Polycentric regions: Proposals for a new typology and terminology

Abstract

Chinese

‘Polycentricity on a regional scale’, or in short the ‘polycentric region’ (PR), is a key concept in both urban and regional research and planning. But the term is used in many different ways, making it an overstretched concept unsuitable for drawing normative conclusions from. This article highlights the fact that, beyond the lowest common denominator – PRs consist of several centres (‘poly’ = many) and are situated on a regional scale (‘region’) – there is no single generally accepted definition of what a PR is. Instead, there are various historical trajectories for the formation of PRs, as well as different substantive perspectives. In this contribution, we will develop a conceptual typology and terminology describing the various dimensions of a PR for the North-Western European context on the basis of a critical literature review. The typology addresses two formation modes, four substantive perspectives and two time perspectives in an attempt to establish conceptual ‘building block(s)’ for use in empirical PR research and policy formulation. Our typology exposes knowledge gaps to be addressed in a recast PR research agenda.

Keywords

agglomeration/urbanisation economic processes polycentricity review theory urban regions

Introduction

‘Polycentricity’ in the broad sense, and ‘polycentric regions’ (PR) (or polycentricity on a regional scale) in the narrow sense, are key concepts in both urban and regional research and planning. But despite the fact that the concept has been discussed in the literature for several decades, it is used in many different ways, thus remaining a fuzzy and overstretched concept (Burger and Meijers, 2012; Van Meeteren et al., 2016). Up to now, no unified understanding of the concept of polycentricity or of PRs has existed in urban studies and related disciplines (or in planning policy). The only consensus is on a minimum definition. A PR is a spatially contiguous area:

with several (= ‘poly’) – at least two – centres

positioned between a nation state and an individual city (region).

Beyond this minimum consensus, there is a very wide and interdisciplinary spectrum of highly different approaches to the definition and operationalisation of the concept. Existing side by side, these approaches can lead to entirely different assessments of the degree of polycentricity of the area in question, creating the risk of conceptual misunderstandings and a lack of determinacy of the object under investigation (Burger and Meijers, 2012; Davoudi, 2003; Rauhut, 2017; Van Meeteren et al., 2016). At the same time, polycentricity in planning practice and policy has positive connotations and is often perceived as something good per se, e.g. a balanced distribution of demographic and economic growth in a polycentric system is often identified as a suitable middle way between excessive density (mega-urbanisation) and urban sprawl (WGBU, 2016). However, without first defining the concept or positioning oneself within it, wrong conclusions may be reached.

Numerous authors are thus calling for a conceptual clarification, as well as concrete typologies and operationalisations (e.g. Hoyler et al., 2008: 10ff; Lambregts, 2009; Rauhut, 2017), posing such questions as: Does the definitional core of a PR include the supposition that the region’s main centres are more or less the same size (see e.g. Kloosterman and Musterd, 2001)? Is it just a matter of the presence of centres or also of their functional interconnection (e.g. Burger and Meijers, 2012)? What characterises the ‘regional’ scale in relation to the phenomenon of polycentricity (e.g. Rauhut, 2017)? These questions are answered very differently in the literature, with the concept still used so broadly that – depending on which definition is taken as the basis in the individual case – every urban region in the global North can potentially be characterised as polycentric.

In this article, we argue that this is due to the fact that many studies and planning concepts use the term ‘polycentricity’ without reflection. In addition to the typologies already called for to systematise the concept, it also seems necessary to stop using the term for the concept in its entire diversity, and instead to develop different terminologies for its individual dimensions. The objective of this article is to develop such a terminology based on a systematic typology of the different dimensions of polycentricity on a regional scale, thus contributing to the clearer use of the concept of polycentricity and its sub-concepts in the literature and providing a common language on PRs for both urban and regional research and planning practice. Furthermore, we intend to use this typology as a starting point for a memorandum on issues to be addressed in a recast research agenda on PRs.

The article is structured as follows: the main section constructs a typology and proposes a terminology for the multi-dimensional concept of PRs via a critical review of the comprehensive body of literature: we look first at the two central modes of the formation of PRs, then at the four key substantive perspectives characterising them. Additionally, the different perspectives of the concept are illustrated through empirical spotlights on two German regions (Rhine-Ruhr and Munich), with a view to substantiating our call for a differentiated terminology. In this main section, we discuss the conceptual conclusions of our literature review with respect to the question of how to deal with the overstretched concept of PR in empirical research and planning practice in the future. This results in a new conceptual typology regarding the multi-dimensional PRs concept. The final section aims to contribute to a recast research agenda on PRs.

The multi-dimensional concept of polycentric regions

Over and above the minimum characteristics of a PR (at least two centres and a minimum – regional – size; see above), up to now no unified understanding of what is meant by a PR has existed. In an attempt to bring light to the concept, we start by describing and discussing the two main historical trajectories of the formation of a PR. We then identify the four key substantive perspectives of the concept of polycentricity. This forms the basis for a new conceptual typology and terminology for the PR concept. Initially applicable to the North-Western European context, its transferability to other geographical contexts is obviously possible, but subject to additional research.

Polycentric regions – formation modes

Reference is made to the concept of polycentricity at all spatial scales, from the neighbourhood scale to the supranational scale (Rauhut, 2017): at the scale of the city, for example, with regard to the local provision of goods for the daily needs of the population or in the sense of a city of short distances; or at the European scale in the ‘European Spatial Development Perspective (ESDP) 1999’ and subsequently the Territorial Agenda (2007 and 2011) as a strategy for promoting balanced territorial development within the EU. Our typology of the concept of PRs is restricted to the regional scale (in a European context). To describe polycentric (sometimes also referred to as ‘polynucleated’) spatial structures at this scale, a wide variety of designations are used side by side: ‘polycentric urban region’ (Kloosterman and Musterd, 2001; Meijers, 2005), ‘global city-region’ (Scott, 2001), ‘mega-city region’ (Hall and Pain, 2006) or, in a German-speaking context, the term ‘Polyzentrale Metropolregion’ (in English: ‘polycentric metropolitan region’).

