BORROWED SIZE IN THE METROPOLITAN REGION OF AMSTERDAM
A mixed methods research about the influence of the city of Amsterdam on its surroundings
Eva van der Meulen Student number: 10757279
E-mailadres: evavandmeulen@gmail.com Supervisor: Niels Beerenpoot Second reader: Evert Meijers
Preface
Zaandam, een unieke stad onder de rook van Amsterdam, is de stad waar ik geboren en getogen ben. Al sinds jongs af aan was het Zaanse leven voor mij echter verworven met Amsterdam. Veel van mijn vrije tijd bracht ik door in Amsterdam. Toen ik ging studeren ben ik naar Amsterdam verhuisd en is mijn leven naar de grote stad verplaatst. Veel van mijn vrienden namen dezelfde stap. Maar in de straat van mijn ouders kwamen juist steeds meer mensen uit Amsterdam wonen. Deze dynamiek riep bij mij vragen op. Hoe verhouden deze steden met elkaar en wat zorgt ervoor dat de een wegtrekt naar de grote stad, en de ander getrokken wordt door de regionale stad. Profiteren de steden van elkaars nabijheid? Of ervaren ze er juist nadelen van? Ruimtelijke regionale dynamieken zoals deze hebben altijd al mijn aandacht getrokken vanwege de economische, sociale en fysieke aspecten die bij een vraagstuk als deze spelen. Ik heb dan ook met plezier mijn masterthesis geschreven over Borrowed size in de metropool regio Amsterdam. Tijdens de eerste fase van mijn onderzoek ben ik een artikel tegengekomen van Evert Meijers over Borrowed size. Gedurende mijn thesis heb ik samen met Evert gesprekken gehad over Borrowed size in de metropool regio Amsterdam. Ik wil Evert graag bedanken voor de leerzame, maar vooral ook leuke gesprekken die mij in het proces veel hebben gebracht. Daarnaast wil ik Niels Beerenpoot graag bedanken voor de begeleiding gedurende het gehele proces. Tot slot zou ik graag de respondenten van het interview willen bedanken voor hun inzet en interessante standpunten.
Dutch Abstract
In deze scriptie is onderzocht hoe de nabijheid van Amsterdam de gemeenten van de Metropool Regio Amsterdam beïnvloeden. Grote steden ervaren agglomeratie voordelen door de kritieke massa van de stad (Glaeser, 2011). Amsterdam heeft door zijn unieke geschiedenis een belangrijke positie in de wereld. Met 868.000 inwoners is Amsterdam een kleine stad in vergelijking met de megasteden van deze wereld. Amsterdam maakt onderdeel uit van een polycentrische stad en profiteert van de nabijheid van steden en dorpen die de kritieke massa van de stad vergroten. Maar hoe profiteert de regio nou van de nabijheid van Amsterdam en de groei die de stad ervaart? Het concept ‘borrowed size’ gaat in op de vraag in hoeverre steden in de nabijheid van een grote stad profiteren van de massa van de grote stad (Alanso, 1973). Is er sprake van positieve effecten, of schaduw effecten? De regio Amsterdam blijkt niet altijd zoveel te profiteren als ze zouden willen (Parool, 2019). Netwerk integratie is een belangrijke verklarende factor voor de mate waarin een stad voordelen of nadelen ervaart van de nabijheid van een grote stad (Meijers et al., 2018). De afgelopen jaren is de integratie tussen de gemeenten van de regio toegenomen. In 2016 is de MRA opgericht, een informeel platform waar de regio samenwerkt op verschillende thema’s. Maar de integratie is ook toegenomen omdat verschillende actoren de regio in trekken vanwege de gestage groei van Amsterdam. Ook padafhankelijkheid, de grootte van een stad en de locatie van een stad spelen een rol. Dit heeft geleid tot de volgende hoofdvraag:
“In welke mate ervaren steden en plaatsen in de metropool regio van Amsterdam ‘Borrowed size’ en waar wordt dit door beïnvloed?”
De prestaties van gemeenten is in 2013 en 2018 onderzocht aan de hand van statistische analyses op basis van negen indicatoren. Het concept van borrowed size is geoperationaliseerd naar de drie dimensies van ‘performance’, ‘functionality’ en ‘demographics’. Door de verwachte waarde te vergelijken met de werkelijke waarde is de disconnectie tussen populatiegrootte en de werkelijke ‘size’ van een stad in beeld gebracht. Het kwalitatieve onderzoek heeft de verklarende factoren in beeld gebracht door middel van zeven diepte-interviews met belangrijke betrokkenen in de metropool regio van Amsterdam.
Uit de resultaten is gebleken dat Amsterdam de enige gemeente is die op alle drie de dimensies van ‘borrowed size’ beter presteert dan verwacht. De kleine gemeenten nabij Amsterdam doen het ook goed. Zij hebben vaak een leefomgeving die aansluit bij de grote stad. Maar de middelgrote steden en gemiddeld grote steden van de regio doen het over het algemeen niet goed op het gebied van deze leefomgeving. Ze hebben minder functies dan verwacht en de populatie is minder grootstedelijk dan verwacht. De trek van actoren naar de regio vanuit de grote stad komt nog niet terug in de cijfers. Deze effecten zijn tussen 2013 en 2018 over het algemeen sterker geworden. Sterkere integratie leidt tot betere groei prestaties in de region, maar het resulteert ook in sterkere functionele schaduwen. De toegenomen integratie resulteert in sterkere regionale rollen van gemeenten in de MRA.
De regio trekt nog onvoldoende samen op en zou volgens verschillende experts meer kunnen bereiken wanneer Amsterdam en de regio dit wel zouden doen. Amsterdam “denkt te veel aan zichzelf”, en de regiogemeenten “gaan onderling de competitie aan”. De regio voelt zich niet altijd eigenaar over het MRA-platform doordat Amsterdam met zijn massa en denkkracht het platform vaak domineert. Een formele bestuurslaag dat bestuurd wordt door de gehele regio zou in het voordeel zijn van de prestaties van de regio. Maar de regio zit niet te wachten op deze extra bestuurslaag. De vraag is daarnaast in hoeverre het uitmaakt hoe
een regionale stad het doet als de centrale stad hoge groei ervaart. Wanneer je als stad geïntegreerd bent in het netwerk kun je profiteren van de voordelen van het netwerk. Maar de voordelen van een netwerk zijn niet altijd voor iedereen toegankelijk (XX). Desondanks zouden de gemeente van de regio moeten inspelen op hun functie binnen de regio. De verschillende dimensies van ‘Borrowed size’ hebben laten zien wat de sterkte en zwaktepunten zijn van gemeenten en zouden hier beleidsmatig op in kunnen spelen. Maar de padafhankelijkheid van een gemeente speelt een belangrijke rol. Het ‘soft capital’ van een plaats beïnvloedt de mate waarin een plaats kan profiteren van de nabijheid van een grote stad. Er zal verder onderzoek gedaan moeten worden naar de precieze invloed van padafhankelijkheid en de mate waarin steden kunnen profiteren van de nabijheid van een grote stad nader moeten onderzoeken.
