENTRIC URBAN REGIONS 175

Figure 1 Major polycentric urban regions in North West Europe



is the Randstad as this region is regarded as a primeor classic example of a polycentric urban region(Hohenberg and Hollen Lees, 1985; Batten, 1995)and, more importantly, is home to several past andrecent attempts to build regional organizingcapacity. Prior to this analysis, we providearguments to underpin the hypothesis that regionalplanning in and for polycentric urban regionsbenefits their competitiveness and that regionalorganizing capacity is needed for an optimal andeffective exploitation of the potentialities of regionalplanning in such regions. For this, we present aninventory of theoretical competitive potentialities ofa regional planning approach in polycentric urbanconfigurations. However, first a brief discussion ofsome examples of applications of the polycentricurban region as a planning concept is presented.

The polycentric urban region in spatialplanning policy

The European Spatial Development Perspective(ESDP) has strongly contributed to the debate onpolycentric development within the EU memberstates and has in this way encouraged theapplication of polycentricity in national or sub-national spatial planning policies, though in somecountries it was already under discussion before theESDP was published. Polycentricity in these cases isoften applied at a regional level, mostly for reasonsof international competitiveness. The integrateddevelopment of polycentric urban regions is ofteninitiated by national or sub-national stategovernments rather than by the regions itself.Examples of such a top down implementation ofthis concept are the Flemish Diamond (Brussels-Antwerp-Ghent-Leuven) (Albrechts, 1998) and theRheinRuhr (including cities such as Cologne, Bonn,Dortmund, Essen and Düsseldorf) (Blotevogel,1998; Knapp, 1998).

The Dutch national government as well paysmuch attention to polycentric urban regions, but hasbeen, in contrast, actually spurred into doing so bysome of these regions, in particular the Randstad.The government’s new spatial policy3 refers topolycentric urban regions as stedelijke netwerken(urban networks). It has designated six urbannetworks of national importance (see Figure 2). The

Randstad4 is one of these six polycentric regions andis composed of a ring of cities around a relativelyopen ‘Green Heart’, including the country’s fourlargest cities – Amsterdam, Rotterdam, The Hagueand Utrecht (see Figure 3). In the Dutch case, thepolicy is considerably far-reaching. The cities thatmake up the urban networks are required to draw upspatial programmes and plans in mutualconsultation while they also have to see to it thatthere is an integrated system of both public andprivate transport (Ministry of VROM, 2001). Thepolicy document puts great emphasis on regionalco-operation and policy-making, but remains vagueas to how that must be arranged. This holds true forthe policy documents concerning the FlemishDiamond and RheinRuhr as well.

Potentialities of a regional approach inpolycentric urban regions

In polycentric urban regions, many spatial decisionson location and mobility by the major decision-makers, firms and households take into accountwider sets of assets and broader spatial scopes thanjust individual cities. It seems plausible thereforethat policies that aim at strengthening economiccompetitiveness at the strictly local level ofindividual cities make increasingly less sense nowthese cities are becoming part and parcel of a largerfunctional geographical entity. The emergingcoherent polycentric configuration appears a moreappropriate entity for policy and planning than theindividual cities it is composed of for an increasingnumber of issues. Planning at the city or city-regionlevel leaves much room for competition withinpolycentric urban regions for investments in high-level services and hi-tech industries, for professionalworkers, for tourists, and even for a marketableimage. It leaves opportunities that are offered by thelarger regional system unutilized and may even leadto waste through, for example, duplications. Theobservation by Scott et al. (2001: 11) in their essayon the rise of global city regions that the ‘individualcity in the narrow sense is less an appropriate orviable unit of social organisation than this regionalnetworks of cities’5 is also applicable to the policyissue of territorial competitiveness.

Building on research findings of EURBANET,





three potentialities of regional co-ordination andaction in polycentric urban regions can bedistinguished. Regional co-operation and co-ordination in these regions may open the road to (1)pooling resources in order to share facilities andservices and to achieve ‘critical mass’, (2) developingand exploiting balanced complementarities and (3)optimizing spatial diversity, which mainly relates toimproving the quality of open spaces. Defined sobroadly, these three potentialities are more or lessacknowledged in all four regions examined inEURBANET.