The extensive literature distinguishes between two essential historical trajectories of the formation of a PR. The term ‘intra-urban’ polycentricity is used to characterise polycentric spatial structures that have come into being in the process of post-industrial or post-suburban developments in the hinterlands of previously monocentric core cities; whereas ‘inter-urban’ PRs are regions that have developed through an increasing functional interconnection of spatially close core cities, as well as the increasing overlap of their respective hinterlands (Kloosterman and Musterd, 2001).

Formation mode 1: Inter-urban polycentric regions or ‘polycentric urban regions’ historically showing multiple cores

Inter-urban PRs – in the literature, frequently also referred to as ‘polycentric urban regions’ (PURs) – are a spatially selective phenomenon. Only a few metropolitan regions can be characterised as such. According to Kloosterman and Musterd (2001), inter-urban PRs are regions consisting of several both historically and administratively distinguishable larger and smaller cities in immediate proximity (generally within daily commuting distance) to one another, as well as their respective – intra-urban polycentric – hinterlands. The larger cities of these regions do not differ significantly in terms of their population size and their economic and political importance. In addition, inter-urban PRs are characterised by the fact that the core cities are increasingly intertwined, with their surrounding regions fused into a larger regional urban system (Champion, 2001; Parr, 2014). Compared to urban regions with a clearly dominant core, inter-urban PRs exhibit very little in the way of developed spatial structural hierarchies, hence not showing any obvious point of crystallisation for the emergence of (both positive and negative) agglomeration effects (Meijers, 2005, 2016; Wiechmann et al., 2016).

Prime examples of inter-urban PRs in Europe are the Rhine-Ruhr region in Germany, the Randstad in the Netherlands and the Upper Silesian industrial region in Poland. The historical origins and trajectories of inter-urban PRs vary. In most cases, they have benefited from a specific location advantage in a particular economic era (e.g. the Ruhr from its coal deposits during industrialisation). The original structure of centres continues to leave its stamp on today’s morphology of these regions, but is supplemented by more recent (intra-urban) sub-centre formations (Kloosterman and Musterd, 2001; Münter and Volgmann, 2014).

Besides these prime examples, inter-urban PRs also include urban regions with several large centres but with one core dominating in many functions – in particular, global business functions and global visibility (e.g. Frankfurt within the Rhine-Main region or Milan within Northern Italy) (Garavaglia, 2014; Münter, 2011) – as well as smaller metropolitan regions with multiple cores, such as the Rhine Neckar region in Germany (with Mannheim and Heidelberg as the major centres) or bipolar regions like Manchester–Liverpool in the UK.

Formation mode 2: Intra-urban polycentric regions as an expression of regional urbanisation and post-suburbanisation

Urbanisation processes are increasingly a regional phenomenon where spatial development processes are played out at the scale of urban regions rather than individual cities (Phelps et al., 2010; Soja, 2015). The hinterlands of core cities are now being used not only for housing but also for business activities, while a (formerly) suburban hinterland’s dependence on the core city is increasingly being replaced by a division of labour. As a result, the sharp phenomenological division between ‘city’ and ‘countryside’ is becoming increasingly blurred (Phelps, 2019), replaced by a patchwork of traditional and new centralities – in the European context also referred to as ‘Zwischenstadt’ (the in-between city) (Sieverts, 1997) – and increasingly polycentric spatial structures with complex patterns of interconnections (Anas et al., 1998). The intra-urban formation of centres outside the traditional central business districts (CBDs) of the core cities (both in the hinterland and within the administrative boundaries of the core cities) – conceptualised for example as sub-centres, new economic clusters or edge cities – is an especially important expression of this (for an overview from a European perspective, see Bontje and Burdack, 2011). These developments have frequently been framed under the umbrella term ‘post-suburbanisation’ (Hesse and Siedentop, 2018; Phelps et al., 2006). Such spatial reconfigurations in a narrower city-regional context are to be observed worldwide in all major urban regions participating in global economic processes. They result in polycentric spatial structures, both in historically PURs like the Rhine-Ruhr and the Randstad, or in (what were in the past) typically monocentric metropolitan regions like London, Paris and Tokyo. From this perspective, the transition to a polycentric system of settlement can be regarded as a universal phenomenon in post-industrial countries (Hall, 1997; Krehl and Siedentop, 2018).

Discussion

The well-established differentiation between inter-urban and intra-urban polycentricity to describe the different evolutionary trajectories of PRs is grounded in different strands of literature and also shows a geographical focus. Van Meeteren et al. (2016) show that the debates on inter-urban PRs are mainly limited to the (by global standards, small-scale) European context, whereas the analysis of intra-urban polycentric spatial development often looks at large-scale US regions. Hence, they suggest a differentiation of the concept of polycentricity based not on spatial scales but instead on the geographical context (Van Meeteren et al., 2016). We take up this suggestion in this article, though restricting ourselves to the (North-Western) European context. Within this geographical context, the differentiation between intra- and inter-urban PRs remains justified, as it distinguishes between both different historical trajectories in their formation and spatial scales. In this European context, inter-urban PRs regularly cover a far larger territory and have a longer history of polycentricity than intra-urban PRs. This contrasts with mega-regions in other geographical contexts (e.g. in the USA or in China, with their much larger scales than both inter- and intra-urban PRs in Europe) (Harrison and Hoyler, 2015; Liu et al., 2017). Moreover, intra-urban PRs or post-suburbanisation processes differ in the European context: European sub-centres, for example, are not comparable with the edge cities in North America, even though they are also a formative component of post-suburbanisation there. European sub-centres are generally much smaller and closer to the CBD of the core city than in North America, tending to complement rather than replace the functions of the CBD (Bontje and Burdack, 2011; Hesse and Siedentop, 2018; Krehl and Siedentop, 2018).

