Differences in the location of urban museums and their impact on urban areas

Traditionally, museums spaces have tended to be located in the central areas of cities, thereby reinforcing existing dynamics and limiting their potential impact on more peripheral areas. The present study examines the locations chosen for museums that have been opened in three European cities: Barcelona, Paris and Turin, between the years 2000 and 2013. It shows how the locating of museums and galleries has not been homogeneous and how each potential location has its strengths and weaknesses. This is particularly evident when we consider such factors as the environmental impact of museums and how they influence the commercial offer, image and attractiveness of their host city. This paper identifies and examines the different types of impact that museum location strategies have on urban areas.


Introduction
In recent times, museums 1 have added a further set of new functions. As a result, some museums have become prime tourist attractions in their own right, while others have played leading roles in processes of urban renewal. In some cases, museums have led the transformation of the economic base of a particular urban area, while in others they have played a key role in promoting greater social cohesion. These changes have implied that some museum activities have assumed a relevance that reaches out beyond their own internal spaces.
The present and potential future impact of museums on their respective urban areas is a subject that has aroused interest in various sectors, both public administration and private sector. The economic, touristic and social successes of cities and their hinterlands have become increasingly based on their ability to create new attractions and that demonstrate their dynamism. As a result, one of the most popular strategies now used for modernising a city's image involves the creation of new museums and/ or cultural events (Navarrete 2008, de Graaff et al. 2009, Gibson 2013. This new function has turned museums into a leading element within urban policy. A number of cities have envisaged the need to create, or perhaps update, their cultural offer. The growing demand for cultural attractions has changed the traditional panorama. In response, a series of new museums have been added to the existing cultural offer. However, in general, this has not been the result of a reflective planning process or of reaching a consensus with the local community (Evans and Foord 2008). In fact, there has often been no real debate as to the purported advantages and disadvantages offered by each type of museum site. As a result, some of the potential benefits associated with the establishment of a new museum may well have been lost. There are many examples of failure to take advantage of the opportunities offered by this type of initiative due to a lack of appropriate analysis of specific territorial contexts. This can produce such anomalous situations as the locating of museums in areas of difficult access and/or museums being concentrated in areas that lack adequate complementary services for visitors.
The same can occur when museums undergo deep-seated transformations (whether physical or relating to content). In some cases, there is debate and reflection to seek the best possible solution and an answer to the question of whether to keep a museum in the same area or to move it to a new location. 2 However, the most common solution is to simply reopen the museum at the same place, despite the fact that this location would no doubt have evolved considerably since its initial inauguration. The potential benefits of changing the location of a museum are rarely considered. In such cases, the opportunity costs associated with unconsidered alternatives may be significant and certainly merit analysis.
It is within this context that the present research seeks to emphasise the importance of choosing museum locations on the basis of relevant economic, cultural, social and urban considerations. It should be remembered that the location chosen may, to a large degree, determine the extent of the impact that the museum in question has on the city as a whole.
2. The location of museums within the urban context; a practice that calls for analysis Interest in studying the location of artistic activities within the urban context is nothing new. In fact, at the beginning of the 1970s, there were notable concentrations of artists, art studios and galleries in the central areas of many cities and this was a phenomenon that had already been associated with important urban changes (Bianchini and Parkinson 1993, Barke and Harrop 1994, Kemp 2004). However, more recently, deliberate attempts have been made to use cultural activity as a motor for urban transformation and this was certainly the case in Europe from the 1980s onwards (Evans 2001, García 2004, Ponzini 2009).
The success of some cultural initiatives has led to the introduction of museumorientated policies affecting a number of different fields. As a result, it is now common to see museums and galleries used as stimuli to help regenerate degraded areas, to attract visitors to cities and/or as leisure spaces in their own right (Monclús and Guardia 2006). Some authors have suggested that this is due to the capacity of cultural facilities to attract population, private and public resources and even companies at a relatively low monetary and political cost (Strom 2002). What is certain is that in recent decades, nearly all Europe's major cities have created new, powerful museums (Landry 2006).
At the global level, museums have acquired an increasingly important role in the marketing of cities (Robertson and Wardrop 2004). Several authors think that in the futureto a greater or lesser extentthe strategic position of the city, understood as a whole, will depend on the cultural attractions that its institutions are able to establish and maintain. This will be of particular importance for cities aspiring to become points of reference for the new economy (Florida 2010).