The first potentiality of a regional approach inpolycentric urban regions is the possibility toeffectively pool assets that are spread across theregion. This provides greater agglomeration orexternal economies for businesses. On the regionalscale, businesses have access to larger and morevaried pools of labour, suppliers, and customersthan in any of its individual urban nodes orlocations. A series of interviews with majorstakeholders in the regions examined inEURBANET reveals that the pooling of highlyqualified professional labour is considered a



Figure 2 Polycentric urban regions designated in the proposed Dutch national spatial planning policy



particularly important advantage (Ipenburg andLambregts, 2001). In some cases, the pooling oflocal labour markets may solve a situation ofunemployment in one area of the region and scarcityof workers in another. The surplus value of regionalco-ordination and action may consist of policymeasures to improve labour mobility at the regionallevel.

Besides the pooling of resources, theencouraging of interaction between neighbouringcities in a polycentric urban region may result infunctional specializations. Where suchspecializations are complementary rather than

competitive, the polycentric urban region as a wholemay offer a broader package of higher quality,metropolitan services to businesses, households,consumers, workers and tourists. These servicesmay be advanced producer services; educational andR&D institutes; specialized types of retail;recreation, leisure and cultural facilities; and last butnot least residential environments (Ipenburg andLambregts, 2001). A large variety of high qualityand rapidly accessible complementary services,universities, businesses and stakeholders in apolycentric urban region creates a favourableenvironment for innovations, which is definitely an



Figure 3 The Randstad



advantage in the competition for investments. Thisis not to say that it is co-operation instead ofcompetition that creates such a set ofcomplementary local environments. Rather, it maybe co-ordination and co-operation between publicpolicy-makers, removing barriers against privatecompetition (market imperfections) that are at theroot of functional specialization of places.

The third potentiality of regional planning, theimprovement of the quality of open space, addsspatial diversity as a competitive quality to the abovementioned pooled economic resources andcomplementary facilities and amenities inpolycentric urban regions. The current tendencytowards dispersed and unbalanced patchworks of allkinds of ‘constructions, topographies and spaces,with elements of urban as well as rural landscapes’(Schmitt et al., 2001: 18) that develop in theformerly unencumbered open landscapes of manypolycentric configurations means a downgrading oftheir spatial diversity. This must be judgednegatively from the competitive point of viewbecause it harms the variety of urban scenery andrural landscapes across short distances, which isconsidered one of the basic competitive advantagesof polycentric urban regions over single largemetropolitan agglomerations. Avoiding suchuncontrolled urban sprawl, and protecting the‘green’ (and blue) networks for recreationalfunctions requires co-ordinated policy-making froma regional rather than a local perspective.

One must be aware that, though thesepotentialities are promising on a regional scale, theymay have drawbacks on a local scale. For instance,the co-ordinated development of complementaryfacilities and amenities may provide a larger varietyand higher quality of these on the regional scale, butmay also require that individual city-regions shouldmake net sacrifices by subordinating their owninterests to the greater regional good (e.g. Hospersand van Lochem, 2001). For instance, it may bedecided on the regional level that a specific city maylose to a neighbouring city a highly appraisedvocational training institute, theatre, or medicalspecialism provided by the local hospital withoutbeing sufficiently compensated for this loss inanother sector. In our view, this still makes the casefor developing an institutional framework for co-operation and co-ordination pressing, as thisframework should allow exactly for these kind of

trade-offs and a fair distribution of the regionalgood. Without such a framework, local interests willcontinue to prevail, so blocking the exploitation ofregional potentialities.

Limited regional organizing capacity inpolycentric urban regions

Notwithstanding the importance of regionalplanning in polycentric urban regions and thestrong contribution of regional organizing capacityin exploiting its potentialities, the four polycentricurban regions considered in this study show thatregional organizing capacity has hardly beendeveloped. Coupled with the eagerness of policy-makers to have their regions taken seriously byoutsiders, it is striking that there is only limitedinterest in the development of this capacity. Forinstance, initiatives to develop regional organizingcapacity in the RheinRuhr – both morphologicallyand functionally a coherent entity (GEMACA,1996) – are only limited so far. The same holds truefor the Flemish Diamond. Although this polycentricurban region does appear in planning documents,one can hardly see the beginning of a developingregional organizing capacity. Albrechts (1998)concludes that institutional coherence and co-operation within this region are rather weaklydeveloped.