English abstract
The aim of this research was to investigate how the proximity of Amsterdam influences the municipalities in the metropolitan area of Amsterdam (MRA). Large cities often experience agglomeration benefits because of their critical mass (Glaeser,2011). Amsterdam has a unique role in the world because of its unique history. But with 868.000 inhabitants, Amsterdam is a relatively small city in comparison with the mega cities of the world. Amsterdam is a part of a polycentric city and profits from the proximity of cities and towns that enlarge the critical mass of the city. But how does the region profit from the proximity of Amsterdam and the growth that this city experiences? The concept of borrowed size explains how cities and towns near a large primate city profit from the critical mass of the large city (Alanso, 1973). Do cities and towns experience positive effects, or shadow effects? The region stated that it is not always profiting as much as they would like to (Parool, 2019). Network integration is an important explanatory factor for the degree in which a city profits from the proximity of a large primate city (Meijers et al., 2018). The integration of the MRA has intensified the last couple of years. In 2016 the informal governance platform of the MRA was founded where the region cooperates on several subjects. The integration has also intensified because of the crowding out phenomena because of the growth of Amsterdam. Path dependency, the size of a city and the physical location of a city also influence the borrowed size extent. This has resulted in the following research question:
“To what extent are the cities and towns in the metropolitan region of Amsterdam experiencing borrowed size and what influences this?”
The performance of the region has been investigated in 2013 and 2018 by applying a statistical analysis based on nine indicators. The concept of borrowed size has been operationalized into the three dimension of performance, functionality and demographics. By comparing the expected value with the actual size, the disconnection between the population size and the ‘size’ of a city is shown. The qualitative research has investigated the explanatory factors by conducting seven in depth interviews with expert of the metropolitan region of Amsterdam.
The results have showed that Amsterdam is the only city that performs better than expected on all three the dimension of borrowed size. Smaller municipalities also tend to perform well. They often have living environments that are linked to the primate city. However, middle sized and average sized municipalities often perform less than expected in relation to the living environment. The movements of actors towards the region is not visible in the data yet. They have less functions than expected and they have less ‘metropolitan’
inhabitants than expected. These effects often got stronger between 2013 and 2018. The stronger integration has resulted in improved economic performance of the region, but it also results in stronger functional shadows. The increased integration has resulted in stronger regional roles of the municipalities of the MRA.
The region is not collaborating enough and should be able to achieve more of they would increase the collaboration in the future according to the experts. Amsterdam is “thinking about itself too much”, and the regional municipalities “compete with each other”. The region doesn’t feel like they own the MRA platform because Amsterdam often dominates the platform with its mass and thinking power. A formal governance layer that is owned by the entire region could benefit the performance of the region. However, the region is not that enthusiastic for this extra governance layer. Yet, it could be questioned how much it matters how the region performs of the primate city is experiencing high growth levels. When a city is integrated in the network it can profit from the agglomeration externalities of the network. But the network is a club good and not everybody is capable to profit from the externalities of a network (XX). Despite all this, the municipalities should focus on their regional role. The different dimensions of borrowed size have shown wat the strength and weaknesses are of municipalities and policy should be targeted on these strength and weaknesses. But the path dependency of municipalities plays an important role. The soft capital of a place influences the degree to which a municipality can profit from the proximity of a large city. Future research should focus on the precise influence of path dependency and the extent to which cities and towns can profit from the proximity of a large city.
Inhoudsopgave
Preface ... 2 Dutch Abstract ... 4 English abstract ... 5 1. Introduction ... 9 2. Theoretical framework ... 12The role of the city ... 12
Path dependency and the city size ... 12
The network paradigm and the polycentric city ... 14
The influence of a large city on its surrounding ... 16
Factors that influence borrowed size ... 17
Conceptual framework ... 19
3. Research methodology ... 20
3.1 The study area ... 20
3.2 Operationalization of key concepts ... 21
The matrix of borrowed size ... 21
City size, location and time ... 21
Performance, functionality and demographics ... 22
Factors that influence the borrowed size dimensions ... 24
Operationalization table ... 24
3.3. The mixed methods approach ... 26
The quantitative method ... 26
The data collection ... 27
The data analysis ... 27
The qualitative method ... 28
Data collection ... 28
Data analysis ... 29
3.4 Strengths and weaknesses of research methodology ... 30
3.5 Research ethics ... 30
4. The research context ... 32
4.1 The Metropolitan region of Amsterdam ... 32
4.2 The regions of the MRA ... 33
Amstelland-Meerlanden ... 34
Gooi en Vechtstreken ... 35
Almere-Lelystad ... 36
IJmond ... 37
Zaanstreek-Waterland ... 39
4.3 The descriptive data ... 40
Descriptive data of functions ... 40
Descriptive data of demographics ... 40
Descriptive data of performance ... 40
5.1 Borrowed performance ... 42
Borrowed performance in 2018 ... 42
Borrowed performance in 2013 ... 43
Changes in borrowed performance ... 44
5.2 Borrowed functionality ... 45
Functionality in 2018 ... 45
Functionality in 2014 ... 47
Changes of functions from 2014 to 2019 ... 48
5.3 Borrowed demographics ... 48
Demographics in 2018 ... 49
Demographics 2013 ... 50
Changes in demographics ... 50
5.4 Borrowed size in the Metropolitan region of Amsterdam ... 51
6. Explanatory factors of borrowed size in the MRA ... 54
6.1 Network integration and location ... 54
6.2 Path dependency ... 56
6.4 Enlarging the borrowed size process ... 57
7. Conclusion and discussion ... 60
Borrowed size in two different points in time ... 60
Borrowed size and its explanatory factors ... 62
The future of borrowed size in the MRA ... 64
Discussion and further research ... 65
Bibliography ... 67 Appendix 1 ... 73 Appendix 1.2 ... 74 Appendix 1.3, ... 75 Appendix 2. ... 77 Appendix 3. ... 82 Appendix 4. ... 83
1. Introduction
Amsterdam, a city with a unique history that is part of an equally unique urban system. The city has experienced economic growth, population growth and increased international importance in the past years (MRA, 2016). But with 868.300 inhabitants, Amsterdam is a rather small city compared to other world cities. Big cities provide agglomeration externalities over their smaller equivalents and some argue that due to these externalities the big city triumphs over the smaller city (Glaeser, 2011). Agglomeration benefits are expected to result in more economic growth and higher productivity. But the Netherlands lacks cities of mega size and is densely populated with medium and small cities that, due to their size, experience less agglomeration benefits than mega cities (Meijers & Burger, 2017). And yet, Dutch cities still experience economic growth that is not subjected to that of other large metropolitan cities (Frick & Rodriguez-Pose, 2018). The metropolitan area of Amsterdam is one of the five strongest urban regions in Europe (MRA, 2018). But the region is not always profiting as much as they can. Almere wanted to get a dependence of the University of Amsterdam but didn’t get it (Khaddari, 2019), Zaanstad is the 16th city of the Netherlands while culturally it is the 41st
city of the Netherlands. Mayors of the municipalities near Amsterdam stated that Amsterdam is thinking less about the region and more about the city itself (Koops & Meershoek, 2019).