This new situation has caused a fragmentation of the objectives of many museums. Complex elements interrelate within a wide framework and reinforce cultural industries. In this way, it is assumed that they help to improve the urban area, strengthen the self-esteem of the local population and promote greater social cohesion. Amongst other advantages, thisin turnhelps to reinforce the city's image, increase returns on investment, raise its scientific and technological value and/or improve its transport system (Paül i Agustí 2013).
Museums do not, however, only generate positive effects. They can also contribute to, or strengthen, some others processes. The most common of these include: the concentration and homogenisation of cultural policies as the result of the concentration of human and economic efforts on a particular museum or group of (Evans 1999); the creation of oversized amenities that do not meet real needs (Miles 2005); the simplification of the images of certain settlements (Ave and Corsico 1994); and the helping to touristify spaces which may culminate in gentrification processes (Aalst andBoogaarts 2002, Savino 2005).
In some cases, museums can therefore form part of policies that become key elements for defining the city, participate in the global competition to attract investment, visitors and population to it, and also help to increase the self-esteem of its residents. In the quest for greater prominence, urban interventions have tended to become more and more competitive and cities have increasingly resorted to ambitious initiatives that require ever greater amounts of investment (Castello 2010). This general increase in the resources devoted to museums has not, however, been accompanied by a corresponding increase in the number of studies conducted to assess their impact.
One factor which has so far received relatively little attention and analysis is that relating to the location of individual museums. Choosing the right location for a museum is fundamental. Although such institutions can have positive and negative repercussions that affect the whole of a city, and even its surrounding region, it is in the immediate neighbourhood of a museum that the most visible consequences will tend to be seen (Valentino 2003, Pilati andTremblay 2007). Administrators must therefore take great care when choosing where to locate a particular museum. The present research seeks to contribute to the ongoing debate regarding the need for comprehensive planning of the location of different museum services. When choosing a site for a museum, it is necessary to examine and assess the potential impact, risks and opportunity costs associated with each location in order to optimise the investment undertaken. It is not viable to exclusively base such decisions on such factors as the land available, the prestige of the site in question and its historical tradition; instead, it is recommended to assess the wider impact that a certain location could have upon its host city.

Methodology
In the present study, we analysed the locations chosen for new museums in three cities with comparable economic and social contexts: Paris (France), Barcelona (Spain) and Turin (Italy). In order to individually analyse the different museums considered in this study, we took as our reference the lists provided by the official tourism web pages of the cities analysed. This made the sources homogeneous and comparable and also removed the need to consult all of the current rules and regulations in force in each of the cities studied. The final list included the museums and other collections that were open to the public in January 2013. According to the data obtained from these web sites, at the time of the study there were 273 museums, collections and exhibition spaces open in the three cities (see Table 1).
We focused our analysis on museums which opened between the years 2000 and 2013. 3 This included new museums and also those that had been subjected to substantial renovations; in other words, those which had been temporarily closed for such purposes for a period of at least six months. 4 Museums that were being renovated at the time of our study were also included. According to these criteria, the number of museums included in our study was reduced to 109. One of the first elements that we should highlight is therefore the fact that almost 40% of the museums in the three cities had either been created or substantially renovated since the year 2000 ( Figure 1).
The research conducted included: visits to 84 of the 109 museums studied; the gathering of information (from sources ranging from tourist leaflets to annual reports); analyses of web pages; andin the most significant casesinterviews with the managers of the different institutions concerned. Abundant documentation was gathered relating to the different institutions. Some interviews were conducted, 5 with general managers and 27 with the public relations managers of some of the museums studied. 5 For each of these museums, our study sought to observe the main initiatives undertaken in a number of different technical areas. First, we compiled different types of information relating to the location of each museum. To study this factor, we classified the museums into one of four different groups: Historic building with a museum tradition: a building constructed before 1940 which was not initially designed as a museum but which was later transformed into a museum related to its initial function: palaces, artists' studios and churches. Historic building without a museum tradition: a building constructed before 1940 which was not initially designed as a museum but which was later transformed into a museum unrelated to its initial function. Modern building not designed as a museum: a building constructed after 1940 which was not initially designed as a museum. 6 New purpose-built building: a building inaugurated after 1940 which was specifically designed as a museum. Within this part of our study, we also analysed whether the museum was located in an area regarded by the local authorities as being particularly sensitive. When we mention 'sensitive areas', we refer to neighbourhoods or parts of the city which have been recognised by their respective city councils as exhibiting certain special characteristics which need to be preserved or protected. These are areas in which the local administration intervenes directly in order to deal with potential problems and to improve the living conditions of local residents. The actions undertaken in these areas may involve improvements relating to social, economic and/or urban planning issues. To be more specific, in the case of Barcelona we analysed whether a particular museum was located in an area covered by a Pla de barri (neighbourhood plan). In Paris, we studied whether it belonged to a Projet de Renouvellement Urbain, Quartier de la Politique de la Ville or Zone Urbaine Sensible. In Turin, we established whether it formed part of an Urban2; Contratti di Quartieri; Programmi di Recupero Urbano or Azioni Locali area.
