Linking processes and patterns: Spatial planning,governance and urban sprawl in the Barcelona and Milan metropolitan regions

Abstract

Chinese

Governance dynamics and spatial planning regulations are significant factors in the occurrence (or containment) of urban sprawl. However, qualitative investigations of the planning regulatory systems and practices, and governance arrangements that cumulatively stimulate suburbanisation, typically remain detached from land-change analyses. Based on the concept of institutional frames of spatial planning systems, this article elucidates how governance dynamics and spatial planning practices, at different scales, can partially explain suburban land-use patterns. The territorial transformations of two Southern European metropolitan regions, Barcelona and Milan, are examined through land-use data (1990–2012) at different territorial scales. Demographic (1991–2011) and administrative (2011) data are also analysed. In-depth interviews about individual and collective land management practices have been carried out, as well as document analysis concerning spatial planning laws and regulations. This research shows that the metropolitan character of urban sprawl originates from local planning practices mainly performed by municipal authorities through land-use micro-transformations. Further, it highlights the decisive role that higher-level institutions can play in land containment. Urban sprawl is hence not necessarily an unplanned phenomenon, but rather a ‘differently planned’ local and regional land-use strategy.

Keywords

governance land-use change spatial planning suburbanisation urban sprawl

Introduction

Recognised as a global phenomenon (Keil, 2013), urban sprawl has extensive environmental, economic and social impacts, such as land consumption and fragmentation, loss of fertile agricultural land and social costs related to the lack of services for an ageing suburban population (EEA/FOEN, 2016).

Many factors have been identified as influential driving forces of urban sprawl. Demographic growth, affordable private car mobility, rising household incomes and urban land rent can encapsulate suburbanisation in the classical monocentric city model (Antrop, 2004; Brueckner and Fansler, 1983; Brueckner, 2000; Oueslati et al., 2015). Socio-economic determinants of urban sprawl, in particular accessibility, have been shown to be key in stimulating suburbanisation (Ewing, 1997; Herce, 2005; Weilenmann et al., 2017). Spatial fragmentation of rural amenities may also be a powerful driving factor for the occurrence of sprawl, as homebuyers value the closeness to open (although gradually more fragmented) space (Cavailhès et al., 2004). Availability of (flat) agricultural land is positively correlated with urban sprawl, as farm land is often overwhelmed by increasing urban pressure (Mazzocchi et al., 2013). Citizens’ housing preferences and the low quality of the urban environment in inner cities may also trigger urban sprawl (Couch and Karecha, 2006; Reckien and Luedeke, 2014). However, advocates have emphasised the many benefits related to a suburban ‘multi-car based lifestyle’ (Glaeser and Kahn, 2004: 2), and disagreement exists regarding whether urban compaction is necessarily desirable (Breheny, 1997).

Both supporters and critics of urban sprawl have acknowledged that public intervention is important, either for ‘correcting’ the market ‘failures’ of suburbanisation (Brueckner, 2000) or for preventing urban sprawl through active spatial policies (Ewing, 1997). For instance, policies aimed at increasing private transport costs, in combination with intensification of urban densities and multifunctional land-use in cities, are seen as effective measures to restrain urban sprawl (Dieleman and Wegener, 2004). Integrated planning of built-up development and transport infrastructures, and brownfield redevelopment are considered effective measures to prevent urban sprawl (EEA/FOEN, 2016). However, environmental protection and downzoning may actually stimulate sprawl and land consumption because wealthy nature-lovers tend to build detached houses on larger plots ‘in nature’ (Rudel et al., 2011).

The role of governance and planning factors in affecting landscape change (Hersperger and Bürgi, 2009), and urban sprawl in particular (Christiansen and Loftsgarden, 2011; EEA/FOEN, 2016; Nuissl and Couch, 2007; Oueslati et al., 2015; Salvati and Carlucci, 2016; Salvati and Gargiulo Morelli, 2014), has increasingly been acknowledged. More specifically, Siedentop and Fina (2012: 2781) emphasise the need for comparative research on the relationship between institutional arrangements and urban sprawl to assess the performance of administrative systems concerning land containment. The authors argue that the distribution and strength of political authority and spatial planning competencies among governmental tiers can play an important role in regional variations of (sub)urban growth.

Suburbanisation is a social practice of land transformation and allocation, involving both spatial patterns and processes of urban land change, and occurring within specific spatial planning systems, governance scales and multi-actor dynamics. However, suburban land-use patterns cannot be directly attributed to spatial planning regulations and plans, or to their interrelated institutional and governance settings. Efforts are hence needed to clarify how the gap between governance and spatial planning processes and suburban land-use patterns can be bridged. Differences in how governance and spatial planning driving forces act at various territorial scales and in distinct contexts, which may show different suburbanisation patterns, should be a particular focus.

This article compares territorial transformations in the metropolitan regions of Barcelona and Milan over a period of approximately 20 years (1990–2012), and investigates how governance dynamics and spatial planning factors are potentially linked to the observed patterns of suburban development in these regions. CORINE land cover data (1990, 2000 and 2012), demographic census data (1991, 2001 and 2011) and administrative data (2011) have been combined with document analysis of spatial planning regulations and plans and with in-depth interviews about individual and collective practices of land-use management and allocation.

Although expansion trends following the post-Second World War industrial boom have been occurring since the 1950s (Kasanko et al., 2006), the 1990s–2010s timeframe is particularly relevant because the 1990s have been identified as the beginning of the ‘sprawling era’ in Mediterranean (and more broadly Southern European) countries (Salvati and Gargiulo Morelli, 2014). Furthermore, significant differences are recognisable in the recent (sub)urbanisation patterns of Southern European countries (Salvati and Gargiulo Morelli, 2014); hence, a comparative analysis of two case studies in Southern Europe elucidates the connection between observed suburban patterns and governance and planning factors.

The findings reveal that local authorities in Barcelona and Milan adopt similar land management practices of urban development, constituting a sort of ‘baseline’. However, the composition of land uses in the two urban regions is different: while in Barcelona the proportion of suburban areas in relation to the total built-up forms is lower than 50% at all territorial scales, in Milan the values are over 50%. Additionally, the larger the territorial scale (from municipal boundary, to metropolitan area, to province), the lower the overall proportion of urbanised areas to other land uses (agricultural areas, forests and other land types) in the Barcelona case. In contrast, in the Milan case, the proportion of built-up areas decreases comparatively less as the territorial scale increases and is almost three times greater in the Milan province than in the Barcelona province.

The findings of this study suggest that the different suburban patterns observed in the two urban regions at different territorial scales can be explained by different institutional frames of spatial planning (Servillo and Van den Broeck, 2012) that have changed over time in each case. These are shaped by distinct governance processes and spatial planning regulations and practices mainly performed at the regional level. Over the last 40 years, the Catalonian regional government has played a more influential role in territorial planning than the Lombardy regional government, and has more successfully restrained the emergence of sprawl.