Amsterdam is regarded as a polycentric city that is part of the larger polycentric urban region of the Randstad. A polycentric city is built up out of multiple important centres. It consists out of cities and towns that are physically separate, but functionally networked and clustered around one or multiple larger central cities (Hall & Pain, 2006; Musterd & Kloosterman, 2001). The Randstad is one of Europe’s most distinct polycentric mega city region that combines several important centres, with international gateways and a cosmopolitan labour force (Hall & Pain, 2006). The North-Western flank of the Randstad is made up of the Metropolitan Region of Amsterdam. In 2016, this area organized itself in the governmental body the Metropolitan Region Amsterdam (MRA). The MRA consists of over 2,5 million inhabitants and combined it has population equal to mega cities like Paris. The Metropolitan region of Amsterdam consists of cities and towns like Almere with over 200.000 inhabitants and towns as Ouderkerk aan de Amstel with slightly over 8000 inhabitants (MRA, 2019). These towns and cities are very diverse. From old historically important cities like Haarlem, to new cities like Almere with an important residential function. But what they all have in common is their ties with Amsterdam. In the last few years, these ties have gotten stronger (MRA, 2018). The realisation of the MRA organization has been targeted on the intensification of the cooperation between the municipalities within the region. The MRA argues that, to be ready for the future, they must benefit from the scale of the region and form a strong coalition with the actors within the region. But external effects have also resulted in stronger ties between the Amsterdam and its surrounding cities. Many people, and especially families, moved to the surrounding towns of Amsterdam due to the high living costs in Amsterdam. 40% of young families have moved to the cities surrounding Amsterdam (CBS, 2017). Firms also move towards the region. Dutch chocolate company Tony Chocolonely will realise its creative head quarter in Zaandam stating that the area is actually already ‘Amsterdam’ (Van Zoelen, 2018). This raises the assumption that functionally, but also institutionally and culturally, the municipalities of the Metropolitan Region of Amsterdam are growing closer together.
While Erickcek & Mickinney (2006) argue that small and medium sized cities foresee a negative future, others have argued that second tier cities near large cities will have a bright future (Meijers & Burger., 2017; Camangi et al., 2015). As large cities experience
agglomeration externalities, they can also experience agglomeration diseconomies. The high concentrations in large cities can lead to congestion, high land prices and higher crime rates (Meijers, 2007). Small and medium sized cities will experience these effects in a lesser extent. Alanso (1973) argued that cities of smaller size retain from these agglomeration diseconomies but enjoy the advantages of larger sized cities through their access to other centres. The concept of borrowed size describes this phenomenon whereby a smaller city or metropolitan area has characteristics of a larger city if it is near a large city. These cities profit from the proximity of larger cities by borrowing critical mass. They may experience higher population growth, higher economic performance and more amenities in relation to their size (Phelps et al., 2001). Smaller sized cities are often capable to overcome their disadvantages through city networking (Camagni et al., 2013). Thus, where these cities experience less negative side effects of larger metropolitan areas like congestion and high prices, they can benefit from agglomeration externalities through integrating into a network of cities. But cities can also experience negative effects from the proximity of large agglomerations (Beherens & Robert-Nicoud, 2008). In some areas, the proximity of the large city will cast a shadow on a city. A smaller city might experience less well-developed cultural amenities due to the nearby large city that provides high levels of cultural facilities. Firms within smaller or medium sized cities might also experience more competition of inner-city actors that can be more productive (Burger et al., 2014).
The proximity of a large city can thus have positive or negative effects on its surrounding cities and towns. Meijers (2015) found that cities often experience higher economic growth levels than expected due to their proximity to larger cities, but less functions than expected (Burger et al., 2014). The differences between the benefits and the negative effects often differentiate between sectors or categories. The extent to which a city experiences benefits from proximity is influenced by several factors. One of them is the extent to which a city is integrated into the city network. Network integration can be distinguished into three different forms of integration, namely institutional integration, cultural integration and functional integration (Meijers, 2018). Network integration is seen as important for economic performance in the age of the city network paradigm. The city network paradigm claims that, through participation in the network, cities exploit scale economies through relationships and cooperative activities and perform better (Capello, 1998). But there is need for a deeper understanding of the driving forces of borrowed size, Volgmann & Rusche (2019) argued.
The metropolitan region of Amsterdam has anticipated on the increased importance of the network and has realized an institutional cooperation to enlarge the performance of the region. But the question remains how the proximity of Amsterdam benefits the performance and functionalities of the surrounding cities. And what factors influence the extent to which cities and towns near Amsterdam can profit from the proximity. This will expectedly lead towards a higher degree of borrowed size. This has led to the following research question:
“To what extent are the cities and towns in the metropolitan region of Amsterdam experiencing borrowed size and what influences this?”
This research question will be answered by the following sub-questions
1. To what extent do the municipalities of the MRA experience the different dimensions of borrowed size or agglomeration shadow?
2. How has the extent to which cities and towns experience borrowed size changed over the last few years?
3. What are the underlying causes of borrowed size in the Metropolitan region of Amsterdam?
The theoretical framework of the research question will be discussed hereafter. The methodology of this research will follow after the theoretical chapter. The research question will be answered with the use of a mixed methods approach. The first two sub questions will be analysed with the use of quantitative statistical analysis. The last sub question will be answered by applying a qualitative methodology. Thereafter, the research context will be discussed. The metropolitan region of Amsterdam and its regions and municipalities will be highlighted. The descriptive data of the quantitative analysis will also be discussed in this chapter. This will be followed up with the first results chapter. This chapter will discuss the first two sub questions, namely the dimensions of borrowed size in the MRA and in what manner this has changed the last years. The second results chapter will discuss why some municipalities experience borrowed size and others do not. It will also discuss why specific changes over time have taken place. The conclusion and discussion will combine the literature with the results and will make suggestions for further research.
2. Theoretical framework
This chapter will describe the theoretical concepts to answer the research question. First, the city and its size will be discussed. Secondly, the polycentric city and the city network paradigm will be highlighted. Thirdly, borrowed size and agglomeration shadow will be discussed. Lastly, the conceptual framework will be introduced.