While the Randstad as a planning concept hasoccupied a central position in national planningstrategies for at least the last 40 years, even thereattempts to actually build regional organizingcapacity have not been successful. Most of theattempts have entailed the introduction of a formal(fourth) government tier in between the municipaland provincial tiers, but none of them have provento be politically acceptable. The traditional Dutchthree-tier system (national government, provinces,municipalities) has proved to be resistant to changes,making clear that regional organizing capacityshould be based on voluntary co-operation betweenthe three tiers. Recent bottom-up attempts, some ofthem specifically aimed at improving the nationaland international competitiveness of the Randstad,have provided a new impulse for building regionalorganizing capacity in the Randstad. Most notableare the Bureau Regio Randstad (Randstad Region





Agency), an agency that supports the co-ordinationof policies by the four provinces in the Randstad6,and the Delta Metropolis Association, a ratherinformal body in which city authorities, districtwater boards, chambers of commerce and a varietyof other semi-public and private institutions meetand discuss the way the Randstad can develop into acompetitive European metropolis.

Now that the national government has publishedits proposals for a new spatial planning policy, thereis some discussion about building the organizingcapacity in the polycentric urban regions identifiedin the document. The four Randstad provinces andthe mayors of the four largest cities have expressedtheir interest in designing a co-operative structurein which the provinces, the four largest cities andthe four matching formal city regions arerepresented. Tasks could include putting Randstad-scale projects on the regional and national agendaand defining a way of elaborating and implementingthem (Bureau Regio Randstad, 2001). However, thevalue of these more recent initiatives still has to beproven.

Context-bound constraints to the buildingof regional organizing capacity

Apparently, applying the concept of polycentricurban regions and acknowledging the potentialitiesof regional planning in such regions do notconsequently result in the development of regionalinstitutional frameworks for co-operation and co-ordination. Here we identify the reasons for this interms of constraints for building regional organizingcapacity in polycentric urban regions, which we baseon our experience in the EURBANET project. Aclear starting point lies in the analysis of the currentpolitical, institutional, cultural and spatial contextsof polycentric urban regions and the way theseinterfere with the building of regional organizingcapacity.

In building a more general framework to assessthe possible constraints for the development ofregional organizing capacity in polycentric urbanregions, we can partly build on previous researchconcerned with factors determining the potential forsuch developments. For instance, Van den Berg andBraun (1999) list seven factors contributing to urban

organizing capacity, which are: the formalinstitutional framework (the administrativeorganization), strategic networks, leadership, visionand strategy, spatial-economic conditions, politicalsupport, and, finally, social support. Keating (2001:379) describes the concept of a developmentcoalition; a place-based interclass coalition dedicatedto economic development in a specific location.Keating claims that the context for building such adevelopment coalition is determined by the currentcompetitive situation of the region, but also byfactors such as culture, institutions, leadership,social composition and external relations. Withrespect to regions constituting themselves as anactor, he remarks that institutions, leadership and anability to carry a definition of the interests of theregion are required. While the focus here is on theregional rather than the urban level and is morebroadly defined than just a development coalition,the factors named above provide clues to thecomponents of the framework needed in order toanalyse the context for building regional organizingcapacity. Irrespective of the precise categorizationused above, and basing ourselves on theEURBANET experiences, we are able to deducethree general dimensions that play a role indetermining the feasibility of the building ofregional organizing capacity in a polycentric urbanregion: the spatial-functional dimension (which isquite specific for polycentric urban regions), thepolitical-institutional dimension, and the culturaldimension.

Spatial-functional dimension

The further a polycentric urban region isfunctionally tied together and integrated, thegreater the need for regional organizing capacity.Clearly, there must be some spatial logic, orfunctional rationality behind the formation ofregional organizing capacity. This means that actors(enterprises, public and private institutions,households) have to maintain relations throughoutthe region to fulfil their needs for production,consumption and personal needs. Authors dealingwith this spatial-functional dimension often centreon daily flows of people, translated into ‘travel towork areas’, ‘daily urban systems’ (Van der Laan,





1998) or ‘functional urban regions’ (Cheshire andHay, 1989). But the functional dimensionencompasses more. According to Pumain (1999: 6),a functional urban unit presents ‘a concentration ofpeople, activity, capital and buildings, constituted bymarkets of, for example, labour, retail, services,culture or housing. It is structured on major roads,railroads and terminals and functions by flows ofpeople, goods, energy, information and money.’Consequently, whether or not a polycentric urbanregion functions as a functional entity can bededuced from the spatial scopes of markets,infrastructure and flows.