Building on this fundamental conceptual differentiation between intra- and inter-urban PRs, further typologies for conceptualising the spatial scales, spatial forms and historical conditions behind the emergence of polycentricity have been developed (e.g. Davoudi, 2003; Rauhut, 2017). These sometimes distinguish further spatial scales (in particular, supra-regional polycentric urban systems, which Rauhut, 2017, for instance, views as an autonomous concept separate from regional polycentricity). But with respect to the regional scale, such additional typologies do not differ fundamentally from the original typology. In addition, the dichotomous typology is primarily a theoretical-conceptual framework; from an empirical perspective, a continuum would seem to exist between regions with dominant cores and ideal-typical multi-core regions (Hall and Pain, 2006; Wiechmann et al., 2016: 453f.).

Using the typology introduced here, we propose going one step further, sharpening the concept by moving away from the scale-related terms ‘inter-urban polycentric’ and ‘intra-urban polycentric’, terms that imply labelling all urban regions in post-industrial countries as ‘polycentric’ per se. Instead, we propose the following labels addressing the contents of the two sub-concepts and the regions’ historical trajectories:

Polycentric urban regions (or PURs) as the long-established term for inter-urban PRs historically displaying multiple cores. This sub-concept asks whether advantages or disadvantages arise from existing historically polycentric structures (economic, political, etc.) (Meijers et al., 2018; Vasanen, 2013).

Post-suburban (polycentric) regions (PSRs) instead of intra-urban PRs formed through regional urbanisation and the emergence of sub-centres in urban regions in post-industrial countries. The term ‘polycentric’ is explicitly bracketed, as these new (sub-)centres have to integrate into the formerly core-dominated region and find their functional role (in terms of division of labour, specialisation, networking, etc). Though this sub-concept would also work without even using the term ‘polycentric’, the related strand of literature frequently refers to ‘polycentricity’ (Anas et al., 1998; Krehl, 2016). We therefore propose including the word.

Key perspectives on polycentricity on a regional scale

With regard to the key substantive-conceptual perspectives on polycentricity at a regional scale, three or four essential facets of polycentricity in PRs coalesce in the literature (Figure 1). Moreover, some first summary presentations of the substantive dimensions of the concept can be found, for example in Lambregts (2009: 10ff.), Burger and Meijers (2012) and Münter et al. (2016). The first two dimensions concern the analytical description of polycentricity as a spatial outcome. There is consensus in the scholarly debate that this static description of spatial structures can essentially be broken down into (a) a morphological perspective, conceptualising PRs as spaces with several centres of similar importance, and (b) a functional-relational perspective focusing on the relations, interactions and synergies between different-sized centres within a region. Münter et al. (2016) have added a third symbolic perspective addressing the internal and external perception of PRs. By contrast, the fourth perspective (d) of polycentricity is political-normative in nature, referring to the securing or creation of polycentric spatial structures as a planning strategy, as well as to the institutional arrangements within PRs.

Figure 1.

Polycentric regions – a multi-dimensional typology.

In addition, each of the four key perspectives can be viewed both statically (spatial outcome) and dynamically (spatial development process), that is, one can focus either on the observable (static) structures or on the (dynamic) reconfiguration of hierarchies within regions (see e.g. Champion, 2001; Giffinger and Suitner, 2014). Reconfigurations take place for very different reasons, ranging from demographic factors (population growth or suburbanisation) or a redistribution of economic forces within the region, via infrastructural aspects (e.g. changes in accessibility), to technological factors (e.g. a reduction in relational distances thanks to ICT) (Hall and Pain, 2006; Lambregts, 2009).

Combining the four material perspectives on polycentricity, the two modes of formation developed in the previous section and the static versus dynamic focus results in a 4 × 2 × 2 matrix of possible dimensions for conceptualising PRs (see Figure 1). The aim of this conceptual typology is to help researchers and planners position themselves with respect to which ‘building block(s)’ of the concept they take up in both spatial research and planning practice.

We now attempt to explain these four key dimensions of polycentricity both from a static and a dynamic perspective based on a critical review of the comprehensive body of literature on this topic. In addition, we will illustrate and substantiate our theoretical-conceptual remarks using empirical examples drawn from the Rhine-Ruhr region as an example of a PUR, and from the Munich region as an example for a PSR.

Key dimensions (a): Spatial outcome – the morphological perspective

The main research questions that are evident on PRs from a morphological perspective can be summarised as follows: Which centres characterise a region’s polycentric spatial structure? How have the functions of existing centres changed over time, and have new centralities for certain functions emerged?

Most definitions and conceptualisations of polycentricity focus on a region’s morphological spatial structure, conceptualising PRs as areas with multiple centres and thus best corresponding to the minimal definition mentioned above. Burger and Meijers (2012) further differentiate between ‘multicentric/multinuclear’ (for a spatial constellation with multiple centres, thus applying to both PURs and PSRs) and ‘polycentric/polynuclear’ (for spatial units with multiple centres of similar size or importance or that have no or only a slightly developed hierarchy among the centres, and thus only applying to PURs). Regarding the intra-urban formation mode, the morphological perspective on polycentricity is closely related to scientific debates on post-suburbanisation (see above), ‘urban spatial structure’ (static perspective) and processes of economic and demographic concentration and deconcentration in urban regions (dynamic perspective) (Anas et al., 1998; Krehl, 2015; Veneri, 2018). Describing a region’s visible spatial structures, the morphological perspective also partly serves as the basis for the three other perspectives (see below).

Moreover, the question of what characterises morphological polycentricity is not just dependent on these theoretical-conceptual definitions, but also on methodological ones: Which indicators do I use to delimit a PR? Which data sources in which ‘granularity’, and which analytical methods, do I use – and on what assumptions (e.g. threshold values) do I base their use? These decisions similarly have considerable influence on whether and to what extent a spatial unit is to be characterised as polycentric (Krehl, 2016; Möck and Küpper, 2019). Regarding morphological polycentricity, simple yet recognised methods exist: density thresholds can be used to identify the existence of polycentric spatial structures, while rank-size distribution can be used to measure the degree of polycentricity and show the hierarchy or – in the case of polycentric spatial structures – lack thereof in a region’s spatial subunits (e.g. in terms of population or employees) (Burger and Meijers, 2012; Möck and Küpper, 2019; Parr, 2004). However, these simple measurements are based on the assumption that the spatial units to be studied – frequently administrative units – also reflect the structure of a region’s centres. While generally unproblematic for identifying and comparing large-scale inter-urban polycentric spatial structures, they are not appropriate for identifying small-scale intra-urban polycentric structures that have become detached from administrative boundaries. To identify the latter, a wide variety of methods have been developed, permitting a consistent comparative identification of small-scale (sub-)centre systems (e.g. kernel density estimations, locally weighted regression or LISA clusters) (for an overview, see Krehl, 2016).