The role of the city
In 1933, Walter Christaller introduced the central place theory. The central place theory explains the number, size and distribution of cities and towns (Getis & Getis, 1966). Christaller argued that there is a central place that hosts a high level of functions which is of importance for its surrounding region. Christaller argued that a city has a specific level of functions based on the population of the city and on the catchment population of a city. The catchment population entails the number of people that is in a specific distance of a facility. He argued that people are prepared to travel a specific amount of time for specific functions. People want to have a bakery nearby, but they are prepared to travel longer for a hospital. The large central place is surrounded by smaller settlements that are dependent on the central place. City size can thus be explained by a scientific law. But in the contemporary society that is influenced by globalization and informationalization, the central place theory fails to explain the number, size and distribution of cities and towns (Hall & Barret, 2012).
In the twenty-first century, cities are experiencing prosperous times and more people are moving towards the city than out of the city (Van Oort & Meijers, 2015). People and firms experience positive effects from being located in a city through agglomeration benefits. Cities make actors more productive, more innovative and speed up growth. The possibility to share infrastructure, to match skills, suppliers and markets and to exchange information between people are among the factors that make cities more productive (Beherens & Robert-Nicoud, 2008). The scale of agglomeration benefits is large in the Netherlands. The region is often able to profit from the agglomeration benefits that are created by a city (Atzema et al., 2012) But agglomerating in a city can also lead to negative outcomes, such as congestion (Tabuchi, 1977). As city size increases, spatial complexity increases and the average commuting time prolongs (Henderson, 1997). Larger cities are places of high productivity and phenomenal wealth, but also places of inequality and poverty (Beherens & Robert-Nicoud, 2008). Path dependency and the city size
The size of the city is an important stimulator of both agglomeration benefits and agglomeration shadows (Frick & Rodriques-Pose, 2018). This calls for a balance between the two forces. There is a consensus on the fact that positive externalities can turn into diseconomies beyond a specific city size. This could lead to the assumption that there is an optimal city size. However, the presence of one single optimal city size is abandoned (Camagni et al., 2013). Alanso (1973) argued that the optimal city size is dependent on the context of a city. The city is dependent on its historic trajectory. Decisions that have been made in the past and the knowledge that has been acquired over time influences the development of a city. A specific choice sometimes eliminates another one and leads a city towards a specific path. This is referred to as path dependency (Boschma et al., 2012). Past events and historical structures determine the future growth developments of a region. The path dependency of a region has an influence on the economic growth of a place. It can lead to windows of opportunities, but also to lock-in situations (Bontje & Musterd, 2011). Jane Jacobs argued that cities with a rich and long history that developed strong economies will probably be the places
that experience high growth rates in the future (Atzema et al., 2012). This raises the question whether performance of a city can be influenced, or that it is already determined by the path dependency of a place.
Research has shown that larger cities often have greater productivity and economic growth than smaller cities. Productivity increases with 3 to 8% when the size of a city doubles (frick & Rodriques-Pose, 2018). Glaeser (2011) found that cities tend to be 30% more productive than rural areas. This high productivity can be attributed to the agglomeration benefits of a city. But a larger city also houses the factors needed for the contemporary global economy. Modern commerce requires more real-life interaction and a knowledge intensive labour market which are present in the large metropolis. Moreover, the presence of unique consumer amenities such as exotic restaurants, museums and specialized health care facilities contribute to the higher level of productivity of the large city (Partridge et al., 2010). In Europe, Metropolitan areas have five distinct functions according to the BBSR (2011). First, cities have a political function that entails of national government institutions and supranational government institutions. Second, cities have an economic function that contains of large enterprises, Advanced Producer Service (APS) offices, Banks and trading markets. Third, cities have a scientific function that contains of education and research institutions, scientific communication institutions and patent offices. Fourth, cities have a transportation function that contains of passenger transportation, transportation of goods and data transportation. And lastly, cities have a cultural function that consists of arts and sports. Larger cities, or so-called megacities, are the cities that connect the nations in a global network (Hall & Pain, 2006). But megacities are also expensive machines. They require large social investments and they have expensive real estate markets with capital intensive buildings according to Camagni et al. (2015). They argue that second-rank cities (cities with 200.000 to 1 million inhabitants) may have higher public resources efficiency and better quality of life than first rank cities. Second rank cities are often the main driving force behind national economic performances and they are also more resilient in terms of economic slowdown. Cities of intermediate size are able to profit from agglomeration benefits, whilst also being able to avoid agglomeration diseconomies (Capello, 2000).
The economic performance of small (less than 20 thousand inhabitants) and medium (20 to 100 thousand inhabitants) sized cities has received attention from scholars (Adam, 2006). Adam argues that the advantages of the medium sized city lies in their medium population density which will likely result in less congestion. These cities can provide better balance between the advantages of cities and the diseconomies that arise with a large city (Frick & Rodriques-Pose, 2018). But the performance of small and medium sized cities is very different. Some are faced with low or even negative growth, while others experience high growth (Dijkstra et al., 2013). Erikceck & Mickinney (2006) have argued that these cities are held back because they often have a manufacturing heritage, lack tolerance and diversity and have an aging population. The ESPON TOWN project (Servillo et al., 2014) found that these cities are often characterised by larger industrial employment and a smaller proportion of service jobs and that they have a smaller share of working adults with a degree. But service sector jobs and higher educated workers are of great importance for economic growth in the contemporary economy (Hall & Pain, 2006). Medium and smaller sized cities are also in disadvantage of the great metropolis because they lack the critical mass to offer specific amenities and economic activities (Capello, 2000). But by networking together, cities might be capable to overcome these negative externalities.
The network paradigm and the polycentric city
Many scholars argue that cities cannot be studied in isolation (Peris, 2018; Meijers, 2017; Van Oort & Meijers, 2015; Camagni, 2013). Cities are organised in systems of cities that are characterised by strong interdependencies that develop at the scale of a large region, a nation, or a continent, or even on the global scale (Peris, 2018). Cities are connected through a network of flows of information, capital, goods and persons that move along the infrastructure between the cities (Meijers, 2007). A network is a cooperation between individuals, corporations and territorial partners (Capello, 2000). It is built up of nodes (the city) and links that connect these nodes to facilitate interaction (Johanson & Quigley, 2004). In the time of Walter Christaller, cities had to draw from their (catchment) population for performance and functionality. In the 21st century, cities are part of a network and can draw
from cities and towns that are further away. Capello (2000) argues that the city network paradigm is a successful theoretical framework that overcomes the often limited explanatory power of the central place theory. The city network paradigm claims that through participation in a network, cities can exploit scale economies. Being connected in a city network allows cities to achieve a larger equilibrium size (Camagni et al., 2013). The links between the nodes may lead to the same effects that arise from agglomeration benefits that a single large settlement experiences due to its size. Networks of actors divided over space may substitute for agglomeration benefits of a single location (Johanson & Quigley, 2004; Van Oort et al., 2015). Urban networks are often related to synergy according to Meijers (2007). This refers to the situation where cities that are part of a network perform better because, networked together, they are more than the sum of its parts. Synergy can result in larger critical mass in a primate city, or in a situation of complementarity where actors have a distinct role in the network. Johanson & Quigley (2004) warn for the fact that the network is a ‘club good’ where exclusion is possible. Agglomeration externalities can reach everybody around, but not everybody is always capable to profit from the externalities of a network.