The required functional rationality behind thedevelopment of regional organizing capacity doesnot mean that a polycentric urban region needs to beone single, compact, functionally coherent and‘closed’ system. Polycentric urban systems tend tobe open and multi-layered complexes of nodes,networks, flows and interactions of global, regionaland local scales (Albrechts, 2001). Consequently, thespatial scope and spatial orientation of interactionsbetween places do not coincide exclusively with thepolycentric system as a whole, but vary considerablybetween types of interactions and are dynamic. Forsome, the region has become one single polycentric‘urban field’ but for others it is either too large ortoo small. In a recent analysis of the labour marketof the Randstad, Van Ham et al. (2001) show that ithas become one single labour market for some

groups of highly qualified, high income and mobileprofessionals, but is sub-divided into several sub-regional labour markets for other workers. Thelabour markets of lowly or unskilled and less mobileworkers do not even surpass the level of individualcities and their immediate hinterlands. Similarconclusions can be drawn for other functions, likehousing, leisure and recreational services: the urbansystem is one single market only for some selectivegroups of its population. Analysing the Randstadpresents a clear picture: regional spatial relationsincrease, strengthen and get more dispersed, whilethe spatial scope of functional markets (in particularfor labour, shopping, social activities, leisure andsports) is widening to a more regional scale, eventhough this is often not the scale of the entireRandstad (Meijers et al., 2003). Similar tendenciescan be found in the Flemish Diamond and theRheinRuhr. Figure 4 presents a view on the increasein travel for social visits between sub-regions in theRandstad. Social visits account for a large share(about 15 percent in 1999) of the total number oftrips made.

The fact that polycentric urban regions cannotbe defined as single, closed functional units is alsoreflected in the multitude of interactions that theircities maintain with cities in other regions. Some oftheir economic clusters, particularly those related tomain ports (seaports, airports, or high-speed trainstations), are connected more strongly within



Figure 4 Travel for social visits between Corop-regions in the Randstad, 1986 (left) and 1998 (right). Based on data CBS (OVG)



international rather than regional networks. Policiesand planning regarding the competitiveness of theseclusters are not primarily formulated on the level ofthe polycentric urban region. This means that theregion is the appropriate platform to formulate andimplement policies for only some of the spatial issues,while others can be better dealt with on other levels.Keeping these necessary differentiations in mind, wenevertheless claim that, on the basis of the spatial-functional relations and their tendencies, there isreason for building regional organizing capacity inthe polycentric urban regions examined. The scopeof the regional co-ordination, however, should beselective with regard to the spatial issue at stake.

Political-institutional dimension

The attitude and vision towards government, and inparticular spatial planning by administrators in thepolycentric urban region, are not factors to beneglected when it comes to the development ofregional organizing capacity. In general, manyspatial issues these days call for an approach that isformulated and implemented at multiple scales andacross several administrative tiers rather than atonly one. Additionally, an increasing number ofspatial issues are, or preferably should be, addressedthrough a governance rather than a governmentalmode. Regional co-ordination in polycentric urbanregions requires the politicians and administratorsto adopt a view on government and planning thatputs emphasis on co-operation across administrativetiers and sectors and between public, private andorganized interest groups, thereby taking intoaccount that different issues call for differentalliances with different spatial competencies anddifferent life spans (cf. Boelens, 2000). Clearly,governing a polycentric urban region is an intricateaffair. Putting such multi-level governance intopractice is a complex task, even if politicians andadministrators agree on its usefulness. There maybe a lack of understanding on how multi-levelgovernance works. Perhaps this partly explains thevagueness on the political-institutional dimensionfor polycentric urban regions in the national Dutchplanning policy.

Besides the attitude and vision of administratorsand politicians, the formal institutional framework is

critical here as well. The question is whether or notthis framework leaves room for regional co-ordination. Often, it needs to be adjusted to be ableto cope with the interfering and multi-level natureof urban dynamics. The existing frameworks areoften too static and hierarchical to recognize anddeal with this complex, multi-scalar interplay oftrends and forces. As mentioned above, attemptshave been made in the Randstad to add a formaladministrative tier for a long time, but the existinginstitutional and political structure has not allowedit. It has become slowly apparent that multi-levelgovernance requires co-operation across scales andacross actors, including private actors. Thenecessary adjustments (e.g. legislation, a formalredistribution of competencies) are only graduallybeing implemented.