With regard to the morphological dimension of polycentricity, PURs are primarily considered from a static perspective: What characterises a region’s historically grown centre structure? Restructuring processes, on the other hand, take place within this grown hierarchy, challenging less the centre structure per se than the fact that different hierarchies of centres emerge in relation to different functions (e.g. the distribution of people, jobs, knowledge-intensive jobs or metropolitan functions) (Münter and Volgmann, 2014). In contrast to this, the static perspective (urban spatial structure) and the dynamic perspective (emergence of new sub-centres) are of equal importance among PSRs (see Figure 1).

Figure 2 shows a static morphological perspective of the Rhine-Ruhr PUR and the Munich PSR using three indicators: a centre’s (= municipality) share of population, share of employees and share of metropolitan functions.¹ Comparing the two regions, the maps illustrate that, within the Munich PSR, all three functions are heavily concentrated on Munich, the centre accounting for 43% of the region’s population (while none of the other municipalities has a share exceeding 1.5%). By contrast, in the Rhine-Ruhr PUR, the largest centre, Cologne, accounts for a population share of just 8.6%, while the largest 10 cities in sum only account for a population share of 39%.

Figure 2.

The Rhine-Ruhr PUR and the Munich PSR – morphological and functional perspectives.

Despite the clear dominance of Munich within its region, we can identify certain municipalities in the hinterland with a clearly disproportionate share of employees compared to their population. These include in particular Unterhachingen, Freising and Garching, municipalities conveniently located in the immediate vicinity of Munich or Munich Airport and characterisable as employment sub-centres.² As the home of Munich Airport, Freising also has an appreciable surplus of metropolitan functions (which are usually highly centre-related; see below).

As shown by the example of the Rhine-Ruhr PUR, conclusions about the degree of a region’s polycentricity also depend to a considerable degree on the indicator considered. This becomes evident when, in addition to population and employee shares which tend to be evenly distributed throughout the region, metropolitan functions are considered. As these represent the metropolitan attributes of cities, they have a strong tendency to be spatially concentrated (Growe and Volgmann, 2016). Figure 2 makes clear that the Rhine-Ruhr PUR is far more polycentric in relation to population and employment than in relation to metropolitan functions: whereas only 39% of the population live and 48% of employed persons work in the 10 biggest cities in the region, 99% of the metropolitan functions are concentrated there.

Key dimensions (b): Spatial outcome – the functional-relational perspective

Looking at PRs from a functional-relational perspective, the main research questions can be summarised as follows: Which relations and interconnections exist between a region’s centres in various domains, e.g. business, housing and metropolitan functions? Can a functional specialisation or division of labour be observed within the region, and what synergies exist between a region’s centres? And how have the functional interconnections, intra-regional specialisation and division of labour changed over time?

The functional perspective on polycentricity puts the focus on both the interactions and the level of integration between a region’s centres. While morphological characteristics – the presence of centres – are not bracketed in this approach, the functional perspective extends the morphological approach by focusing on both the directions of flows between the centres and the strength of their interrelations. Furthermore, this perspective is less about the general presence of relations between the centres; rather, a region can be characterised as functionally polycentric if flows between centres are balanced. This means that they have no dominant direction – that is, there is a two-way exchange between centres, with flows criss-crossing a region, including between its smaller centres (Cervero and Wu, 1998; De Goei et al., 2010; Green, 2007). With respect to the functional interconnections within PRs, the question also arises as to whether such interconnections are weak or strong and lead to a functional integration and division of labour: that is, whether a region’s different centres fulfil different roles and allow mutual synergies between the centres to emerge (Cowell, 2010; Meijers, 2005). (Related) specialisation and complementarities between (sub-)centres are an important proxy for the division of labour between centres in terms of economic activities, while the level of an urban region’s functional integration is seen as a key driver of regionalised agglomeration economies (Camagni and Salone, 1993; Hanssens et al., 2014).

With a few exceptions, most empirical studies analysing a region’s functional spatial organisation are based on commuting data. In part because it is also often the only data available, such data reflecting the daily lives of a region’s inhabitants is used as a proxy indicator for other types of spatial interaction (Burger et al., 2013). Other data on flows of persons and goods also serve as indicators: for instance, shopping itineraries, intra-company communication and interaction flows and value-creation chains between businesses (e.g. Burger and Meijers, 2012; Hall and Pain, 2006). A division of labour between a region’s centres can, for example, be established, inasmuch as specialisations of centres are determined via location quotients, or complementarities between centres are identified with the help of correspondence analysis (Meijers, 2005; Volgmann and Münter, 2018).

As commuting patterns are the most established indicator for measuring functional polycentricity on a regional scale, we use them at this juncture to visualise differences between the commuting patterns of the Rhine-Ruhr PUR and those of the Munich PSR (see Figure 2). In the former, commuter flows represent a tightly interwoven criss-cross pattern of interlinkages throughout the region, with the strongest interlinkages observed, on the one hand, between the major centres and, on the other, between the core cities and their surrounding municipalities. By contrast, commuter flows in the Munich PSR are dominated by radial flows, with Munich at the centre. Only a few smaller centres show their own, spatially limited commuter catchment area.

Key dimension (c): Spatial outcome – the symbolic perspective

Turning to the symbolic perspective, the main research questions are as follows: Is the PR being marketed externally and, if yes, in what form? Which regional symbols does the PR have at its disposal? In addition to local identifications, does the PR also have regional identifications? Finally, how have the internal and external perceptions of the region changed over time?