Polycentrism explains the existence of multiple centres in one area (Musterd & Kloosterman, 2001). They are historically distinct and independent on administrative and political terms but are well-connected through infrastructure and linkages (Meijers, 2007). A polycentric urban region distinguishes itself from the monocentric urban system in that it has a sharp divide between the city and its suburban surroundings. The polycentric city lacks a dominant centre (Musterd & Kloosterman, 2001). Hall & Pain (2006) argue that a polycentric city region can consist of 10 to 50 cities and towns that might have several dominant centres, or only one dominant centre. Cities within a polycentric urban region are physically separate, but functionally clustered. The settlements within the polycentric city act as separate entities where local residents work and live. They are also part of the wider functional urban region (FUR). Parr (2004) distinguished various conditions that need to be present in a polycentric region. Figure 1. shows the characteristics of a Polycentric Urban Region. The definition of a polycentric urban region thus differs between scholars.
Figure 1. Characteristics of the Polycentric urban region (Parr, 2004).
Cities and towns in the polycentric city seek a spatial integration of particular regions through city network development (Burger et al., 2014). By networking with other cities, they can enlarge their agglomeration benefits and overcome diseconomies that are connected to cities of smaller size (Camagni et al., 2015). It is the question whether distance matters as much as network integration. It is argued that distance is often merely a condition for interaction because it mitigates interaction (Van Oort et al., 2015). But Glaeser et al. (2016) state that distance still matters very much. People often tend to draw from their own urban network. But there is a consensus on the fact that local actors can benefit from the actors in the larger urban network. A polycentric city is sometimes associated with low performance because it can lack agglomeration benefits and has a duplication of lower-order functions (Burger et al., 2014). But the polycentric structure still became a popular urban landscape in advanced economies (Musterd & Kloosterman, 2001). The polycentric urban region is often seen as a city structure that has the potential for superior economic performance (Parr, 2004). The rise of the advanced producer service (APS) firms in the globalized economy has contributed to the rise of the polycentric cities due to their strong network integration (Hall & Pain, 2006). Knowledge intensive services that are provided by these advanced producer service firms are a central feature of the contemporary economy. APS firms often cluster in cities because that is where knowledge can be found and shared (Taylor et al., 2014). APS firms are also extensively networked together, both regional, national and international (Taylor et al., 2014; Hall & Pain, 2006). These firms thus create tighter networks between cities. And in the contemporary global economy, these networks are of great importance (Hall & Pain, 2006; Glaeser et al., 2015)
To be part of a network is not enough. It too matters in which way a city is integrated into the network of cities and towns. The fact that a city is part of a polycentric city network, does not necessarily imply that there are strong functional linkages between the centres (Burger et al., 2014). To benefit from the participation within a network, it is important to be integrated. Cities that are poorly integrated often perform less than those that are integrated (Van Oort et al., 2015). Note that a region can be integrated on one linkage, while it is monocentric and poorly connected with respect to other types of linkages. This is referred to as multiplexity. Actors within a place can also have very different spatial interaction patterns. While some people and firms are very well connected, others are not. This is referred to as individual-level heterogeneity (Burger et al., 2014). Meijers et al. (2018) used three different dimension of network integration to determine the integration of a city into a network.
1. There is a clustering of distinct centres that are separated from each other by land
2. There is a maximum level of separation between the cities. One hour of travel time is the maximum distance.
3. There is a minimum level of centres. A twin city does not count for a polycentric city
4. There is no pronounced differentiation by size. No centre has a population dominance
5. There is interaction between the centres is greater than city interaction in other given regions
6. Centres of the polycentric urban region are more specialized than non-polycentric regions
Namely, functional integration, cultural integration and institutional integration. Functional integration refers to the physical connection between cities and the connectivity of the actors within these cities. If a city is functionally integrated, companies share and match their demand and supply, people are capable to join the labour pool of the network and actors can move around the cities through good infrastructure (Meijers et al, 2018; Dijkstra et al., 2013). Cities are also functionally integrated when there are many travellers between the cities, both consumers and business travellers (Burger et al., 2014). Cultural integration refers to the extent in which the cultures of cities are linked to each other. Emotional ties and a shared identity are factors that influence the cultural integration of a city within a network. Likewise, the shared political views and a common language within regions influences the cultural integration of a network (Meijers et al., 2018). Boschma et al. (2012) state that mutual norms and values bring actors together and will result in a better matching, learning and coupling process. However, Glaeser et al. (2016) has argued that local cultures might get lost when cities are strongly integrated or morphed together. This might not be beneficial for creativity and eventually for economic growth. Institutional integration refers to the extent to which the institutions of a city are linked together. A central governance institution provides cities the opportunity to launch and run joint policies (Capello, 2000). Yet, many cities are only willing to take part in joint governance institutions when there is a win-win situation (Burger et al., 2014).
The influence of a large city on its surrounding
The capacity of a city to access the advantages of another cities is explained by the ‘borrowed size’ concept. Alanso (1973) introduced the concept in the 1970’s in relation to his research about the optimum size of a city. He argued that smaller cities can enjoy the benefits of larger cities through their easy access to other centres, while retaining from the disadvantages of the large city like congestion and high prices. Smaller cities can profit from the high-end facilities of the large city, businessmen can share facilities together with the large city and smaller cities can draw from a larger labour pool (Alanso, 1973). Small cities near large agglomerates have the best of both worlds. They experience agglomeration benefits, while they retain from agglomeration diseconomies (Phelps et al., 2001). Alanso stated that the phenomenon was quite visible, especially in the Netherlands and the other low countries. Nevertheless, the concept remained understudied for a long time. Phelps (2001) applied the concept in a research about the performance of firms in small settlements in greater London. He argued that accessible locations nearby a large city have financial benefits of the smaller settlement such as low rents, while they profit from the positive externalities of the large city such as innovations. This concept gained popularity in the following years in relation to the performance of towns and cities near large settlements, especially in relation to the polycentric city structures in the low countries. Cities within a polycentric city are more likely to borrow size than cities that are part of a monocentric city region or isolated cities (Van Oort et al., 2015). They might also benefit from ‘crowding out’ processes of the large city. Actors could get pushed out of the city due to agglomeration diseconomies such as high land prices. Smaller cities surrounding the large city are in consequence attractive alternatives (Meijers & Burger, 2017).