Albrechts (2001: 734) characterizes polycentricurban regions as ‘socio-spatial conflict zones for thearticulation of multiple interests, identities andcultural differences’. While turning to identities andcultural differences below, here we deal with thedifferent interests in a polycentric urban region.There appears to be a clear need to establishcommon and shared interests for the polycentricurban region. This is far from self-evident as thereare many fields where the interests of places andstakeholders in a polycentric urban region aredifferent or even opposed. Regional disparities(between central cities and between a central cityand its suburban nodes) in, for instance,demographic and economic growth rates, socialproblems like poverty and unemployment, and inthe attractiveness of residential environments meanthat the areas better off in these matters have nointerest in adopting regional policies that may adjustthis situation. Actors in the Randstad have shownthat, to a certain extent, they are capable of definingregional interests. External incentives to do so havebeen important. These stimuli include the need toposition the region externally as a metropolis andthe way the national Dutch government fundsprogrammes or projects (chances for investmentsincrease if the region successfully co-operates andclearly defines a desired and prioritized investmentprogramme).

Finally, leadership is important. The region’scommon and shared interests need to be picked upby leaders willing and able to build on them.Networks in particular need leadership as they lack





a formal hierarchical structure. Such leadership canrely on the specific competencies of key figures andkey institutions or on the charisma of public orprivate individuals (Van den Berg and Braun, 1999).The Delta Metropolis Association, for instance, wasinitiated by a professor who convinced the mayors ofthe largest cities to start a regional discourse on thefuture of the Randstad. Soon other stakeholders inthe region followed and the scope expanded.

Cultural dimension

In recent thinking on urban and regionaldevelopment, much emphasis has been placed onthe cultural dimension. Here, the culturaldimension is concerned with the feeling ofbelonging together and the creation of culturalelements that help in perceiving the polycentricurban region as an entity. Social relationships,shared understandings, and norms of co-operationand reciprocity all ease regional networking. Sharpcultural divides, on the other hand, impose barriersto co-operation. Cultural discontinuities possiblyreduce the opportunities for relationships andinteraction. Again referring to the quote ofAlbrechts (2001: 734) above, polycentric urbanregions are regions in which potentially discordantmultiple identities and cultural differences occur.

Following the distinction between culture andidentity, we can distinguish two elements. The firstone is a common culture and refers to the existenceof a shared history and shared values, norms andbeliefs in a region. Major sources of culturaldifferences are language, ethnicity, religion andpolitical preferences. The second element is regionalidentity. This is a concept that is primarily a socialconstruct and therefore a dynamic phenomenon.Moreover, it is a contextual and multi-layeredconcept. One belongs to many groups that togetherfurnish one with a whole variety of discreteidentities which vary in relative or contextualimportance. Some of these are linked to ageographical entity, for instance the neighbourhood,city or country one lives in, but probably also theregion. The existence of such a regional identity inthe polycentric urban region helps to generate socialsupport for regional co-ordination and action.According to Faludi (1999), a common identity helps

to achieve common, functional or strategic goals.Cultural divides can be present in polycentric

urban regions even if their scale is relatively small.For instance, experiences in the polycentric urbanregions of Central Scotland and the RheinRuhrshow that cultural, if not psychological, cleavageshamper the building of regional organizing capacity.There are strong cultural cleavages between theEdinburgh and Glasgow urban areas and the Rhineand Ruhr areas respectively. In both regions, themost affluent areas, being the Rhine and Edinburgh,are not very enthusiastic about being identified withareas with a reputation for economic downturn,unemployment, and environmental and socialproblems. Moreover, the lack of a ‘regionaldiscourse’ in the RheinRuhr adversely affectsregional organizing capacities. Lurking culturaldivides can be found in the Flemish Diamond aswell, as this polycentric urban region extends overan area consisting of two regions (Flanders andBrussels) that are not culturally homogeneous, forinstance with respect to language. The Randstadperforms comparatively better when it comes to thecultural dimension, at least in the sense that thereare no major cultural cleavages present.

As polycentric urban regions are notinfrequently the result of strategic thinking byplanners in a top-down manner, the regionalidentity in such regions is, in general, weak. Thoughthe Dutch language has an official Dutch word forinhabitants of the Randstad (‘Randstedeling’)suggesting some homogeneity, this does not meanthat the population and administrators have aregional ‘Randstad’ identity. This is because theRandstad lacks identifying power for various reasons:

• there is no undisputed official boundary of theregion;

• there are no symbols (for instance specialbuildings, or a Randstad soccer team) connectedto the Randstad-scale, except for one, itsmorphological form;

• institutions (public, private, social) do not takethe Randstad as their territorial organizingprinciple, therefore not reproducing theRandstad concept in daily life;

• the Randstad is not a political arena, which alsomeans that the media do not operate on theRandstad-scale (no newspapers or regionaltelevision channels).