The important contribution of this research is the establishment of a link between governance and spatial planning factors and territorial dispersion patterns of housing areas, using the concept of institutional frames of spatial planning systems. In addition, it clarifies mechanisms and routine planning practices at different governance scales that are relevant for the occurrence of urban sprawl. This study shows that a stronger governance and planning role played by the regional government can be effective for urban sprawl containment.

The remainder of the article is organised as follows. The theoretical framework is introduced, followed by a description of the empirical methodology. The main results are then presented: the quantitative analysis is shown and then the qualitative investigation is reported for both case studies. The final section combines the discussion and the main conclusions of the study.

Governance actors and institutional frames of spatial planning systems

One of the central tensions characterising contemporary suburbanisation processes is the misalignment between political institutions and the growth of decentralised development across administrative borders within large metropolitan regions, whose boundaries have become difficult to clearly define (Ekers et al., 2012; Phelps and Wood, 2011).

The term ‘post-suburban politics’ coined by Phelps and Wood (2011) accounts for the dissolution of the conventional boundaries between urban, suburban and regional politics given the interconnected and metropolitan character of sprawling built-forms. Post-suburban politics embraces multi-scalar interests and non-local governmental relations, as well as new spatial scales. The analysis of ‘the various agents involved in the process of [post-sub]urban development’ is crucial because, ‘despite the different actors with seemingly different and potentially conflicting interests, there is in reality an accommodation among actors in planning for sprawl’ (Phelps et al., 2006: 200).

The concept of ‘suburban governance’ emphasises ‘the varied agents, methods, relations and institutions through which [suburban development] is managed’ (Ekers et al., 2012: 409). State-driven, capital-driven and private authoritarian-driven sprawl have been identified as the three different but complementary ideal-typical modalities through which suburban processes are governed (Ekers et al., 2012), as suburbanisation has an ‘explicit political character’ (Hamel and Keil, 2015: 21). Governance and politics, however, make up only part of the picture. Spatial planning traditions, regulations and practices should also be a research focus because urban sprawl implies a process of land-use transformation involving actors, institutions and governance dynamics related to spatial planning.

Beyond the technical understanding of spatial planning as sets of rules, directives and building standards to steer urban growth, protect conservation areas and balance public interests with land ownership rights, spatial planning systems are primarily ‘embedded in complex socio-political configurations’ (Servillo and Van den Broeck, 2012: 42). This means that spatial planning systems are associated with specific institutional settings formed by, and reciprocally influencing, the dynamics and practices of certain actors involved in the management and allocation of land (e.g. regional authorities, directors of public works, developers’ associations). These interrelated institutions compose institutional frames of spatial planning systems: technical aspects of spatial planning (e.g. regulations and plans’ objectives) are merged with socio-political dimensions (e.g.specific actors and groups’ practices and their governance dynamics, distribution of planning competencies) at different territorial scales, and are characterised by specific planning traditions and cultures.

In this study, understanding the role of governance and spatial planning factors in influencing land-use patterns of suburbanisation requires examining the institutional frames of spatial planning characterising a specific territorial context at a certain moment. The argument is that different institutional frames are associated with different degrees of suburban land-use patterns. This implies analysing which institutional actors at different governance scales hold, over time, different degrees of planning authority and competencies, how they are involved in the issuing and application of planning regulations and plans and what role they play in multi-scalar governance arrangements related to practices of land-use transformations. The explanatory power of such an analysis can be multiplied if performed comparatively: the different degree of effectiveness of spatial planning as a technical tool for the containment of urban sprawl in different contexts is arguably related to different institutional frames of spatial planning systems.

Methods

Case selection

An increasing number of studies have emphasised the specificities of urban growth and sprawl in Mediterranean countries (Salvati and Carlucci, 2016; Salvati and Gargiulo Morelli, 2014; Salvati et al. 2013). Barcelona, located in Catalonia (northeast Spain), and Milan, located in Lombardy (northern Italy) were chosen for comparison because they present different suburban territorial patterns despite sharing a number of historical and institutional similarities. The comparison between the two urban regions was deemed appropriate to identify governance and spatial planning aspects that could partially explain their diverging suburbanisation patterns.

Both urban regions underwent similar processes of urban growth in the 19th century and have subsequently experienced massive national (since the 1950s) and international (since the 1990s) migration flows that have sustained intense metropolitanisation processes. The post-industrial character of both regions has developed since the 1980s, and they are currently the largest non-capital cities in their countries (García, 2003; Gualini, 2003).

Furthermore, Barcelona and Milan have similar governmental systems. Although Spain has been structured as a type of federal state since the end of the 1970s, and although Italy was a centralised state until the recent 2001 constitutional reform, both countries maintain similar governmental scaffolding. In Italy and Spain the terms ‘city’ (urban level), ‘province’ (provincia or diputación), ‘metropolitan level’ (città or area metropolitana, or área or región metropolitana) and ‘regional authority’ (regione or comunidad autónoma) are employed to refer to the common Napoleonic structure (Newman and Thornley, 1996), and are categorised in the same ‘urbanism’ type of planning tradition (see ESPON, 2007). Similarities in administrative structures are important for minimising both substantial and terminological disagreement when comparing case studies. Case studies were limited to the Barcelona and Milan metropolitan regions to allow an examination of governance dynamics and spatial planning practices and regulations through in-depth interviews and document analysis.

Measurement of urban sprawl

In this study, urban sprawl was operationalised as a land-use type by using the European longitudinal CORINE Land Cover (CLC) datasets, following other comparative studies (Oueslati et al., 2015; Salvati and Gargiulo Morelli, 2014; Salvati et al., 2013). CORINE data use a specific nomenclature composed of 44 land cover classes subdivided into five broad categories (artificial surfaces, agricultural areas, forests and semi-natural areas, wetlands and water bodies). Of the 11 land-use classes that constitute the artificial surfaces, the ‘discontinuous urban fabric’ class (coded 1.1.2) was considered a suitable proxy for quantifying housing urban sprawl in the two case studies. This class predominantly captures residential land uses and the term ‘discontinuous’ refers to the intermediate intensity of impervious surface (30%–80%) compared to that of the ‘continuous’ land-use class (i.e. compact urban fabric; > 80% sealed soil) or agricultural, sport and recreation areas (< 30%). Land-use transformations were calculated as changes between 1990 and 2012.¹ Although urban sprawl is an umbrella term that refers to various morphologies (Keil, 2013; Walks, 2013), this study focuses on scattered housing areas because they take up most of the built-up forms in Europe and are hence the characterising pattern of the European urban environment (EEA, 2013).