For cities to experience borrowed size, Meijers (2015) argues that there must be a disconnection between the expected size and the actual size of a city. He has decoupled the concept of size into two measurements. He distinguished a functional approach and a performance approach. The functional performance relates to the borrowed functions that a
city can experience. A city experiences borrowed functions when it has more functions than expected in relation to the size. The performance approach relates to the borrowed performance. A city experiences borrowed performance when a city is economically performing better than expected. But there is also a different side of the coin. The development of a large, productive city might hinder the further development of nearby cities (Beherens & Robert-Nicoud, 2008). Krugman refers to the concept of ‘agglomeration shadow’ in his new economic geography. He predicts a shadow effect from large cities over their surroundings due to the higher competition level (Krugman, 1998). It can thus be expected that places close to large cities have less amenities than isolated cities of the same size. Meijers
et al. (2015) introduced a matrix of borrowed size where the different dimensions are visualized structurally (figure 2). In the matrix, the connection between size and functionality and size and performance is made.
Figure 2., dimensions of borrowed size (Meijers et al, 2015)
Factors that influence borrowed size
Research showed very mixed results on the question whether cities near large agglomerations experience borrowed size. The characteristics and context of a city influence the extent to which a city can borrow size from its surrounding. Meijers et al. (2015) found that large cities often improve the performance of its surrounding towns and cities compared to isolated towns of similar size. They experience borrowed performance. However, these towns often experience less functions than expected in comparison to isolated towns of similar size. The ESPON town project (Servillo et al., 2014) found that smaller settlements in remote areas experience negative growth trends, while their counterparts near large settlements exhibit better performances. They do however state that they could become ‘dormitory towns’ where people live but economic activity is lacking. But when such a ‘dormitory town’ is sufficiently connected with a large city, they are capable to profit from the nearby large city. However, these smaller cities often fail to fully exploit the network. Van Oort et al. (2015) found that only a small portion of Dutch cities borrow functions. The majority of the cities and towns near large cities have less functions than expected. These places were often smaller settlements, larger settlements were more likely to experience borrowed functionality. Especially cities that have more than 400.000 inhabitants. Meijers & Burger (2017) came to the conclusion
that borrowing size is not reserved for the smaller settlements and that larger cities are even more likely to borrow size. Sometimes it is the primate large city that borrows the most size from its surroundings. It can thus be concluded that city size has an influence on the borrowed size performance of a city. Volgmann & Rusche (2019) found that small cities in close proximity to larger growing cities can also benefit from the spatial proximity and connectivity to the functions. They can enrich themselves functionally and structurally.
Camagni et al (2015) state that cities are capable to overcome the disadvantages of their small size by borrowing size from nearby large metropolitan areas by getting access to the functions and network these areas host. Their study has shown that outperforming cities are those that are capable to borrow size from large metropolitan areas through network participation or through innovation. However, according to Glaeser et al. (2016), greater network connectivity creates more competition and can generate agglomeration shadows on the city region scale. Meijers & Burger (2017) found that stronger integrated cities perform better than cities that have neighbouring cities but are moderately or weakly integrated. However, stronger integration often results in lower functional levels. Especially functional integration contributes to borrowed size, followed by institutional integration (Meijers et al., 2018). They found no significant relationship between cultural integration and borrowed size. They did argue that the scale of network integration matters. On the national or international scale, deeper network integration sometimes resulted in borrowed size, while local network integration sometimes resulted in agglomeration shadow. The institutional integration of the Randstad is limited according to Salet et al. (2009) because governments compete with each other on the spatial levels and they have different loyalties. Yet, the Randstad is still performing good. However, they argue that it could be better if the institutional integration within the region would improve.
Proximity also still matters for borrowed size. Gunnar Myrdal argued that the direct area around a city can profit from spreading out, or crowding out, effects when the city is experiencing growth. On the other hand, the city can also have a backwash effect on the region where people and companies move from the region to the central city (Atzema et al., 2012). On the other hand, geographic remoteness is associated with lower productivity in the United States and areas that are more remote from high-tier urban areas often perform less good economically then cities and towns that are closer to the high-tier urban areas (Partridge et al., 2010). However, remote cities often host more functions and show a lower correlation between size and functionality. This could also be linked to the historic trajectory and political patterns (Meijers & Burger, 2017). The historic trajectory matters a lot for the performance of a city or town. Meili & Mayer (2017) found that Swiss towns often performed better due to their historic trajectory and where hosting high tech firms, even though their regional context did not differ a lot from their counterparts. The cumulative causality of past events can lead to higher growth. Cumulative causality refers to the notion where, for example, the settlement of a specific company in a place can lead to stronger growth. The company attracts other actors and these actors can result in the creation of a cluster. The cumulative causality of a place can influence the entire city and region around it (Atzema et al., 2012). Meili & Mayer (2017) also state that the economic sectoral composition of a city matters for the ability to borrow size from a nearby city. Service orientated towns may be able to benefit more from the economic dynamics in a near city than industry orientated towns. Thus, a city that has an economic sectoral composition that is linked to the primate city is more likely to experience a degree of borrowed size.
It can be concluded that the borrowed size effects of cities differs to a great extent. It matters what the size of the city is, how the city is connected and on what scale a city is connected, what the geographical location of a city is in the region, what the economic sectoral composition is and what the historic path of a city is. It can also be expected that cities and towns are more likely to borrow performance, than functionality. Table 1 shows the scholars that refer to borrowed size or agglomeration shadow in relation to a specific factor. Table 1. Factors that influence borrowed size
Factors Borrowed size or agglomeration shadow Larger cities Van Oort et al., 2015; Meijers & Burger, 2017
Smaller cities Servillo et al., 2014; Meijers & Burger, 2017; Van Oort et al., 2015; Meijers et al., 2015; Volgmann & Rusche, 2019
Network integration Servillo et al., 2014; Meijers et al., 2018; Camagni et al., 2015; Glaeser et al., 2015; Van Oort et al., 2015; Salet et al., 2009
Proximity Partridge et al. (2010); Servillo et al., 2014;
Contextual influence Meili & Mayer, 2017; Meijers & Burger, 2017; Atzema et al., 2012 Sectoral composition
economy Meili & Mayer, 2017
Conceptual framework
The literature has resulted in the following conceptual framework (Figure 3). The borrowed size or agglomeration shadow a city experiences exists out of the performance of a city, its functionality and the demographic composition of a municipality. The functional and performance dimension are derived from the stretched definition of borrowed size by Meijers & Burger (2017). The demographic dimension will complement these two dimensions because it has been argued that the human aspects matter a lot to profit from growth in the contemporary economy (Erikceck & Mckinney, 2006; Servillo et al., 2014). The dimension of borrowed size are influenced by the size of a city, the location, the network integration and path dependency. The concepts will be operationalized in the following chapter.
Figure 3, conceptual framework
City performance Borrowed size Agglomeration shadow Network integration Size Location City functionality
3. Research methodology
The role of borrowed size in the metropolitan region of Amsterdam was researched with the use of a mixed methods approach. This chapter will first discuss the research area and the time frame. Secondly, the variables of the research will be operationalized and an operationalization table will be introduced. Thereafter, the mixed method research design will be discussed. Both the qualitative approach and quantitative approach will be discussed. The data collection and analysis method will be discussed per approach. To conclude, the strengths and limitations and research ethics will be discussed.