The symbolic perspective addresses a region’s external visibility, as well as its internal perception in the sense of a regional identity. This perspective refers to the symbolic function of metropolitan regions, one of four functions determining the degree of ‘metropolitanity’ (see above). As externally directed interpretative models, a region’s national and international visibility or its image is produced, in particular, by cognitive, historical and architectural symbols (e.g. flagship projects), as well as targeted marketing activities. By way of these symbols and marketing activities, all urban or metropolitan regions do their best to stand out from other regions (Blotevogel and Danielzyk, 2009; Growe and Volgmann, 2016). Regarding a region’s symbolic value, the question arises as to whether these externally directed interpretive models only or primarily concern a region’s core(s) or whether they have an explicitly regional character. In the latter case, a region would, for example, have regional symbols at its disposal, marketing not only the image of a city but the potential of the region as a whole.

Internally, the symbolic reading of polycentricity concerns regional consciousness or identity: that is, the perception of regional players; Meijers et al. (2018) also speak of cultural integration. Regional identity takes into account the extent to which the PR is the guiding frame of reference for the action of residents, businesses and regional stakeholders. Regional identity is sub-divided into a regional ‘belongingness’ in the sense of a population’s region-specific identification with a certain territory, and a regional ‘developmental consciousness or identity’ in the sense of a regional economic and work culture or mentality. Existing functional interconnections within the region (see Key dimension (b)) have a positive influence on regional identity (Kübler, 2017). The same applies to regional institutions (see Key dimension (d)), although, in practice, a simultaneity of or a complex relationship between cooperation and competition in PRs is often to be found here. Whereas such regions, in order to gain supra-regional visibility, often present themselves as unified wholes externally, at the same time they internally attempt to stand out from their neighbouring centres (Goess et al., 2016).

Comparing the Rhine-Ruhr PUR and the Munich PSR, we again arrive at contrasting findings: the Rhine-Ruhr PR is a living space and space of action for a wide variety of players. Nonetheless, up to now, no regional belongingness has emerged within the region, and neither is it jointly marketed to the outside world. For several reasons (e.g. the sheer size of the region or the uneven economic development of its sub-areas), there is no perception or self-understanding of the Rhine-Ruhr region as a unified whole. The regional belongingness of local stakeholders thus relates to the region’s more manageable, homogeneous and institutionalised sub-areas: in particular, the Ruhr, the Rhineland and the Bergisches Land (Goess et al., 2016: 2046). By contrast, the Munich PSR’s external visibility and internal regional consciousness are strongly associated with the city of Munich with its high degree of global recognition. Munich and its name also serve as the region’s standard-bearer. For example, ‘Munich’ is the basic geographical point of reference for the ‘Munich Metropolitan Region’, a voluntary cooperation between the city of Munich and its surrounding area (EMM e.V., 2019). The Munich ‘Oktoberfest’ serves as an external symbol for the whole region, while in the Rhine-Ruhr region external symbols such as Cologne Cathedral have no symbolic value for the whole Rhine-Ruhr region, but only for the city of Cologne and its direct surroundings. The same can be observed with regard to the identity-forming role of football clubs for a city and region. While in the case of the Rhine-Ruhr, Borussia Dortmund (BVB) assumes an identity-forming role for the city of Dortmund and its direct hinterland, the club plays no such role for the Rhine-Ruhr region as a whole. Quite the opposite, the Rhine-Ruhr region is instead characterised by local rivalries (especially with the Schalke 04 football club from Gelsenkirchen, just 35 km away from Dortmund). By contrast, the sense of belonging to FC Bayern Munich radiates out to the whole of Bavaria (Spiegel, 2018).

Key dimension (d): The political-normative perspective – planning concepts and governance arrangements

Last but not least, the main research questions regarding the political-normative dimension are as follows: Which goals and strategies are followed in the region with regard to the maintenance and creation of polycentric spatial structures? Which informal and formal institutional arrangements does the region have at its disposal? In what way have planning concepts and institutional arrangements been modified over time?

The political-normative perspective comprises, on the one hand, conceptual strategic aspects (polycentricity as a planning concept) and, on the other, the institutional arrangements within PRs. Prominent examples of planning concepts aiming at a polycentric spatial development are situated at the European scale (e.g. the ESDP; see above). On the regional scale, such planning strategies aim either at maintaining existing polycentric structures or at creating polycentric structures, for example with a view to avoiding urban sprawl (METREX, 2010). Normative planning strategies promoting polycentric spatial development often build on the given spatial-structural conditions (in particular in PURs), but are also employed in a visionary way detached from real-life conditions (mainly in PSRs).

With respect to the institutional arrangements in a PR, the focus is placed on questions of governance structures in a multi-scalar system; of the delimitation of the region or the acceptance of fluid forms of (non-)participation of stakeholders within ‘soft spaces’ (Othengrafen et al., 2015); of legal dispositions; and of modes of decision-making, including their democratic legitimation. The fact that PRs, in particular PURs, regularly lack strong regional stakeholders able to assume a leading role, combined with the particularly complex spatial-structural conditions and functional interconnections, place high demands on the design of institutional arrangements (Brenner, 2004; Keil et al., 2017; Zimmermann, 2017).

Comparing the Rhine-Ruhr PUR and the Munich PSR, we should start by highlighting that both regions are European Metropolitan Regions, recognised as such by the German Ministerial Conference on Spatial Planning since the 1990s. However, the Rhine-Ruhr PUR has never been able to convert this label into a visible organisational form or practice of regional cooperation, for instance due to the lack of a regional belongingness (e.g. the large number of centres of similar size and the associated complex stakeholder structures or the unequal development of the Rhineland and Ruhr sub-areas; see above) (Schmitt, 2007). Indeed, in the meanwhile two metropolitan cooperation areas have emerged, the ‘Ruhr Metropolis’ and the ‘Rhineland Metropolitan Region’, both of which see their function as improving their respective competitiveness and related issues. Thus, two sub-regional organisational forms, not originally intended, have proven themselves to be functional within the Rhine-Ruhr PUR. By contrast, in the Munich PSR, the Munich Metropolitan Region under the leadership of Munich’s Bürgermeister has for years been the undisputed central organisation for a regional development strategy promoting international competitiveness and internal networking (EMM e.V., 2019).