This lack of territorial, symbolic andinstitutional shape (see Paasi, 1996) and politicalspace (see Keating, 1997) in the Randstad has notprevented the establishment of this region inpeople’s consciousness, especially in the way theymentally structure their spatial environment. So,while cultural factors hamper the development ofregional organizing capacity in Central Scotland,the Flemish Diamond and the RheinRuhr, they playa rather neutral role in the case of the Randstad.

Conclusions

Like other interpretations of the principle ofpolycentricity, the concept of the polycentric urbanregion appeals very much to spatial policy-makers,and for good reasons. In theory, the regional co-ordinated spatial development embodied by theconcept entails certain potentialities over a stand-alone development strategy of the cities within sucha region. These potentialities are:

• to pool resources in order to share facilities andservices and to achieve critical mass;

• to develop and exploit balanced complementarities;• to optimize spatial diversity and to better protect

the quality of open spaces.

By arguing that all these potentialities requireregional co-ordination and policy-making to createthe best result, we have made plausible our beliefthat the challenge for actors in polycentric urbanregions is to generate the regional organizingcapacity to be able to seize and use thesepotentialities. Regional organizing capacity involvesregional co-ordination through regional policynetworks, i.e. some kind of co-operative regionalforum, in which all relevant stakeholders in apolycentric urban region (different public actors,but also private market parties and non-governmentalorganizations) meet, discuss and decide uponplanning policies and their implementation.

Despite the seemingly apparent advantages ofbuilding regional organizing capacity to makeeffective and optimal use of the potentialities ofpolycentric urban regions, examples of such regionswhere this takes place are rather exceptional. Thisarticle has explored the reasons for this. Basing ourargument on evidence from four polycentric urban

regions in North West Europe, and in particular theRandstad, we found that the building of regionalorganizing capacity is conditioned by the spatial-functional, political-institutional and culturalcontext of the region. In general, the mainconstraints found in these regions can becategorized as institutional fragmentation combinedwith an internal orientation of key persons (such aspoliticians and policy-makers) and the lack ofidentification with the region at large.

Actors in both polycentric urban regions and atother levels (e.g. national) who feel attracted toapplying the principle of polycentricity and itsregional interpretation in their spatial developmentpolicies need to be aware that this involves newforms of regional co-ordination and the creation ofregional organizing capacity. Even this does notguarantee success – that is, the realization of thetheoretical potentialities of polycentric urbanregions – as success is also dependent on thefunctioning of such co-ordinating networks orpartnerships. However, without regional organizingcapacity, there is a danger that the concept of thepolycentric urban region will remain a rather emptyone. Reviewing current policies introducing theconcept of the polycentric urban region, it appearsthat the building of regional organizing capacity toimplement it is often simply forgotten.

Besides being aware of the need for regionalorganizing capacity, these actors must also bemindful of the possible contextual constraints theymay encounter in developing capacity and must dealwith them strategically. Lack of functional, culturalor political-institutional coherence does not meanthat the building of regional organizing capacity isimpossible. It rather poses limits on what isachievable in the first stage. In the situation wherelarge political-institutional and or culturalconstraints exist, and where the functionalcoherence may also be relatively limited, the beststart is a small start. In such a case, voluntary co-operation between a limited number of actors onsimple, not too sensitive issues or well-definedprojects with clear benefits to the individual actorsis best. Mutual trust, understanding and strongerworking relationships are likely to evolve, thusenabling the addressing of more complex policiesand projects in a later stage. In cases such as theRandstad, where constraints are relatively limited,more structured co-operation yields more





advantages. Here, regional organizing capacity inthe truest sense of the word must be developed tomake possible ongoing deliberation, debate,negotiation and decision-making by all interestedparties on a wide variety of more or less complexprojects and policies that benefit the competitivenessof the region as a whole. This does, however,sometimes require concessions to the wider regionalinterest by individual actors. Some first attempts inthis direction deserve our future attention.

Acknowledgement

The authors would like to thank Bart Lambregts forhis valuable comments on an earlier version of thepaper.