3.1 The study area
The metropolitan region of Amsterdam consists of 32 municipalities that are divided over seven regions (Figure 4). The metropolitan region of Amsterdam (MRA) was used as the study area. The MRA area is part of an informal governance platform that will be further discussed in the context chapter. The 32 municipalities of the platform were chosen for the analyses due to the cooperation of the municipalities in the platform. These municipalities are part of the daily urban system of the city of Amsterdam and have a relation with Amsterdam and each other on diverse aspects. Therefor the MRA is the suitable study area to study the concept of borrowed size in the Amsterdam city region. The Netherlands divides the space in four dimensions. The Netherlands is divided into 12 provinces, 355 municipalities, over 3000 boroughs and over 13.000 neighbourhoods (CBS, 2018). The CBS provides data on all these different spatial levels. The level of the municipality is the most suitable level for this research
due to the importance of this governance level and the data availability. From here on, cities and towns will be referred to as municipalities.
3.2 Operationalization of key concepts
To answer the research question, the following concepts were operationalized; borrowed size, agglomeration shadow and the factors that influence the borrowed size dimensions. First, the matrix of borrowed size will be discussed. Second, the city size variable will be introduced. Third, the different dimensions of borrowed size are highlighted. And lastly, the factors that influence borrowed size will be discussed.
The matrix of borrowed size
The stretched definition of borrowed size that was introduced by Meijers et al. (2015) was used to analyse the extent to which municipalities profit from the proximity of Amsterdam. Meijers introduced a stretched version that includes two dimensions, namely borrowed functionality and borrowed performance. A third dimension of borrowed demographics was added to the analysis. This dimension adds a demographic aspect to the analysis and analyses to what extent a municipality experiences demographic characteristic of a large city. By adding a demographic dimension, the economic gaze will be complemented with a human centred dimension. This dimension aims to indicate how large the number of inhabitants is that have characteristics of large city inhabitants, or so called ‘metropolitans’. Table 2 displays the expended dimensions of borrowed size. A municipality is experiencing a borrowed dimension if it performs better than expected. A municipality can also perform less than expected and experience an agglomeration shadow dimension. But a municipality can also perform as expected and not experience a borrowed size effects. If a municipality outperforms on all dimensions, it experiences borrowed size. If a municipality outperforms in two out of the three indicators, it is described by one of the following dimensions. Borrowed economics refers to a municipality that performs better on the more economic dimensions of performance and functionality. Borrowed living environment refers to an outperforming municipality on the dimensions of functionality and demographics that is linked to a living environment that is ‘larger’ than expected.
Table 2, dimensions of borrowed size
Dimension of Borrowed size Performance Functionality Demographics
Borrowed size X x x
Borrowed economics x x
Borrowed living environment x x
Borrowed human capital x x
Borrowed demographics x
Borrowed functionality x
Borrowed performance x
Expected
City size, location and time
Borrowed size was investigated in two different points in time in the MRA. By comparing two different points in time, the changes of the dimensions can be linked to certain developments. The concept of borrowed size has been analysed in 2018 and 2013. In 2018, the MRA
partnership was realized two years ago and Amsterdam was experiencing high growth levels. In 2013, the MRA partnership was not realized yet and the economic crisis still held the region in its power. Possible changes in 2018 could be linked to stronger integration, but also to the changed economic times. The spatial location and population size are derived from quantitative information from the CBS. These indicators can play a role in the degree in which municipalities profit from the proximity of Amsterdam. Municipalities were categorized into different sizes. Large cities are those with over 250.000 inhabitants. Medium sized cities are those that have more than 100.000 inhabitants. Averaged sized cities have between 100.000 and 50.000 inhabitants. Small cities have less than 50.000 inhabitants. The physical proximity was categorized into municipalities that are physically connected to Amsterdam and those that are not physically connected to the city of Amsterdam.
Performance, functionality and demographics
The different dimensions of borrowed size were measured with the use of several indicators. The performance of a municipality is based on three indicators, namely average house value, growth of jobs and population growth (Table 3). These three indicators are diverse and provide a holistic image of the performance of a municipality. The growth of jobs and population growth have been used by other scholars to measure the performance (Volgmann & Rusche, 2019; Meijers et al, 2015). The growth of jobs and population were controlled for by possible mergers with other municipalities to overcome corrupt outcomes. Edam-Volendam and Bloemendaal both underwent a merger in the time frames and the number of inhabitants from the joined municipalities was extracted from the total. The average property value per square meter will complement these indicators. Amsterdam has experienced exponential growth of its property value and this indicator can indicate how the proximate municipalities were influenced by this phenomenon (3).
Table 3, indicators to measure performance
DESCIRPTION OF DATA SOURCE YEAR SCALE LEVEL GEOGRAPHI
C SCALE AVERAGE HOUSE VALUE PER
M2 (WOZ VALUE) Huizenzoeker 2008 - 2018 municipality The Netherlands
GROWTH OF JOBS CBS 2010
-2017 municipality The Netherlands
POPULATION GROWTH CBS 2011 -
2017 municipality The Netherlands The growth of jobs and population were compared in percentages so the municipalities can be compared with each other. Ranges of 3 to 5 years were used for these indicators to enlarge the validity and limit the influence of random events. The average property value per square meter is compared in absolute numbers. Some scholars (frick & Rodriques-Pose, 2018; Glaeser, 2011) have argued that a large city often shows more growth compared to smaller and medium sized cities. If a smaller or medium sized city is experiencing more economic growth, it is expected that the proximity of a large city plays a role in this performance.