Land-use data are presented at the urban, metropolitan and provincial scale to account for the different composition of land distribution across the different scales and between the two case studies. CORINE data are presented for Barcelona and Milan administrative, metropolitan and provincial boundaries (NUTS-3).² For the Barcelona case, both the Área metropolitana (BMA, 36 municipalities) and the Región metropolitana (BMR, 164 municipalities) were included, while for the Milan case the municipalities affiliated in 2016 with the Centro Studi Programmazione Intercomunale Area Metropolitana Milanese (PIM), a voluntary metropolitan association of local authorities, were considered.

Qualitative data

A total of 30 (16 in Barcelona and 14 in Milan) in-depth interviews with planners, politicians, stakeholders and experts at different governance scales (i.e. urban, provincial, metropolitan, regional) were carried out in both cities during 2012 and 2014. Interviewees were selected through a snowball technique. The average duration of the interviews was 65 minutes. Interviews were fully transcribed by the author and manually coded.

The intention of the interviews was to investigate actors’ decisions and routine practices of land-use allocation and management, and their possible influence on the contextual emergence of suburban residential areas. Interviewees were first asked how they would describe the process of (sub)urban growth that had occurred in their urban regions. The aim of subsequent questions was to identify particular stages and moments (e.g. issuing of national laws, local regulations, political scenarios), institutional actors (e.g. builders, land owners, local and regional authorities) and driving factors (e.g. housing preferences, private mobility) that were considered key in consolidating or shifting actors’ practices on spatial planning and land management, and in changing or maintaining current trends of (sub)urban growth.

The aim of the document analysis of planning regulations in Italy and Spain, with a particular focus on Milan and Lombardy, and Barcelona and Catalonia, was to test the evidence gathered through the interviews, and to highlight the main spatial planning tools at different scales related to land-use management. This analysis was linked with the rich qualitative interview data to identify the spatial planning institutional frames assumed to be related to the different degrees of suburbanisation patterns observed in the two case studies.

Demographic and administrative data

Demographic trends are typically related to urban decentralisation processes (Couch et al., 2007; EEA/FOEN, 2016; Kasanko et al., 2006; Siedentop and Fina, 2010), hence the analysis was complemented with census demographic data (1991, 2001, 2011; IDESCAT, 2011; ISTAT, 2011) that cover a time horizon similar to CORINE data on land-use transformations, and are presented as variations.

In addition, administrative data on both provinces were employed to test the connection between the occurrence of urban sprawl and administrative fragmentation (i.e. number of municipalities; see Nuissl and Couch, 2007: 230). The number of municipalities for both Barcelona and Milan was considered for 2011 because it did not substantially change during the analysed period in either case study.

Analysis and findings

Suburbanisation patterns in the Barcelona and Milan metropolitan regions

Within the cities’ administrative boundaries (Tables 1a and 2a), Barcelona has a lower proportion of discontinuous residential built-up forms (21.2% in 2012) than Milan (52.8%). Similarly, the metropolitan area of Barcelona (Table 1b) has a smaller proportion of discontinuous urban fabric (31.3% in 2012) than Milan (56.8%), where continuous (compact) urban fabric takes up only 9.4% (Table 2b). Although discontinuous urban fabric takes up almost 50% of the area in the Barcelona metropolitan region (Table 1c) and province (Table 1d), the Milan province (Table 2c) can be considered more dispersed because urban sprawl occupies 63.3% of built-up forms.

Table 1.

CORINE Land Cover classes (1990, 2000, 2012) at administrative, metropolitan and provincial scales for the Barcelona case (values in hectares and percentages).

(a) Barcelona administrative boundaries
CLC classes	1990	%	2000	%	2012	%	% 1990–2012
Continuous urban fabric	3637	47.0	3641	46.3	3641	45.9	0.1
Discontinuous urban fabric	1649	21.3	1681	21.4	1681	21.2	1.9
Industrial and commercial areas	1201	15.5	1247	15.9	1265	15.9	5.3
Road and railways	160	2.1	180	2.3	185	2.3	15.6
Airports and ports	332	4.3	365	4.6	377	4.7	13.6
Green and sport areas	683	8.8	742	9.4	791	10.0	15.8
Other uses	72	1.0	11	0.1	0	0.0	−100.0
Total urbanised area	7734	100.0	7867	100.0	7940	100.0
		77.5		78.9		79.6	2.7
Agricultural areas	215	2.2	104	1.0	96	1.0	−55.3
Forests	1933	19.4	1915	19.2	1909	19.1	−1.2
Other open areas	91	0.9	87	0.9	27	0.3	−70.3
Total surface	9973	100.0	9973	100.0	9973	100.0
(b) Barcelona Metropolitan Area (BMA)
CLC classes	1990	%	2000	%	2012	%	% 1990–2012
Continuous urban fabric	7237	28.0	7335	26.4	7376	25.1	1.9
Discontinuous urban fabric	8437	32.6	8935	32.2	9197	31.3	9.0
Industrial and commercial areas	5589	21.6	6508	23.4	7192	24.5	28.7
Road and railways	832	3.2	950	3.4	1129	3.9	35.7
Airports and ports	804	3.1	911	3.3	1245	4.2	54.9
Green and sport areas	2218	8.6	2419	8.7	2526	8.6	13.9
Other uses	747	2.9	731	2.6	719	2.4	−3.7
Total urbanised area	25,864	100.0	27,789	100.0	29,384	100.0
		40.7		43.7		46.3	15.5
Agricultural areas	10,087	15.9	8559	13.5	7388	11.6	−26.8
Forests	26,609	41.9	26,209	41.3	25,973	40.9	−2.4
Other open areas	960	1.5	963	1.5	775	1.2	−19.1
Total surface	63,520	100.0	63,520	100.0	63,520	100.0
(c) Barcelona Metropolitan Region (BMR)
CLC classes	1990	%	2000	%	2012	%	% 1990–2012
Continuous urban fabric	12,859	20.0	13,098	18.9	13,294	18.1	3.4
Discontinuous urban fabric	30,826	48.0	32,302	46.6	33,352	45.4	8.2
Industrial and commercial areas	12,616	19.6	14,838	21.4	16,646	22.7	31.9
Road and railways	1432	2.2	1576	2.3	1862	2.5	30.0
Airports and ports	858	1.3	975	1.4	1341	1.9	56.3
Green and sport areas	3539	5.5	4070	5.9	4579	6.2	29.4
Other uses	2159	3.4	2470	3.5	2377	3.2	10.1
Total urbanised area	64,289	100.0	69,329	100.0	73,451	100.0
		19.8		21.4		22.6	14.3
Agricultural areas	80,842	24.9	77,015	23.7	73,725	22.7	−8.8
Forests	178,227	54.9	177,033	54.6	176,386	54.4	−1.0
Other open areas	1094	0.4	1075	0.3	890	0.3	−18.6
Total surface	324,452	100.0	324,452	100.0	324,452	100.0
(d) Barcelona NUTS-3
CLC classes	1990	%	2000	%	2012	%	% 1990–2012
Continuous urban fabric	14,485	18.7	14,728	17.7	14,955	16.9	3.2
Discontinuous urban fabric	39,217	50.6	40,880	49.1	42,254	47.9	7.7
Industrial and commercial areas	14,424	18.6	16,982	20.4	19,203	21.8	33.1
Road and railways	1577	2.0	1722	2.1	2070	2.3	31.3
Airports and ports	920	1.2	1041	1.3	1431	1.6	55.5
Green and sport areas	3786	4.9	4422	5.3	5000	5.7	32.1
Other uses	3035	4.0	3430	4.1	3373	3.8	11.1
Total urbanised area	77,444	100.0	83,205	100.0	88,286	100.0
		10.0		10.7		11.4	14.0
Agricultural areas	197,510	25.5	195,766	25.2	192,059	24.8	−2.8
Forests	498,521	64.2	494,535	63.8	493,359	63.6	−1.0
Other open areas	2260	0.3	2229	0.3	2031	0.2	−10.1
Total surface	775,735	100.0	775,735	100.0	775,735	100.0