Notes

1 In the USA, such processes could be observed severaldecades earlier.

2 The EURBANET-project was part of the INTERREGIIC programme of the North Western Metropolitan Areaand co-funded by the European Commission.

3 The Fifth Memorandum on Spatial Planning still has tobe considered by the Parliament. Its current status is thatof a policy proposal. However, the concept of urbannetworks is well received.

4 The Fifth Memorandum refers to the Randstad as theDeltametropolis. We prefer to stick to the ‘old’ name fortwo reasons: the Memorandum has not yet passedParliament and the public of foreign readers is morefamiliar with it.

5 Scott et al. mention the polynuclear Randstad Holland asan example of a global city region.

6 These are the provinces of North Holland, SouthHolland, Utrecht and Flevoland.

References

Albrechts, L. (1998) ‘The Flemish Diamond: PreciousGem and Virgin Area’, European Planning Studies 6 (4):411–24.

Albrechts, L. (2001) ‘How to Proceed from Image andDiscourse to Action: As Applied to the FlemishDiamond’, Urban Studies 38 (4): 733–45.

Bailey, N. and Turok, I. (2001) ‘Central Scotland as a

Polycentric Urban Region: Useful Planning Concept orChimera?’, Urban Studies 38 (4): 697–715.

Batten, D.F. (1995) ‘Network Cities: Creative UrbanAgglomerations for the 21st Century’, Urban Studies 32(2): 313–27.

Berg, van den L. and Braun, E. (1999) ‘UrbanCompetitiveness, Marketing and the Need forOrganising Capacity’, Urban Studies 36 (5–6): 987–99.

Blotevogel, H.H. (1998) ‘The Rhine-Ruhr MetropolitanRegion: Reality and Discourse’, European PlanningStudies 6 (4): 395–410.

Boelens, L. (2000) ‘Inleiding’, in L. Boelens (ed.) NederlandNetwerkenland, pp. 5–12. Rotterdam: NAi Uitgevers.

Boer, N. de (1996) De Randstad bestaat niet, de onmacht totgrootstedelijk beleid. Rotterdam: NAi Uitgevers.

Bontje, M. (2001) ‘Dealing with Deconcentration:Population Deconcentration and Planning Response inPolynucleated Urban Regions in Northwest-Europe’,Urban Studies 38 (4): 769–85.

Bundesministerium für Raumordnung, Bauwesen undStädtebau (ed.) (1995) RaumordnungspolitischerHandlungsrahmen. Bonn: Selbstverlag der BfLR.

Bureau Regio Randstad (2001) NL/EU: From Randstad toDelta Metropolis: A Contribution to the Discussion of theCooperation in the Delta Metropolis by the Queen’sCommissioners of the Provinces of South Holland, NorthHolland, Utrecht and Flevoland and the Mayors ofAmsterdam, Rotterdam, The Hague and Utrecht. Utrecht:BRR.

Camagni, R. and Salone, C. (1993) ‘Network UrbanStructures in Northern Italy: Elements for aTheoretical Framework’, Urban Studies 30 (6): 1053–64.

CEC (Commission of the European Communities) (1999)European Spatial Development Perspective: TowardsBalanced and Sustainable Development of the Territory ofthe EU. Luxembourg: Office for Official Publications ofthe European Communities.

Champion, T. (2001) ‘Urbanization, Suburbanization,Counterurbanization and Reurbanization’, in R.Paddison (ed.) Handbook of Urban Studies, pp. 143–61.London: Sage.

Cheshire, P.C. and Hay, D.G. (1989) Urban Problems inWestern Europe: An Economic Analysis. London: UnwinHyman.

Clark, W.A.V. and Kuijpers-Linde, M. (1994)‘Commuting in Restructuring Urban Regions’, UrbanStudies 31 (3): 465–83.

Dieleman, F.M. and Faludi, A. (1998a) ‘Randstad, Rhine-Ruhr and Flemish Diamond as One PolynucleatedMacro-region’, Tijdschrift voor Economische en SocialeGeografie 89 (3): 320–27.

Dieleman, F.M. and Faludi, A. (1998b) ‘PolynucleatedMetropolitan Regions in Northwest Europe: Theme ofthe Special Issue’, European Planning Studies 6 (4):365–77.





Faludi, A. (1999) De architectuur van de Europeseruimtelijke ontwikkelingspolitiek. Nijmegen: KatholiekeUniversiteit Nijmegen.