The third aspect that we studied was whether the neighbourhood in which the museum was located had been subjected to any form of physical transformation during the period 2006-2012. We understood this to include any urbanisation work, the creation of new public spaces, housing renewal projects and/or similar interventions undertaken within a radius of 500 m from the museum in question. 7 The second large group of indicators analysed was related to questions of image. First, we analysed whether the touristic image of the area had undergone any type of change (such as an increase in its promotion, improvements in sign-posting or plans for tourism-related improvements). We then analysed whether any hotels within a radius of 200 m of the museums studied had been either created or remodelled. 8 Finally, we looked for improvements made to local transport systems (such as bus routes, underground services and/or special tourist buses). This qualitative information was complemented with the collection of quantitative data: the total number of visitors to these museums and their profiles, the number of people working in these museums and the budgets of the museums studied. Unfortunately, there were some serious deficiencies in the information obtained, which made it difficult to compare; as a result, we had to discard it. 9 Finally, we collected information about the type of management employed at the different museums. We divided this information as follows: State-owned museums: museums 100% owned by the state or by some other type of state-owned institution (such as hospitals, police stations and post offices). Local corporation museums: museums 100% owned by a municipal corporation or local authority (such as the Ajuntament de Barcelona, Mairie de Paris or Comune di Torino). Other public museums: museums belonging to public bodies that are neither national nor local (such as regions, provinces or federations of municipal authorities …). Private museums: museums which are not public property (including those owned by NGOs, foundations and church-owned museums …). Public-private museums: museums jointly managed by public and private agents. Other cases not mentioned above: other museums, including university museums and public museums that are run by universities and/or similar academic and/or social institutions.
Our analysis of the different museums allowed us to identify examples of both local and general types of behaviour, some of which are outlined below. We first present our findings for each individual city and then offer more general data that are applicable to all three of the cities studied.

The locations of the museums within the different cities
To make the results for the three cities analysed comparable, we mapped the different museums on a scatter graph. The centre of the graph corresponded to the historic locations of the three city council buildings (the Plaça Sant Jaume in Barcelona; the Place de l'Hôtel de Ville in Paris; and the Palazzo di Città in Turin). We then plotted the positions of the different museums according to their distances from their respective city council seats. 10 This comparison was possible because the three cities have comparable surface areas. Barcelona has a surface area of 101 km 2 ; Paris of 105 km 2 ; and Turin of 130 km 2 . The distances between the points on the peripheries of the three cities also presented similar values when measured from east to west and north to south: Barcelona We used these distances to provide a framework and to give dimensions to the graph in order to make it possible to observe the concentration of the museums. Finally, it is important to add a note concerning the methodology applied: the maps only include museums with physical locations that are open to the public; virtual museums have not been included. Similarly, where a given museum has several different physical locations under common management, we have only mapped the building that serves as its headquarters.
As it is possible to observe from Figure 2, there is a clear concentration of museums in the most central spaces. An area within a 2 km radius of the respective city council headquarters, 11 corresponding to approximately 10% of each municipal territory, contained 120 museums (44% of the overall total). If we extend this central area to a radius of 4 km (50% of the municipal territory), we find 199 museums (72.9% of the total). In other words, only 27.1% of the museums in these cities were located in more peripheral areas (74 museums). One of the first elements to highlight is therefore the high concentration of museums in and around historic city centres.