Discussion

In this article, we subdivided the concept of ‘polycentric regions’ into the two main historical trajectories of PR formation and four key substantive perspectives on polycentricity, each analysable from both a static and a dynamic point of view. Summarising our findings, Figure 1 provides an overview of our conceptual typology of the multi-dimensional features of a PR.

In line with our proposal to replace the scale-related terms ‘inter- and intra-urban polycentric’ for PRs with ‘PURs’ and ‘PSRs’, terms more precisely addressing the contents of the sub-concepts, we also propose content-related labels for the four key perspectives:

‘Polycentric urban spatial structure’ for the morphological perspective

‘Polycentric regional networks’ for the functional-relational perspective

‘Polycentric identities and symbols’ for the symbolic perspective

‘Polycentric planning concepts and government arrangements’ for the normative perspective.

As the concept of a PR is so broad and diverse that it can hardly be dealt with comprehensively in empirical studies (i.e. addressing all dimensions), it is only rarely advisable to make reference to the PR concept as such. The typology presented in this article allows a positioning with respect to which ‘building block(s)’ of the concept are being taken up (e.g. ‘regional networks in PURs’ or ‘polycentric planning concepts for PSRs’).

Both PURs and PSRs can be analysed from both a static and a dynamic perspective. But, as Figure 1 indicates, the initial conditions differ:

In the case of PURs, the initial condition is the static perspective: the focus is on historically polycentric spatial structures where any development takes place within a spatially persistent setting and tends to have a step-by-step effect, at least with regard to material aspects. Restructuring processes in these regions are mainly functional (e.g. changing commuter flows or hierarchies of centres)

In the case of PSRs, however, the starting condition is the process: the focus is on the decentralisation of functions and the formation of sub-centres within a region, promoting the emergence of new post-suburban spatial realities, even if still comparatively unstable.

Building on the typology presented here makes it easier to look at the causes and effects of polycentricity in a more targeted fashion and thus also at normative questions in the context of developing planning strategies. But without clear substantive positioning, it is hardly possible for policymakers and planners to draw any normative conclusions from the concept of PRs (e.g. ‘more centres in a region create more opportunities’). At the same time, however, from an institutional perspective, PRs are more complex: Policymakers have to keep in mind that the four perspectives on polycentricity are closely connected and partially dependent on one another. Thus, for instance, strengthening the economic competitiveness of a PR depends on the extent to which one succeeds in considering the region ‘as a whole’. The more morphologically PRs succeed in growing together functionally, symbolically and institutionally, the better are the preconditions for reciprocal synergies to arise within the region (see also Meijers et al., 2014). Looking at the Rhine-Ruhr PUR, while it is clear that the latter is a morphologically and functionally polycentric settlement agglomeration, it lacks both symbolic-cultural and institutional integration.

Conclusion: Towards a renewed research agenda on polycentric regions

It is generally acknowledged in the literature that PRs consist of several centres, have a minimum size and are situated on a regional scale in the respective geographical context. This article highlights, however, that over and beyond this lowest common denominator there is no generally accepted definition of PRs, but rather various formation modes and content-related perspectives. Polycentricity is both an analytical concept and a normative planning concept, it can be interpreted both morphologically and functionally-relationally, it concerns both the material and the symbolic significance of areas and it can be grasped as both a static spatial structure and a dynamic process of development. In this article, we have summarised these opposites in a 4 × 2 × 2 matrix of dimensions for conceptualising PRs, and have proposed (old and) new labels for different sub-concepts (e.g. ‘urban spatial structure of PURs’). Addressing these sub-concepts in empirical research and planning practice instead of using the umbrella terms ‘polycentricitiy’ and ‘polycentric regions’ would help overcome problems resulting from the present fuzziness of the concept. Without clear substantive positioning, it is hardly possible to draw normative conclusions from the concept, as the often-heard political supposition that ‘polycentricity is something good’ certainly does not apply to all dimensions in all regions.

Depending on the aspects and indicators chosen in analysing the four key perspectives on the PR concept, one arrives at different findings with respect to a region’s degree of polycentricity (e.g. a morphologically polycentric population distribution versus a clearly hierarchical distribution of metropolitan functions) and its evolution (e.g. stable morphological spatial pattern versus increasing functional interlinkages).

Moreover, the question of what constitutes a PR is not solely dependent on these theoretical-conceptual definitions, but also on methodological ones: Which indicators do I use to delimit a PR? Which data sources in which ‘granularity’, and which analytical methods, do I use – and on what assumptions (e.g. threshold values) do I base their use when addressing a specific research question? These decisions have considerable influence on whether and to what extent a spatial unit is to be characterised as polycentric (Krehl, 2016). We have only hinted at such methodological challenges in this article. These have as much to be addressed in empirical studies, against the background of specific research questions, as the conceptual aspects discussed in detail in this article. Additionally, the typology developed in this article aims to offer conceptual support for positioning oneself within the various sub-concepts for the North-Western European context in particular. This typology would have to be expanded or adapted for other geographical settings, as, besides the different formation modes and content-related perspectives, the PR concept is interpreted differently in different spatial contexts (Liu et al., 2017; Van Meeteren et al., 2016).

What can we conclude from our critical literature review for a renewed research agenda on PRs? The comprehensive empirical body of literature on the PR concept shows that extensive empirical findings for different spatial scales and geographical contexts are already available on certain dimensions, especially the morphological. On the other hand, hardly any empirical research has been done on the questions of the internal and external perception of PRs and which symbols serve as representative functions for which cities and regions. Additionally, our view on PRs remains rather static, neglecting longitudinal research on their spatio-temporal dynamics and path reconfigurations. Knowledge gaps are also apparent with regard to cross-national comparative studies on the substantive perspectives on the PR concept, taking regional contexts into account: the emergence of sub-centres is a universal phenomenon in post-industrial countries (see above), but they differ significantly in different spatial contexts (e.g. small-scale CBD-oriented sub-centre formations in many European countries vs. large edge cities in the USA; see Hesse and Siedentop, 2018; Keil and Addie, 2015). For example, very little research has been conducted on the impacts of various planning and regulatory practices in different regional contexts.