Table 4, indicators to measure functionality
DESCIRPTION OF DATA SOURCE YEAR SCALE LEVEL GEOGRAPHIC
SCALE NUMBER OF ADVANCED PRODUCER
SERVICE FIRMS KvK Feb 2019 Towns and cities The Netherlands
NUMBER OF CREATIVE FIRMS AND
INSTITUTIONS KvK Feb 2019 Towns and cities The Netherlands
NUMBER OF SCIENTIFIC FIRMS AND
INSTITUTIONS KvK Feb 2019 Towns and cities The Netherlands
NUMBER OF CAFES AND RESTAURANTS KvK Feb 2019 Towns and
cities The Netherlands The functional level of municipalities was measured with the use of four indicators (table 4). The number of creative firms, advanced producer service firms (APS), higher educational institutions & firms and the number of restaurants and cafés were analysed. The presence of advanced producer service firms, creative firms and scientific institutions and firms can be linked to large cities according to the BBSR (2011). The number of restaurants complements the analyses because it focusses on consumer functions and is more directly linked to the number of inhabitants of a municipality. The dataset of the chamber of commerce of the Netherlands provides data about all the registered firms in the Netherlands. The dataset provides information about the location of firms, what the core activity is, when it was founded and when the firm located in the municipality. The activity of a firm is described with SBI codes. Multiple SBI codes were merged together to create the desired categories. Initially, the dataset was also not on the desired spatial level of the municipality. This has resulted in a data editing process to translate the spatial level towards the municipality by using ArcMap and SPSS. The creative firms and institution indicator consist of categories that are linked to art, publishing, design and the movie, tv and radio sector. The advanced producer service firms indicators consists of ICT firms, financial firms, insurance firms, law firms and consultancy firms. These firms are knowledge intensive and meet the requirements of advanced producer service firms provided by Taylor et al. (2014). The indicator is broad and also consists of smaller firms. However, the data does not provide the possibility to exclude smaller firms from the analysis. The data provides information about the pattern and relations, but it might not provide information about the precise number of advanced producer firms. The scientific and educational indicator consists of universities and research firms. The restaurant and café indicator consist of the restaurants and cafés. Appendix XX provides more detailed information about the included firms and institutions.
Table 5, Indicators to measure demographics
DESCIRPTION OF DATA SOURCE YEAR SCALE LEVEL GEOGRAPHIC
SCALE NUMBER OF VOTES FOR GREENLEFT
AND D66 Kiesraad 2018 and 2014 Municipality The Netherlands
NUMBER OF HIGHER EDUCATED
INHABITANTS CBS 2018 and 2013 Municipality The Netherlands
The demographic dimension is measured with the use of two indicators (table 5). A large city has different demographic characteristics than smaller cities and towns. This dimension of
borrowed demographics can indicate how inhabitants of Amsterdam have spread out over the region and how the demographic composition of the municipalities has changed over the years. The media has written about the pull to the region for the inhabitants of Amsterdam. These people are often higher educated (CLO, 2018). They also are more likely to vote on left wing progressive parties (Couzy, 2019). D66 and GroenLinks both perform very well in the city of Amsterdam and the areas where inhabitants get ‘pushed out’. Especially within the city centre and the older more dynamic parts of the city these political parties do well (De Voogd, 2012). By comparing the amount of higher educated inhabitants and the number of votes for the ‘large city parties’ to the expected value based on the city size, it was analysed whether a city is acting different than expected on demographic characteristics.
Factors that influence the borrowed size dimensions
The literature framework introduced several factors that influence the concept of borrowed size (see table 1). The network integration, path dependency, location and city size are the factors that were derived from the literature that influence the extent in which a city profits from the proximity of a large city. The city size and spatial location have been discussed before and were derived from a secondary data file from the CBS. The network integration is divided into the three aspects of institutional integration, cultural integration and functional integration by Meijers et al (2018). They have provided several indicators that can measure the extent to which a municipality is integrated in a network. The influence of path dependency has also been acquired through primary qualitative data collection. Path dependency refers to the historical development. The indicators that measure the network integration and path dependency are displayed in operationalization table (table 6).
Operationalization table
Table 6 provides a detailed operationalization of the key concepts. The concept of borrowed size is operationalized into quantitative factors. The explanatory factors of borrowed size are operationalized into qualitative factors. The next section will discuss the research methods and the data analysis.
Table 6., operationalization of key concepts
CONCEPT DEFINITION INDICATORS MEASUREMENT
DIMENSIONOS OF BORROWED SIZE PERFORMANCE
LEVEL A city is performing better in relation to the national average due to the proximity of other cities
- Growth of jobs - M2 value of property - Population growth - Discrepancy real value and average value.
FUNCTIONAL
LEVEL A city has more (high-tier) functions than expected in relation to the size due to the proximity of other cities - Number of firms in the creative industry - Number of Advanced producer service firms - Discrepancy real value and expected value based on a linear regression in
- Number of restaurants and cafes - Number of educational and scientific firms and institutions relation to city size DEMOGRAPHIC
COMPOSITION A city has more inhabitants that are characterized as ‘metropolitans’ than expected in relation to the population size
- Number of progressive voters (GroenLinks and D66) - Number of higher educated inhabitants - Discrepancy real value and expected value based on a linear regression in relation to city size AGGLOMERATION
SHADOW A city performs less than expected and has less functions than expected in relation to its size due to the proximity of other cities - Negative dimension of borrowed size - A majority of the indicators of a dimension is a negative residual of -1 standard deviation.
CITY SIZE The size of a city based on
the number of
inhabitants per
municipality
- Large city: over 500.000 inhabitants - Medium large city: over 100.000 inhabitants - Medium sized city: over 50.000 inhabitants - Small city: less
than 50.000 inhabitants - CBS data on population size NETWORK INTEGRATION FUNCTIONAL
INTEGRATION A city is functionally integrated when firms and people can share and match their supply and demand and if it is
physically well connected
- Common labour pool - Number of travellers between cities - Good infrastructure - Matching of supply and demands of firms
- See topic list interview
- Physical location CULTURAL
INTEGRATION A city is culturally integrated when the cultures of other cities are linked to each other.
- shared identity - emotional ties - political homogeneity - common language
- See topic list interview
INSTITUTIONAL
INTEGRATION A city is institutionally integrated if the institutions of the cities are linked with each other
- central governance institution - Engagement in central governance institution
- See topic list interview
PATH
DEPENDENCY Decisions or events from the past influence the city for a long period of time. A specific choice sometimes eliminates another one and leads a city towards a specific path. - Historic development trajectory - Economic composition
- See topic list interview
3.3. The mixed methods approach
A mixed methods approach combines a qualitative and quantitative research method within one project. It can provide a more complete answer to the research question (Bryman, 2012). Quantitative data analysis is combined with expert interviews to enrich the results. The mixed methods approach enriches the completeness, but it has also been used to answer a different component of the research question. Bryman (2012) has distinguished completeness and answering different research questions as reasons to apply mixed methods. The quantitative method is complimented with a qualitative approach to provide a more detailed understanding of the research.
The quantitative method
The extent to which a municipality experiences borrowed size is researched quantitatively. The aim of this phase is to discover which municipalities are experiencing specific dimensions of borrowed size. A deductive research approach was applied during this phase. The existing literature on the concept of borrowed size were tested in the metropolitan region of Amsterdam in two different points in time. Namely 2018 and 2013. By analysing two different points in time, the changing innerworkings of borrowed size were discovered. The aim was to gather data from 2018 and 2013. Due to data availability this was stretched between 2017 – 2019 and 2013 – 2014. The concept of borrowed size translates into a disconnection between city size and the performance and functionality of a municipality (Meijers, 2015). The quantitative analysis unravels this disconnection by comparing the expected value to the real value. The real value is based on the acquired data about the performance, functionality and demographics of the municipalities. The expected value is based on quantitative analyses that was conducted with the use of the aggregated dataset.