In this study, Paris was the city with the greatest number of museums (144). It was also the city that had opened or renewed most museums since 2000 (54). Figure 3 shows that 37.5% of the Paris's museums have opened or been renovated since 2000. In Paris, 36.1% of the city's museums (52) are located within a radius of 2 km from the City Hall. We could therefore say that the construction of new museums has tended to follow a pattern of relatively central location. Outside this central area, it is relevant to highlight the area around the Tour Eiffel (west), where 10 new museums have been created. The location of new museums in Barcelona is rather different from what is usually observed elsewhere. Unlike in the other cities, the creation of museums in its historic centre has been extremely limited: only 28 museums have opened in this area (35.9%). New museums, whether public or private, have tended to be located over a larger area of the city and particularly in two specific areas: up on Montjuïc hill (south), which hosts five of the new museums (three municipal and two private museums) and in an area near the end of Avinguda Diagonal (east; three municipal and one private museum) ( Figure 4).
Turin, the other city studied, is the city with the greatest percentage of new or renovated museums. To be more precise, 52.9% of the city's museums have either opened or reopened since 2000; 27 museums in total. Most of Turin's new museums are located within a 2 km radius of the City Hall (18 museums, 62%) ( Figure 5).
Having made this introduction, it is now time to analyse the peculiarities of the distribution of museums in each of the cities studied.

The influence of ownership on museum location
Traditionally, within the context analysed, museums have tended to be located in central areas. From the data obtained, we can state that the majority of museums with a certain historical tradition tend to be located in the central areas of their respective cities (Table 2). Although there has been a certain tendency for the presence of museums in historic buildings located outside central areas to increase, for the moment, their numbers remain rather testimonial. This information is valid for all of the ownership groups.   The various changes experienced by these cities over the last few decades could have modified this location pattern, but the data we have analysed show that this has generally not been the case, particularly as 46% of the new museums have been given central locations (52 museums). This is a trend that becomes even clearer when we analyse the distribution of museums based on their ownership. Generally speaking, it is possible to clearly differentiate between the behaviour of the different administrations (Table 3). In both Paris and Barcelona, the public administrations (whether municipal or state) have tended to establish or renew museums in areas which are quite distant from the historic city centre. In the case of Turin, the majority of new museums continue to be located within the area nearest to the City Hall. But, although they do not yet constitute the majority, in Turin, it is also possible to observe a tendency for museums to move out of the historic city centre. There is a centre-periphery movement which could potentially advance a more general tendency towards the location of museums in areas further from historic centres (specially the 2-4 km radius). These data also made it possible to observe another change. In contrast to what was observed with the public museums, in all three cities, the majority of the new museums with private participation tended to be located in historic city centres. In Paris, the majority of these museums opened before the year 2000 were located within 2 and 4 km of the City Hall. The majority of the museums which opened after this date were located within 2 km of the historic city centre. The same occurred with museums of mixed (public-private) ownership. In Barcelona, the percentage of private museums occupying central locations remained at 30% (9 museums). Finally, despite a fall in percentage terms, the number of private museums occupying central locations in Turin increased from 4 to 8. As a result, in contrast to the tendency observed for public museums, private museums do tend to be located in central areas.

The types of building that house museums
One possible explanation for location of these museums would be the type of building in which they are housed. In the three cities analysed, the areas around the respective City Halls are also those containing their most historic buildings. This partly helps to explain the concentration of museums observed. 71.6% (78) of the new museums are housed in buildings constructed before 1940 ( Figure 6). The    desire to conserve certain buildings, which are often centrally located palaces and churches, has favoured the location of museums in central spaces. Even so, it should be stressed that only 57.1% (45) of the new museums housed in historic buildings are located in central areas. The progressive changes that have taken place with respect to what is understood as a heritage building has meant that a number of different types of building have gradually come to host museums and has resulted in the previously mentioned association between former palace and museum disappearing. The Pinacoteca Giovanni e Marella Agnelli museum, which occupies part of the site of the former FIAT factory of Lingotto (Turin), provides an excellent example of this. As a result, tradition and the reuse of what were once monumental buildings only partly explain decisions to locate museums in central spaces; an important part of such decisions come down to the desires of their administrators. This will on the part of the administrator can be more clearly observed in the distribution of museums in buildings constructed after 1940 (31 museums) (Figure 7). Only 50% of the museums that have taken advantage of buildings constructed after 1940 are located within a radius of 2 km from the respective City Halls (15). Amongst the purpose-built museums, this percentage is lower: 9.5% (3 museums). As we might expect, the location of new buildings in consolidated central areas of cities can be a complex issue. The situation changes, however, if we consider a radius of 4 km. There were 13 museums (61.9%) within this area, with 9 of them (90%) housed in buildings constructed since 1940. This shows that while there may be certain difficulties associated with locating museums in the most central areas, this possibility does exist, as demonstrated by major projects like that of the DHUB in Barcelona.