Source: Copernicus Land Monitoring Services (2016).

Table 2.

CORINE Land Cover classes (1990, 2000, 2012) at administrative, metropolitan and provincial scales for the Milan case (values in hectares and percentages).

(a) Milan administrative boundaries
CLC classes	1990	%	2000	%	2012	%	% 1990–2012
Continuous urban fabric	3119	23.3	3119	23.1	3145	22.8	0.8
Discontinuous urban fabric	7226	54.0	7272	53.8	7276	52.8	0.7
Industrial and commercial areas	1259	9.4	1328	9.8	1580	11.5	25.5
Road and railways	442	3.3	442	3.3	535	3.9	21.0
Airports and ports	58	0.4	58	0.4	58	0.4	0.0
Green and sport areas	1148	8.6	1110	8.2	1161	8.4	1.1
Other uses	123	1.0	194	1.4	29	0.2	−76.4
Total urbanised area	13,375	100.0	13,523	100.0	13,784	100.0
		73.6		74.4		75.9	3.1
Agricultural areas	4671	25.7	4523	24.9	4263	23.5	−8.7
Forests	40	0.2	40	0.2	40	0.2	0.0
Other open areas	81	0.5	81	0.5	80	0.4	0.0
Total surface	18,167	100.0	18,167	100.0	18,167	100.0
(b) Milan metropolitan area (PIM)
CLC classes	1990	%	2000	%	2012	%	% 1990–2012
Continuous urban fabric	4003	10.1	4003	9.8	4029	9.4	0.6
Discontinuous urban fabric	23,011	58.1	23,400	57.5	24,282	56.8	5.5
Industrial and commercial areas	8445	21.3	8809	21.7	10,066	23.5	19.2
Road and railways	794	2.0	794	2.0	947	2.2	19.3
Airports and ports	515	1.3	515	1.3	515	1.2	0.0
Green and sport areas	2376	6.0	2648	6.5	2670	6.2	12.4
Other uses	461	1.2	494	1.2	260	0.7	−43.6
Total urbanised area	39,605	100.0	40,663	100.0	42,769	100.0
		42.2		43.3		45.5	8.0
Agricultural areas	51,663	55.0	50,368	53.6	48,407	51.6	−6.3
Forests	2135	2.3	2351	2.5	2220	2.4	4.0
Other open areas	495	0.5	516	0.6	502	0.5	0.8
Total surface	93,898	100.0	93,898	100.0	93,898	100.0
(c) Milan NUTS-3
CLC classes	1990	%	2000	%	2012	%	% 1990–2012
Continuous urban fabric	4651	6.7	4651	6.5	4678	6.2	0.6
Discontinuous urban fabric	44,942	65.1	45,821	64.4	47,986	63.3	6.8
Industrial and commercial areas	14,030	20.3	14,860	20.9	17,158	22.6	22.3
Road and railways	800	1.2	800	1.1	1043	1.4	30.4
Airports and ports	599	0.9	599	0.8	599	0.8	0.0
Green and sport areas	2914	4.2	3413	4.8	3656	4.8	25.5
Other uses	1082	1.6	1048	1.5	634	0.9	−41.4
Total urbanised area	69,018	100.0	71,192	100.0	75,754	100.0
		34.9		36.0		38.3	9.8
Agricultural areas	119,128	60.2	116,561	58.9	112,912	57.1	−5.2
Forests	8873	4.5	9295	4.7	8450	4.3	−4.8
Other open areas	863	0.4	834	0.4	766	0.3	−11.5
Total surface	197,882	100.0	197,882	100.0	197,882	100.0

Source: Copernicus Land Monitoring Services (2016).

In addition, the overall percentage of urbanised areas compared to areas with other land uses (agricultural areas, forests and other land types) is more contained in the Barcelona case. Although the proportion of built-up forms to open and agricultural land is similar for the Barcelona and Milan administrative boundaries (Tables 1a and 2a) and for the Barcelona and Milan metropolitan areas (Tables 1b and 2b), substantial differences can be found in the larger Barcelona metropolitan region and province (Tables 1c and 1d). In the Milan province (Table 2c), the percentage of built-up forms is over three times greater than in the Barcelona case.³

Demographic trends

Descriptive demographic data support the less dispersed character of Barcelona in comparison to Milan (Table 3a). In the case of Barcelona, while variations in demographic data show negative trends, the majority of inhabitants live in the Barcelona Metropolitan Region (63.5%) and province (73.4%).⁴ Other studies corroborate the increase in the demographic weight of the second metropolitan ring, which is mostly related to considerable population growth in smaller municipalities (López et al., 2011; see Salvati and Carlucci, 2016). In contrast, Milan shows a limited combined capacity of the metropolitan area and of the provinces to concentrate population (26.2% and 40.0% in 2011; Table 3b), as shown in earlier studies (Balducci, 2003; Caiello and Colleoni, 2013).

Table 3.

Total number of inhabitants, population change and demographic percentage (in italics) for Barcelona and Milan administrative, metropolitan and provincial scales, and for Catalonia and Lombardy (1991, 2001, 2011).