This brings us to one of the main knowledge gaps in research on PRs: the interaction and mutual connections between the different dimensions of polycentricity in a specific region. How and why do morphological, functional-relational, symbolic and political processes interact in a particular PR, and what does this imply for both its material appearance and its economic performance compared to other regions? How do, for example, new regional institutions in PRs affect the regional sense of belonging or a region’s day-to-day functional interlinkages? Or is greater institutional integration only feasible where a regional identity already exists? And what are the differences between PURs and PSRs in this vein? Or (how) can regional governance enhance the division of labour between the various centres of a PUR to avoid dispensable of material (e.g. high-class infrastructure) and immaterial (e.g. sector-related promotion strategies) duplication?

Footnotes

Acknowledgements

We would like to thank our colleagues and former colleagues from the ILS – Institute for Regional and Urban Development and from the Competence Field ‘Metropolitan Research’ within the University Alliance Ruhr (in particular, Rainer Danielzyk, Wolfgang Knapp, Angelika Krehl, Stefan Siedentop and Thorsten Wiechmann) for discussing and working on PRs with us for several years. Additionally, we would like to thank the reviewers and editor of Urban Studies for their fruitful and engaged comments which improved the article to its current state.

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

ORCID iD

Angelika Münter

Notes

References

Anas

Arnott

Small

(1998) Urban spatial structure. Journal of Economic Literature 36(3): 1426–1464.

Blotevogel

Danielzyk

(2009) Leistungen und Funktionen von Metropolregionen. In: Knieling

(ed.) Metropolregionen: Innovation, Wettbewerb, Handlungsfähigkeit. Hannover: Akademie für Raumforschung und Landesplanung (ARL), pp. 22–29.

Bontje

Burdack

(2011) Post-suburbia in Continental Europe. In: Phelps

(eds) International Perspectives on Suburbanization. London: Palgrave Macmillan, pp. 143–162.

Brenner

(2004) New State Spaces: Urban Governance and the Rescaling of Statehood. Oxford and New York: Oxford University Press.

Burger

Meijers

(2012) Form follows function? Linking morphological and functional polycentricity. Urban Studies 49(5): 1127–1149.

Burger

Van der Knaap

Wall

(2013) Polycentricity and the multiplexity of urban networks. European Planning Studies 22(4): 816–840.

Camagni

Salone

(1993) Network urban structures in Northern Italy: Elements for a theoretical framework. Urban Studies 30(6): 1053–1064.

Cervero

K-L

(1998) Sub-centring and commuting: Evidence from the San Francisco Bay Area, 1980–90. Urban Studies 35(7): 1059–1076.

Champion

(2001) A changing demographic regime and evolving polycentric urban regions: Consequences for the size, composition and distribution of city populations. Urban Studies 38(4): 657–677.

10.

Cowell

(2010) Polycentric regions: Comparing complementary and institutional governance in the San Francisco Bay Area, the Randstad and the Emilia-Romagna. Urban Studies 47(5): 945–965.

11.

Davoudi

(2003) Polycentricity in European spatial planning: From an analytical tool to a normative agenda. European Planning Studies 11(8): 979–999.

12.

De Goei

Burger

Van Oort

, et al. (2010) Functional polycentrism and urban network development in the Greater South East, United Kingdom: Evidence from commuting patterns, 1981–2001. Regional Studies 44(9): 1149–1170.

13.

EMM

e.V.

(2019) The Munich metropolitan region. Available at: https://www.metropolregion-muenchen.eu/en/ (accessed 19 December 2019).

14.

Garavaglia

(2014) The distribution of advanced business services in Northern Italy: Towards a polycentric metropolis model? Métropoles 14. Available at: http://journals.openedition.org/metropoles/4889 (accessed 19 December 2019).

15.

Giffinger

Suitner

(2014) Polycentric metropolitan development: From structural assessment to processual dimensions. European Planning Studies 23(6): 1169–1186.

16.

Goess

De Jong

Meijers

(2016) City branding in polycentric urban regions: Identification, profiling and transformation in the Randstad and Rhine-Ruhr. European Planning Studies 24(11): 2036–2056.

17.

Green

(2007) Functional polycentricity: A formal definition in terms of social network analysis. Urban Studies 44(11): 2077–2103.

18.

Growe

Volgmann

(2016) Exploring cosmopolitanity and connectivity in the polycentric German urban system. Tijdschrift voor economische en sociale geografie 107(2): 214–231.

19.

Hall

(1997) Modelling the post-industrial city. Futures 29(4/5): 311–322.

20.

Hall

Pain

(eds) (2006) The Polycentric Metropolis: Learning from Mega-City Regions in Europe. London: Earthscan.

21.

Hanssens

Derudder

Van Aelst

, et al. (2014) Assessing the functional polycentricity of the mega-city-region of Central Belgium based on advanced producer service transaction links. Regional Studies 48(12): 1939–1953.

22.

Harrison

Hoyler

(eds) (2015) Megaregions: Globalization’s New Urban Form?Cheltenham: Edward Elgar.

23.

Hesse

Siedentop

(2018) Suburbanisation and suburbanisms: Making sense of continental European developments. Raumforschung und Raumordnung/Spatial Planning and Research 76(2): 97–108.

24.

Hoyler

Kloosterman

Sokol

(2008) Polycentric puzzles: Emerging mega-city regions seen through the lens of advanced producer services. Regional Studies 42(8): 1055–1064.

25.

Keil

Addie

J-PD

(2015) ‘It’s not going to be suburban, it’s going to be all urban’: Assembling post-suburbia in the Toronto and Chicago regions. International Journal of Urban and Regional Research 39(5): 892–911.

26.

Keil

Hamel

Boudreau

J-A

, et al. (eds) (2017) Governing Cities Through Regions: Canadian and European Perspectives. Waterloo: Wilfrid Laurier University Press.

27.

Kloosterman

Musterd

(2001) The polycentric urban region: Towards a research agenda. Urban Studies 38(4): 623–633.

28.

Krehl

(2015) Urban spatial structure: An interaction between employment and built-up volumes. Regional Studies, Regional Science 2(1): 289–307.