The types of area that house museums
One final aspect to analyse with respect to museum location is the area in which a museum is established. In general terms, 12.8% of new museums (14 museums) have been located in what are regarded as sensitive areas. Even so, the percentages vary enormously from city to city. Paris (3 museums, 5.5%) and Turin (2 museums, 7.4%) have hardly opened any new museums in sensitive areas. In contrast, in Barcelona, 32.1% of the new museums (9) have been opened in these areas.
The five museums that have been opened in sensitive areas of Paris and Turin have all formed part of relatively unambitious projects, in terms of the investment, number of workers employed and promotional campaigns. In contrast, the nine museums in Barcelona exhibit two different types of profile. The typology of the private museums is very similar to that seen in the other two cities, though with the notable exception of Can Framis. In contrast, in Barcelona there has been a clear public preference, particularly exhibited by the municipal authority, in favour of sensitive locations. The municipal administration has sought to locate major museums in these areas in order to attract the maximum possible number of visitors and thereby regenerate them both socially and economically. This has clearly been the case with two newly constructed museums. The DHUB has helped to structure the major communications junction at the Plaça de les Glòries Catalanes and to renew the technological district of Barcelona: 22@, while the Museu Blau has been located near the space occupied by Barcelona's great waste water treatment plant.
In this case, Barcelona therefore clearly constitutes a particular case, as it has chosen to house important cultural amenities in sensitive spaces. We shall now go on to examine some of the repercussions of this choice.

The repercussions of the museums according to their locations
Once we have analysed the locations of the museums, we can start to look at the repercussions that they have had. To do this, we focused on the four aspects that we mentioned earlier: changes to transport; changes in the offer of hotels; changes in the promotion of tourism; and changes related with urban planning. We must, however, make it clear that in this section we shall only analyse the most basic and direct repercussions that have been associated with the establishment of new museums. We have not studied repercussions that cannot be directly related to museums. As a result, the indirect repercussions of these museums, which mayin some caseshave been greater than those directly associated with them, have remained outside the scope of the present study.
With regard to the first aspect: improvements in public transport services, we only observed improvements in the vicinity of 17 centres (15.6% of the museums studied). The values observed were similar for Barcelona (7 museums, 25%) and Turin (6 museums, 22.2%), but significantly lower for Paris (4 museums, 7.4%). These values were conditioned by a number of factors, the first of which could have been an aspect that we were not able to analyse: the greater density of Paris's transport network than those of the other two cities.
Secondly, we must be conscious of the fact that historic city centres tend to have the best public transport services. In consequence, a central location would tend to enjoy greater accessibility, while the converse would apply to more peripheral locations. As a result of this, peripherally located museums would probably have a greater capacity to contribute towards improvements in the public transport systems of the areas in which they are located. This statement does, however, need to be qualified, as this phenomenon was only observed in the cases of 17 of the museums analysed. There are some particularly relevant cases, such as changes in the location of urban bus stops (Caixaforum Barcelona). Even so, when a museum claims that there has been an improvement in its local transport service, this tends to reflect major changes that have affected wide areas of the city, as in the cases of the new underground train service in Turin or the tram service in Paris. The first thing that we learn is therefore that despite offering a certain potential, new museums tend to make little or no significant contribution to the development of transport services in the peripheral neighbourhoods of cities (Figure 8).
The second aspect that we analysed related to the repercussions of tourism. To assess this, we analysed the evolution of the offer of hotel accommodation in the areas surrounding the museums. The data collected showed that 46.4% of the museums in Barcelona (13) and 44.4% of those in Paris (24) had observed changes in the offer of hotel accommodation in the areas surrounding their museums, although this percentage was only 7.4% in the case of Turin (2 museums). Again, it is important to note here that the specific circumstances of each city could have affected the results obtained. The dynamism of the tourism sector in Barcelona and Paris is much greater than in Turin and this is a factor that has conditioned the development of new hotel accommodation. In some cases, this relationship is patently evident: the incorporation of museums as attractions on web sites coinciding with the inaugurations of both new museums and hotels. Even so, it is again important to understand that the results observed could have various causes. It would be necessary to carry out a much more in-depth analysis to observe whether museums have a direct influence on the opening of new hotel services. In the present study, the only thing that we can say for certain is that the presence of a nearby museum tended to be exploited by hotels as an element of positioning, without there being any significant differences between central and peripheral locations (Figure 9).