(a) Barcelona
Territorial scales	1991	2001	2011	% 1991–2011
Barcelona municipality	1,643,542	1,503,884	1,611,013	−2.0
Demographic percentage of Catalonia	27.1	23.7	21.4	−21.0
Barcelona Metropolitan Area (BMA)	3,048,479	2,936,563	3,220,476	5.6
Demographic percentage of Catalonia	50.3	46.3	42.8	−14.9
Barcelona Metropolitan Region (BMR)	4,264,422	4,390,390	4,772,130	11.9
Demographic percentage of Catalonia	70.4	69.2	63.5	−9.8
Barcelona province	4,654,407	4,805,927	5,522,566	18.7
Demographic percentage of Catalonia	76.8	75.8	73.4	−4.4
Total Catalonia	6,059,494	6,343,110	7,519,842	24.1
(b) Milan
Territorial scales	1991	2001	2011	% 1991–2011
Milan municipality	1,369,231	1,256,211	1,242,123	−9.3
Monza municipality	120,651	120,204	119,856	−0.7
Demographic percentage of Lombardy	16.8	15.2	14.0	−16.5
Milan metropolitan area (PIM)	2,568,306	2,472,527	2,539,567	−1.1
Demographic percentage of Lombardy	29.0	27.4	26.2	−9.7
Milan province	3,006,632	2,938,557	3,038,420	1.1
Monza province	729,347	766,631	840,129	15.2
Demographic percentage of Lombardy	42.2	41.0	40.0	−5.2
Lombardy	8,860,596	9,038,118	9,704,151	9.5

Source: Census data IDESCAT (2011), ISTAT (2011).

Considering the previously shown tables on land use, almost 65% of the population in Catalonia is located in the 22.6% of the territory in the Barcelona metropolitan region, and nearly 75% of inhabitants occupy 11.4% of the territory in the Barcelona province. In contrast, 40.0% of the population in Lombardy occupies 38.3% of the entire territory of the Milan and Monza provinces (Milan NUTS-3).

Municipal fragmentation

Although a desire to locate outside the inner city centre can partially explain suburbanisation processes (Reckien and Luedeke, 2014), municipal fragmentation appears to be related to urban sprawl because independent, medium- and small-sized local authorities try to attract investment, for instance in the form of suburban houses or decentralised industrial areas, to the detriment of other municipalities (Siedentop and Fina, 2012). Such interurban fiscal competition potentially stimulates (sub)urban growth as a way to harvest resources (Couch et al., 2007: 18; Herce, 2005; Nuissl and Couch, 2007: 225).

In both case studies, competencies over urban planning remain strongly tied to local authorities. In a context of decreasing state grants (García, 2003), several interviewees highlighted the crucial role of local taxation in fostering urban sprawl. An interviewee working at the Àrea Metropolitana de Barcelona (AMB) stated that:

Planning fees are the best way for local governments to ensure revenues, together with the state grants to municipalities for providing services according to the number of inhabitants (…) [all this] allows municipalities to keep afloat.

Similarly, in the Milan case interviewees emphasised that, following a common practice, municipalities can strategically convert land (at the beginning only on paper) to put municipal balance sheets on an even keel (cash inflows). The ‘complex issue of support for local economic development’ is then dealt with ‘at the local level, by the simple allocation of land for use’ (Balducci, 2003:68).

Compared with other European countries, where the estimated median of Local Administrative Units (LAUs; i.e. municipalities) per NUTS-3 area is 36, the average being 178 (Eurostat, 2015),⁵ both the Barcelona and Milan NUTS-3 areas (provinces) have a rather large number of municipalities (Table 4).

Table 4.

Municipal sizes in the provinces of Barcelona and Milan (and Monza) (2011).

	Barcelona			Milan
Municipal size(no. of inhabitants)	Freq.	%	Cum	Freq.	%	Cum.
< 5000	179	57.6	57.6	48	25.4	25.4
< 25,000	94	30.2	87.8	112	59.3	84.7
< 50,000	19	6.1	93.9	23	12.2	96.8
< 100,000	12	3.9	97.8	4	2.1	98.9
< 250,000	5	1.6	99.4	1	0.5	99.5
< 500,000	1	0.3	99.7	0	0.0	0.0
> 500,000	1	0.3	100.0	1	0.5	100.0
Total	311	100.0		189	100.0

Source: IDESCAT (2011); ISTAT (2011).

Looking at the number of municipalities, Barcelona appears to be more fragmented than Milan. However, the province of Barcelona is almost four times larger than the Milan and Monza provinces together (see Tables 1d and 2c). Indeed, the average municipal surface in the province of Barcelona is 24.9 km², compared to 10.5 km² in the Milan and Monza provinces. Based on the two considered case studies, this finding indicates that the more municipal fragmentation is limited, the more discontinuous urban areas are reduced.

Institutional frames of suburbanisation

In the Barcelona case, the 1956 law no. 12 (Ley de 12 de mayo de 1956 sobre régimen del suelo y ordenación urbana), promulgated during the Francoist regime and defining land-use regulations in Spain, set the first institutional frame of spatial planning. During the 1950s to 1970s, commonly known as the época desarrollista (a period of sustained planned economic growth), illegal land development through scattered and isolated low-density built-up forms was relatively common. These were mainly (second) residential houses built on agricultural land (urbanizaciones ilegales) with no public services whatsoever (e.g. roads, water or sewage systems, public lighting or electricity). These services had to be provided afterwards, especially during the 1980s, implying a continuous drain on public expenses. The interviewee at the Área Metropolitana de Barcelona (AMB) reported that, during the época desarrollista, ‘there was no control on the part of higher institutions’ with regard to land management. Another interviewee, who is a professor at the Barcelona Autònoma University, explained that illegal sprawled areas were:

Not like the Italian phenomenon of abusivismo, where someone illegally builds his detached house, or villetta, on his cousin’s land plot. Rather, urbanizaciones ilegales emerged when there was a builder or a real estate developer that bought a land plot, or found an agreement with the owner of the land plot, who was generally a farmer, and illegally subdivided the land, and then sold these land plots without any basic services, and with the promise that the land subdivision would be legalised and that services would be provided.

These suburban developments can only partly be considered illegal, as they were put into practice under the informal authorisation of municipalities, which promoted the idea of a ‘detached house with a garden’ (la caseta i l’hortet, in Catalan; see also Herce, 2005: 45).

In 1975, a new land-use law was promulgated (Legislación del suelo) that rectified the shortcomings of the 1956 law. The suelo rústico was defined as ‘non developable land’, and its subdivision into land plots was prohibited. The 1975 law also indicated (art. 9) that urban extensions should be built in morphological continuity with the already existing settlements, thereby avoiding the uncontrolled spread of residential areas in isolated morphological structures, although low densities were still possible. The 1975 national law constituted a tight normative corpus on land-use regulations.

In 1978, at the onset of the democratic era in Spain, devolution of authority in territorial spatial planning to regional governments (Comunidades Autónomas) was carried out. Such devolution represented the beginning of a second institutional frame.

Law no. 23 on territorial planning and law no. 7 on territorial organisation, approved by the Catalonian Parliament in 1983 and 1987, respectively, were key moments in strengthening the role of the Catalonian regional government (Generalitat) regarding territorial planning. The approval of both laws was also favoured by political continuity and stability, as the 23-year-long centre-right presidency of Jordi Pujol as president of the Catalonian regional government began in the early 1980s.