Faludi, A. and Waterhout, B. (2001) Evaluation of theESDP. Report produced within the framework of theEURBANET project. Nijmegen: KatholiekeUniversiteit Nijmegen.

GEMACA (Group for European Metropolitan AreasComparative Analysis) (1996) North-West EuropeanMetropolitan Regions. Paris: IAURIF.

Ghent Urban Studies Team (1999) The Urban Condition:Space, Community and Self in the ContemporaryMetropolis. Rotterdam: 010 Publishers.

Ham, van M., Hooimeijer, P. and Mulder, C.H. (2001)‘Urban Form and Job Access: Disparate Realities in theRandstad’, Tijdschrift voor Economische en SocialeGeografie 92 (2): 231–46.

Hohenberg, P. and Hollen Lees, L. (1985) The Making ofUrban Europe 1000–1994. Cambridge, MA: HarvardUniversity Press.

Hospers, G.-J. and Lochem, R. van (2001) ‘Stedelijknetwerk Twente: over economie, het spel en de spelers’,in F. Boekema and E. Kuijpers (eds) StedelijkeNetwerken; theoretische en empirische reflecties. Nieuweideeën in Nederlands ruimtelijk onderzoek, pp. 65–76.Maastricht: Shaker Publishing.

Ipenburg, D. and Lambregts, B. (eds) (2001) PolynuclearUrban Regions in North West Europe: A Survey of KeyActor Views. Delft: Delft University Press.

Keating, M. (1997) ‘The Invention of Regions: PoliticalRestructuring and Territorial Government in Europe’,Environment and Planning C: Government and Policy 15(4): 383–98.

Keating, M. (2001) ‘Governing Cities and Regions:Territorial Restructuring in a Global Age’, in A.J. Scott(ed.) Global City Regions, Trends, Theory, Policy, pp.371–90. Oxford: Oxford University Press.

Kloosterman, R.C. and Musterd, S. (2001) ‘ThePolycentric Urban Region: Towards a ResearchAgenda’, Urban Studies 38 (4): 623–33.

Knapp, W. (1998) ‘The Rhine-Ruhr Area inTransformation: Towards a European MetropolitanRegion?’, European Planning Studies 6 (4): 379–93.

Laan, L. van der (1998) ‘Changing Urban Systems: An

Empirical Analysis at Two Spatial Levels’, RegionalStudies 32 (3): 235–47.

Lambooy, J.G. (1998) ‘Polynucleation and EconomicDevelopment: The Randstad’, European PlanningStudies 6 (4): 457–66.

Meijers, E., Romein, A. and Hoppenbrouwer, E. (2003)Planning Polycentric Urban Regions in North WestEurope: Value, Feasibility and Design. Housing andUrban Policy Studies 25. Delft: Delft University Press.

Ministry of Housing, Spatial Planning and theEnvironment (VROM) (2001) Ruimte maken, ruimtedelen, Vijfde Nota over de Ruimtelijke Ordening (FifthMemorandum on Spatial Planning), deel 3:kabinetsstandpunt. The Hague: SDU.

Paasi, A. (1996) Territories, Boundaries and Consciousness:The Changing Geographies of the Finnish-Russian Border.Chichester: Wiley.

Pumain, D. (1999) Summary Report on 2.2 Typology ofCities and Urban-Rural Relationships, version17/12/1999, Study Programme on European SpatialPlanning 1998–9, Reports from the Working Groups.

Schmitt, P., Knapp, W. and Kunzmann, K. (2001)Regional Report on the RheinRuhr Area. Reportproduced within the framework of the EURBANETproject. Dortmund: Institut für Landes- undStadtentwicklungsforschung des Landes Nordrhein-Westfalen.

Scott, A.J. (ed.) (2001) Global City Regions, Trends,Theory, Policy. Oxford: Oxford University Press.

Scott, A.J., Agnew, J., Soja, E.W. and Storper, M. (2001)‘Global City Regions’, in A.J. Scott (ed.) Global CityRegions, Trends, Theory, Policy, pp. 11–30. Oxford:Oxford University Press.

Correspondence to:

Evert Meijers, Delft University of Technology,OTB Research Institute for Housing, Urban andMobility Studies, PO Box 5030, 2600 GA Delft,The Netherlands. [email: [email protected]]





realizing potential: building regional organizing capacity in polycentric urban regions

Documents