29.

Krehl

(2016) Analyzing polycentricity. PhD Thesis, TU Dortmund University, Germany.

30.

Krehl

Siedentop

(2018) Towards a typology of urban centers and subcenters: Evidence from German city regions. Urban Geography 36(3): 1–25.

31.

Kübler

(2017) Citizenship in the fragmented metropolis: An individual-level analysis from Switzerland. Journal of Urban Affairs 40(1): 63–81.

32.

Lambregts

(2009) The polycentric metropolis unpacked: Concepts, trends and policy in the Randstad Holland. PhD Thesis, University of Amsterdam, the Netherlands.

33.

Liu

Derudder

Wang

(2017) Polycentric urban development in China: A multi-scale analysis. Environment and Planning B: Urban Analytics and City Science 45(5): 953–972.

34.

Meijers

(2005) Polycentric urban regions and the quest for synergy: Is a network of cities more than the sum of the parts? Urban Studies 42(4): 765–781.

35.

Meijers

(2016) Polycentric metropolitan regions in Europe: State of the art. In: Danielzyk

Münter

Wiechmann

(eds) Polyzentrale Metropolregionen. Lemgo: Rohn, pp. 122–138.

36.

Meijers

Hoogerbrugge

Cardoso

(2018) Beyond polycentricity: Does stronger integration between cities in polycentric urban regions improve performance? Tijdschrift voor economische en sociale geografie 109(1): 1–21.

37.

Meijers

Hoogerbrugge

Hollander

(2014) Twin cities in the process of metropolisation. Urban Research & Practice 7(1): 35–55.

38.

METREX (ed.) (2010) Intra-Metropolitan Polycentricity in Practice: Reflections, Challenges and Conclusions from 12 European Metropolitan Areas. Glasgow: METREX.

39.

Möck

Küpper

(2019) Polycentricity at its boundaries: Consistent or ambiguous? European Planning Studies 59(12): 1–20.

40.

Münter

(2011) Germany’s polycentric metropolitan regions in the world city network. Raumforschung und Raumordnung/Spatial Planning and Research 69(3): 187–200.

41.

Münter

Volgmann

(2014) The metropolization and regionalization of the knowledge economy in the multi-core Rhine-Ruhr metropolitan region. European Planning Studies 22(12): 2542–2560.

42.

Münter

Wiechmann

Danielzyk

(2016) Polyzentrale Metropolregionen – Zur Einführung. In: Danielzyk

Münter

Wiechmann

(eds) Polyzentrale Metropolregionen. Lemgo: Rohn, pp. 10–44.

43.

Othengrafen

Haughton

Knieling

, et al. (2015) Soft Spaces in Europe: Re-negotiating Governance, Boundaries and Borders. London: Routledge.

44.

Parr

(2004) The polycentric urban region: A closer inspection. Regional Studies 38(3): 231–240.

45.

Parr

(2014) The regional economy, spatial structure and regional urban systems. Regional Studies 48(12): 1926–1938.

46.

Phelps

(2019) In what sense a post-suburban era? In: Hanlon

Vicino

(eds) The Routledge Companion to the Suburbs. London and New York: Routledge, Taylor & Francis Group, pp. 39–47.

47.

Phelps

Parsons

Ballas

, et al. (2006) Post-suburban Europe: Planning and Politics at the Margins of Europe’s Capital Cities. Basingstoke: Palgrave Macmillan.

48.

Phelps

Wood

Valler

(2010) A postsuburban world? An outline of a research agenda. Environment and Planning A: Economy and Space 42(2): 366–383.

49.

Rauhut

(2017) Polycentricity – one concept or many? European Planning Studies 25(2): 332–348.

50.

Schmitt

(2007) Raumpolitische Diskurse um Metropolregionen: Eine Spurensuche im Verdichtungsraum Rhein-Ruhr. Dortmund: Rohn.

51.

Scott

(2001) Globalization and the rise of city-regions. European Planning Studies 9(7): 813–826.

52.

Sieverts

(1997) Zwischenstadt: Zwischen Ort und Welt, Raum und Zeit, Stadt und Land. Braunschweig and Wiesbaden: Vieweg.

53.

Soja

(2015) Accentuate the regional. International Journal of Urban and Regional Research 39(2): 372–381.

54.

Spencer

(2015) Knowledge neighbourhoods: Urban form and evolutionary economic geography. Regional Studies 49(5): 883–898.

55.

Spiegel (2018) Freunde und Rivalen: Der Fan-Atlas von SPIEGEL Online. Available at: http://www.spiegel.de/sport/fussball/fussball-wer-sind-die-freunde-und-rivalen-ihres-lieblingsvereins-a-1219354.html (accessed 16 November 2018).

56.

Van Meeteren

Poorthuis

Derudder

, et al. (2016) Pacifying Babel’s Tower: A scientometric analysis of polycentricity in urban research. Urban Studies 53(6): 1278–1298.

57.

Vasanen

(2013) Spatial integration and functional balance in polycentric urban systems: A multi-scalar approach. Tijdschrift voor economische en sociale geografie 104(4): 410–425.

58.

Veneri

(2018) Urban spatial structure in OECD cities: Is urban population decentralising or clustering? Papers in Regional Science 97(4): 1355–1374.

59.

Volgmann

Münter

(2018) Specialization of and complementarities between (new) knowledge clusters in the Frankfurt/Rhine-Main urban region. Regional Studies, Regional Science 5(1): 125–148.

60.

WGBU (2016) Humanity on the move: Unlocking the transformative power of cities. Flagship report; German Advisory Council on Global Change. Berlin: WGBU.

61.

Wiechmann

Münter

Danielzyk

(2016) Stand und Perspektiven der Erforschung polyzentraler Metropolregionen. In: Danielzyk

Münter

Wiechmann

(eds) Polyzentrale Metropolregionen. Lemgo: Rohn, pp. 447–483.

62.

Zimmermann

(2017) Re-scaling of metropolitan governance in Germany. Raumforschung und Raumordnung/Spatial Research and Planning 75(3): 253–263.