The progressive accumulation of political power and authority on territorial spatial planning by the Generalitat is exemplified by the abolishment of the Barcelona metropolitan authority in 1987. This event exacerbated the political tensions between the right-wing Catalonian government (led by Pujol’s party Convergència i Unió, CiU) and the left-wing Catalonian Socialists, who traditionally governed Barcelona municipality and the majority of the local authorities in the Barcelona metropolitan area (García, 2003: 355).

With the 1987 law, 41 comarques were established under the administrative level of the four provinces composing Catalonia (Barcelona, Girona, Lleida and Tarragona). Comarques still remain administrative delimitations without real powers, despite being the local ‘tiles’ upon which the spatial strategy envisioned by the Generalitat is based. The territorial plan for Catalonia (Pla General Territorial de Catalunya, PGTC), approved in 1995 more than a decade after the 1983 law, refers to comarques as the fundamental geographical and administrative arrangement of Catalonia.

Although they were envisioned with the 1995 plan, it was only between 2006 and 2010 that regional plans for the six territorial àmbits⁶ (‘areas’), as sets of comarques, were issued, including the 2010 plan for the metropolitan region of Barcelona (BMR). These plans are spatially explicit and define where and how urban growth should occur, thus binding local plans to regional directives. The approval of the six territorial plans was related to a political change in the Catalonian government following the 2003 elections, indicating the onset of a third institutional frame. A left-wing coalition (called Tripartit) took over leadership in the Catalonian government from the Pujol’s centre-right-wing party and strongly supported a revision of the Catalonian territorial spatial policy (Nel·lo, 2011).

These three institutional frames highlight how the Catalonian regional government concentrated substantial political power and authority regarding spatial planning policies. A rather lengthy, but significant, extract from an interview with one of the reference persons on planning in Catalonia, who had extensive experience in plan making and political responsibilities, clarifies the strengthened governance and political role played by the Generalitat regarding spatial policies:

The local government proposes the urban plan, discussing it with its inhabitants, and then it delivers it to the regional authority, which checks it in certain commissions, where there are competent people, who are anyway connected with the political basis of the regional government, which are the municipalities. Hence, the urban plans are discussed in such commissions, however the last word is for the director of the territorial planning department [director general de ordenación del territorio y urbanismo] at the Generalitat. The commission can make some comments, and the director can advise the city officials that local plans should conform to certain requirements. But it is obvious that the urban plan will be approved. However, it is the director’s final choice to approve the plan. Hence, the director of the territorial planning department muddles through a very complex situation, because politically a local government can have quite an influence in the party, and in the party that is ruling the Generalitat. Therefore, the decisions that the director wants to make can be short-circuited. Obviously, the territorial planning director at the Generalitat receives certain instructions (…) in order to accommodate municipalities’ expectations on urban plans and projects. At the end of the day, this is quite normal. (…) That is why being the Generalitat’s planning director is the most difficult political post in Catalonia.

Regarding the Milan case, the 1942 national law no. 1150 on planning (Legge urbanistica statale), modified in the subsequent decades, is still valid in Italy. Approved during Fascist rule and the Second World War, it anticipated the needs for urban development during the reconstruction period after the war. The 1942 law can be considered a spatial planning tool that was advanced for its time (De Lucia, 2006; Settis, 2010). However, the ‘urgency’ for reconstruction postponed its implementation, and spatial planning regulations were considered ‘useless barriers’ for a certain type of urban development and growth (De Lucia, 2006). For instance, in Milan the rito ambrosiano was applicable especially during the 1950s and 1960s, meaning that rural or protected land was unofficially urbanised for the provision of public services (De Lucia, 2006: 48; Settis, 2010: 273). One of the most striking results of a legislation prioritising the interests of local governments for land allocation, and of a diffuse tendency towards unregulated building activities, has been the three national building amnesties (condono edilizio) in 1985, 1995 and 2003.⁷

The institutional frame set up by the 1942 planning law did not change substantially when, in 1970, regional governments (Regioni) were formally constituted, despite the fact that the 1948 Italian Constitution had already envisioned them. The delay in their establishment resulted in the absence of supervisory bodies for land allocation strategies performed by local authorities, creating an institutional vacuum (Gualini, 2003: 268).

While local authorities were entitled to exclusive competencies in urban planning, only competencies in landscape and environmental protection were delegated by the state to the new regional governments. The Regioni received all local plans issued by local authorities and enforced, with the cooperation of provincial authorities, their compliance to national and regional environment and landscape protection laws, performing a supervisory role (enti di controllo). However, planning power, held by municipalities, belongs to the Ministry of Public Works according to the 1942 national law, which has always been much stronger than the regulations on landscape protection delegated to the Regioni and linked to the Ministry of Education (Settis, 2010: 248).

The 2001 constitutional reform marked the beginning of a new institutional frame, when planning competencies were devoluted to regional governments. The Regioni were converted into governing authorities (enti di governo), and could be endowed with proper territorial plans. In 2005, the Lombardy regional government (Regione Lombardia) promulgated law no. 12 on territorial spatial planning. Subsequently, the regional territorial plan (Piano territoriale di coordinamento regionale, PTCR) for Lombardy was approved in 2010, and the subsidiary territorial provincial plan (Piano territoriale di coordinamento provinciale, PTCP) for the Milan province was issued in 2012. These plans have a strategic character and orient territorial development trajectories of municipalities in accordance with national and regional environment and landscape protection laws.

However, there is still a diffuse sentiment that consolidated local practices of land management are not easy to overcome. One of the interviewees at the Lombardy regional government stated that:

‘Planning power is still in the hands of local governments. With instructions, but just instructions [in environmental and landscape protection], coming from the territorial regional and provincial plans (…) we try to limit the power of municipalities’. The regional and provincial administrative tiers function as ‘buffers’ that limit municipalities’.

Growth expectations through the political work done by its functionaries, and reduce the impacts of amendments to local plans (varianti di piano) on urban growth. One of the interviewed provincial officers stated that:

It is really a political role, to find a balance and a dialogue with the mayors. (…) When necessary, I call the mayors and say that the plan does not comply with the requirements of the [provincial or regional] plan, which is an important political message. Then they lower their expectations. But not every municipality does it. (…) An agreement has to be found. (…) We reason together that some areas cannot accommodate development, especially if it implies urban sprawl. (…) It’s a political job basically.

This finding is also confirmed by previous studies emphasising that higher-level authorities have attempted to coordinate and regulate local development through ‘informal’ supra-municipal efforts, meaning that the ‘greater results were obtained by informal [political] activities considered secondary’ (Balducci et al., 2011: 4).

Interviewees emphasised that land management strategies at the local level imply an instrumental use of land, where suburban residential areas are particularly functional in attracting private investments and residents. One of the interviewed stakeholders has particularly clarified the lobbying practices of collective stakeholders with city officials:

We [builders] are some of those actors who occupy, work and transform land. (…) For us, land is the raw material. (…) We transform land according to the laws and regulations defined by public authorities. (…) However, each local authority has its own rules [i.e. the local urban plan]. (….) What happens is that the mayor comes here [to the Building Constructors Association] – well, actually we go there – and says: gentlemen, I [the mayor] want to provide houses for all. (…) As land is a municipal property, I will give it to you for free, so you don’t have this cost. (…) So we make a deal, a real written contract. (…) Most of the times, builders are also the landowners. (…) So when a new mayor comes and reclassifies our plot into agricultural land, well, this makes us angry, because we have paid taxes on it because it was developable before.

Finally, similar to in the Barcelona case, the Milan metropolitan authority (Piano intercomunale Milanese, PIM), which had a pivotal role in coordinating often oversized local plans during the 1970s, saw its governance and political influence significantly reduced by the 1981 no. 23 Lombardy regional law: the PIM was abolished as an administrative tier and converted into a public consultancy and research centre to support its affiliated municipalities.

Conclusions: Linking processes and patterns

The aim of this study was to clarify the link between suburban land-change patterns and governance processes and spatial planning regulations and practices. The article has shown how different actors, interrelated in specific governance and political dynamics within certain institutional frames of spatial planning, have influenced the territorial expansion patterns of dispersed housing areas. In a constant tension between urbanism and suburbanism (Walks, 2013), urban sprawl is a territorial materialisation of actors’ decisions and dynamics over land-use allocation and management.

By comparing two different case studies during a period of over 20 years, this research demonstrates how the metropolitan character of urban sprawl originates from local planning practices, mainly performed by municipal authorities, through land-use micro-transformations. In both Barcelona and Milan, local governments are entitled to similar responsibilities in urban planning and constitute a sort of ‘baseline’, suggesting that a similar ‘sprawl culture’ is present in both cases. Urban sprawl is built locally but has a metropolitan effect.

Nevertheless, the land-use data analysis shows that the Barcelona case exhibits less urban sprawl than the Milan case. Demographic data reinforce the greater urban compactness of the Barcelona case, where a larger proportion of the population is concentrated in a smaller amount of territory compared with in the Milan case. Furthermore, it has been discussed how municipal fragmentation is associated with the emergence of suburbanisation because of the tendency of local authorities to transform land to sustain their local fiscal tax base. The less severe fragmentation in the Barcelona case can be related to a less dispersed built-up form. However, the identification and comparison among evolving institutional frames of spatial planning demonstrate that it is the broader, institutional context where such localised decisions on land-use management are made that leads to a difference between the two regions.

The identified factor explaining the different occurrence of urban sprawl in Barcelona and Milan is the divergent degree of authority held, over time, by the two regional governments in managing territorial transformations. Both regional authorities engage in an intense political ‘labour’ with the municipalities to mitigate their growth expectations. However, in the Barcelona case the regional government has taken on more substantial responsibilities in territorial planning, exercising a greater authority on municipalities’ decisions, and these competencies were assumed on an earlier date. The Generalitat has issued laws and regulations on spatial planning since the 1980s, while the Regione Lombardia has been entitled to do so starting only in the 2000s. Additionally, in Catalonia local plans regarding spatial development are approved by the regional government, while in Lombardy they are monitored concerning their compliance with landscape and environmental (and not land-use) regulations.

This finding confirms those of previous studies emphasising the key role of regional governments in the containment of urban sprawl (Nuissl and Couch, 2007: 225; Siedentop and Fina, 2012). However, it also highlights the complexities characterising the emergence of institutional frames of spatial planning. The delayed approval of the 1995 territorial plan for Catalonia more than a decade after the 1983 law, and the approval of the regional plans during 2006–2010 under a newly elected left-wing coalition, signal the leniency of the Generalitat conservative government towards planning practices stimulating suburbanisation, which have actually spread more clearly since the 1990s (Salvati and Gargiulo Morelli, 2014). Hence, the gradual reinforcement of the Generalitat’s authority on territorial planning does not come without drawbacks. However, compared to the Milan case, the role of the planning director at the Catalonian regional government has been stronger for mediating with the growth expectations of local authorities. Further studies on Southern European countries should examine the consistencies and contrasts characterising changing institutional frames of spatial planning in influencing regional variations in (sub)urbanisation patterns. The role of higher-level institutions should be a particular focus.

This study suggests that substantial competencies on territorial spatial planning held by higher-level institutions, in particular regional governments, could tackle the urban growth expectations foreseen by local authorities in their urban plans, and hence be an effective counter-force against suburbanisation.

Footnotes

Acknowledgements

I am grateful to many people for their help, guidance, insightful comments and suggestions on earlier versions of the manuscript. I would especially like to thank Frank Moulaert, Serena Vicari, Mario Boffi, Salvatore La Mendola and Matteo Colleoni for their guidance during my doctoral studies at University of Milano-Bicocca (Italy) and at Katholieke Universiteit Leuven (Belgium), and Anna M Hersperger and Robert Pazur for their support during my post-doc at the Swiss Federal Research Institute WSL (Switzerland). I am additionally thankful to Charlotte Steinmeier (WSL) for her critical insights into Corine Land Cover data. I am also grateful to all the interviewees for their time and commitment in participating in this research. Finally, I would like to warmly thank the three anonymous reviewers for their constructive comments, and Melissa Dawes for her meticulous proof-reading. Any inaccuracies are my own.

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) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: A post doctoral grant between 2011 and 2013 was received from the Italian Ministry of Education to partially carry out this research.

Notes

References

Antrop

(2004) Landscape change and the urbanization process in Europe. Landscape and Urban Planning 67(1): 9–26.

Balducci

(2003) Policies, plans and projects governing the city-region of Milan. DISP 152: 59–70.

Balducci

Fedeli

Pasqui

(2011) Strategic Planning for Contemporary Urban Regions: City of Cities: A Project for Milan. Farnham: Ashgate.

Breheny

(1997) Urban compaction: Feasible and acceptable? Cities 14(4): 209–217.

Brueckner

(2000) Urban sprawl: Diagnosis and remedies. International Regional Science Review 23: 160–171.

Brueckner

Fansler

(1983) The economics of urban sprawl. The Review of Economics and Statistics 65(3): 479–482.

Caiello

Colleoni

(2013) Il peri-urbano e i suoi caratteri socio-territoriali. Una proposta analitica e empirica in Lombardia [Peri-urban patterns and socio-territorial features: An analysis in Lombardy]. Sociologia urbana e rurale 102: 97–115.

Cavailhès

Peeters

Sékeris

, et al. (2004) The periurban city: Why to live between the suburbs and the countryside. Regional Science and Urban Economics 34(6): 681–703.

Christiansen

Loftsgarden

(2011) Drivers behind urban sprawl in Europe. Institute of Transport Economics Report 1136. Norwegian Centre for Transport Research.

10.

Copernicus Land Monitoring Services (2016) CORINE land cover. Available at: http:://land.copernicus.eu/pan-european/CORINE-land-cover (accessed 20 August 2016).

11.

Couch

Karecha

(2006) Controlling urban sprawl: Some experiences from Liverpool. Cities 23(5): 353–363.

12.

Couch

Leontidou

Arnstberg

(2007) Definitions, theories and methods of comparative analysis. In: Couch

Leontidou

Petschel-Held

(eds) Urban Sprawl in Europe: Landscapes, Land-use Change & Policy. Oxford: Blackwell Publishing, pp. 3–38.

13.

De Lucia

(2006) Se questa è una città [If this is a city]. 3rd edn. Rome: Donzelli Editore.

14.

Dieleman

Wegener

(2004) Compact city and urban sprawl. Built Environment 30(4): 308–323.

15.

Ekers

Hamel

Keil

(2012) Governing suburbia: Modalities and mechanisms of suburban governance. Regional Studies 46(3): 405–422.

16.

ESPON (2007) Governance of territorial and urban policies from EU to local level. Project 2.3.2, final report, part II, 31 May. Luxembourg: ESPON Coordination Unit.

17.

European Environmental Agency (EEA) (2013) Land accounts data viewer 1990, 2000, 2006. Available at: http://www.eea.europa.eu/data-and-maps/data/data-viewers/land-accounts (accessed 26 January 2013).

18.

European Environment Agency/Federal Office for the Environment (EEA/FOEN) (2016) Urban sprawl in Europe. European Environment Agency Report 11. Office for Official Publications of the European Communities.

19.

Eurostat (2015) Local Administrative Units (LAU). Available at: http://ec.europa.eu/eurostat/web/nuts/local-administrative-units (accessed 10 January 2016).

20.

Ewing

(1997) Is Los Angeles-style sprawl desirable? Journal of the American Planning Association 63(1): 107–126.

21.

García

(2003) The case of Barcelona. In: Salet

Thornley

Kreukels

(eds) Metropolitan Governance and Spatial Planning: Comparative Case-Studies of European City-Regions. London: Spon Press, pp. 337–358.

22.

Glaeser

Kahn

(2004) Sprawl and urban growth. Handbook of Regional and Urban Economics 4: 2481–2527.

23.

Gualini

(2003) The region of Milan. In: Salet

Thornley

Kreukels

(eds) Metropolitan Governance and Spatial Planning: Comparative Case-Studies of European City-Regions. London: Spon Press, pp. 264–284.

24.

Hamel

Keil

(eds) (2015) Suburban Governance: A Global View. Toronto: University of Toronto Press.

25.

Herce

(2005) Urbanización, precios del suelo y modelo territorial: La evolución reciente del Área Metropolitana de Barcelona [Urban growth, land rent and territorial model: The recent evolution in the Barcelona Metropolitan Area]. Revista Eure, Santiago de Chile 31(93): 35–51.

26.

Hersperger

Bürgi

(2009) Going beyond landscape change description: Quantifying the importance of driving forces of landscape change in a central Europe case study. Land Use Policy 26(3): 640–648.

27.

IDESCAT Catalonian Institute of Statistics (2011) Population and housing censuses. Available at: http://www.idescat.cat/pub/?id=censph#FORM (accessed 9 June 2014).

28.

ISTAT Italian National Institute of Statistics (2011) Population and housing censuses. Available at: http://www.istat.it/it/censimento-popolazione (accessed 15 July 2014).

29.

Kasanko

Barredo

Lavalle

, et al. (2006) Are European cities becoming dispersed? A comparative analysis of 15 European urban areas. Landscape and Urban Planning 77(1): 111–130.

30.

Keil

(ed.) (2013) Suburban Constellations: Governance, Land and Infrastructures in the 21st Century. Berlin: Jovis Verlag.

31.

López

Pujadas

Bayona

(2011) Recent Internal Migration and Demographic Dynamics in the Urban Transformation Processes of the Metropolitan Area of Barcelona. Madrid: Spanish Ministry of Science and Innovation.

32.

Mazzocchi

Sali

Corsi

(2013) Land use conversion in metropolitan areas and the permanence of agriculture: Sensitivity Index of Agricultural Land (SIAL), a tool for territorial analysis. Land Use Policy 35: 155–162.

33.

Nel·lo

(2011) El planeamiento territorial en Cataluña [Territorial planning in Catalonia]. Cuadernos Geográficos 47: 131–167.

34.

Newman

Thornley

(1996) Urban Planning in Europe. London: Routledge.

35.

Nuissl

Couch

(2007) Lines of defence: Policies for the control of urban sprawl. In: Couch

Leontidou

Petschel-Held

(eds) Urban Sprawl in Europe: Landscapes, Land-Use Change & Policy. Oxford: Blackwell Publishing, pp. 217–241.

36.

Oueslati

Alvanides

Garrod

(2015) Determinants of urban sprawl in European cities. Urban Studies 52(9): 1594–1614.

37.

Phelps

Parsons

Ballas

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

38.

Phelps

Wood

(2011) The new post-suburban politics? Urban Studies 48(12): 2591–2610.

39.

Reckien

Luedeke

MKB

(2014) The social dynamics of suburbanization: Insights from a qualitative model. Environment and Planning A 46(4): 980–1000.

40.

Rudel

O’Neill

Gottlieb

, et al. (2011) From middle to upper class sprawl? Land use controls and changing patterns of real estate development in Northern New Jersey. Annals of the Association of American Geographers 101(3): 609–624.

41.

Salvati

Carlucci

(2016) The way towards land consumption: Soil sealing and polycentric development in Barcelona. Urban Studies 53(2): 418–440.

42.

Salvati

Gargiulo Morelli

(2014) Unveiling urban sprawl in the Mediterranean region: Towards a latent urban transformation? International Journal of Urban and Regional Research 38(6): 1935–1953.

43.

Salvati

Sateriano

Bajocco

(2013) To grow or to sprawl? Land cover relationships in a Mediterranean city region and implications for land use management. Cities 30: 113–121.

44.

Servillo

Van den Broeck

(2012) The social construction of planning systems: A strategic-relational institutionalist approach. Planning Practice & Research 27(1): 41–61.

45.

Settis

(2010) Paesaggio Costituzione Cemento [Landscape Constitution Concrete]. Turin: Einaudi.

46.

Siedentop

Fina

(2012) Who sprawls most? Exploring the patterns of urban growth across 26 European countries. Environment and Planning A 44(11): 2765–2784.

47.

Walks

(2013) Suburbanism as a way of life, Slight return. Urban Studies 50(8): 1471–1488.

48.

Weilenmann

Seidl

Schulz

(2017) The socio-economic determinants of urban sprawl between 1980 and 2010 in Switzerland. Landscape and Urban Planning 157: 468–482.