Where we did observe certain changes was in relation to new museums and the way of promoting a given area to tourists. We observed similar trends in all three cities: 32.1% of the museums in Barcelona (9) and 22.2% of those in Paris (12) and Turin (6) said that the promotion of tourism in their immediate areas had changed since they opened. Examples of this include the 'colline des musées', which unites the Parisian museums of Quai Branly, Musée d'Art Moderne de la Ville de Paris and Palais de Tokio and the Cité de l'Architecture et du Patrimoine. These are all museums near tourist attractions such as the Eiffel Tower but which have traditionally had difficulty attracting visitors. Similar cases can be observed in Barcelona, with the promotion of museum tourism in the new 22@ technological Along these lines, Figure 10 shows how it is possible to observe a certain correlation between the location of museums and changes in the promotion of tourism. New museums located in the most central areas of the three cities rarely  help to change their touristic image, whereas more peripherally located museums do tend to change the images of the areas in which they are located. Promotion associated with tourism is also a factor element that can have an important multiplier effect.
The capacity of museums to bring about a change of image related to tourism is also repeated in their capacity to change the neighbourhoods in which they are located. Figure 11 shows the museums that have been located in sensitive areas. In these cases, the objective has often been to accompany processes of physical and social change in these neighbourhoods. Barcelona has clearly opted for this strategy, with 32.1% of its new museums being located in sensitive areas (9). In the other two cities, the respective percentages are much lower: 7.4% for Turin (2) and 5.5% for Paris (3).
Analysing the effects that museums have had on their immediate vicinities, it is possible to observe how 32.1% of the museums in Barcelona (9), 29.6% of those in Turin (8) and 14.8% of those in Paris (8) claim to have contributed to the transformation of their surrounding areas ( Figure 12). Once again, these values tend to be highest for areas further from the centres of the different cities. In other words, opening a new museum could lead to improvements in several physical aspects of cities, but principally in areas furthest from the historic city centre.
Despite the range of possible impacts, we can state that the establishment of a museum at a particular location may favour a number of processes. In other words, museums are capable of becoming motors for change in certain areas, but it is important to choose their locations with care, to evaluate whether a central location is better than one on the periphery, and to be conscious of the possible impacts. We should not automatically expect major changes in all sectors. If inappropriate locations are chosen, their impacts will be limited, as they will tend to concentrate in already consolidated spaces where most of the economic and cultural, and even governmental, activity already takes place. It could, however, be argued that choosing such sites could also have positive effects as they would tend to reinforce the existing offer. Whatever the case, it is necessary to evaluate different locations in line with the possibilities and opportunities presented by each promoting institution. Such a strategy results in museums being housed in a wider variety of locations where the presence of a cultural institution can help to enrich the surrounding area and to give it greater dynamism.
In short, we can group the different possible repercussions that were previously mentioned into three large groups. According to the group that the museums fit into, they will have different physical, social and/or cultural repercussions for their host cities. One element to highlight here is that these three categories, which have emerged from studying Spanish, Italian and French cities, are similar to those suggested by other authors for other European contexts, and particularly for the UK (Evans 2005, Miles and Paddison 2005, Vickery 2007). These main characteristics of each group would be: Museums that are seen as having a strategic role within processes of change: these are museums that have mainly been promoted by local administrations. The museum is used as a driving force and as a symbol of regeneration. Examples of this type of initiative taken from our study would be the creation of cultural facilities such as the Parco Arte Vivente, in Turin, and its relationship with the recovery of the General Market area. Alternatively, there is the establishment of the Museu de la Música, in Barcelona and the recovery of the area around the Glòries Catalanes square, a space that Barcelona City Council wishes to give a new centrality. These are, however, initiatives that are unlikely to interact with the people living in their immediate surroundings. In these cases, the aim has been to reach out to a wider audience. Museums, however, often lack the promotional and mediating tools to do this and so end up attracting the same public that typically frequents other museums: middle-high class and mainly local people (Mironer 2001). Museums that are seen as actively participating in processes of change but which do not lead them: cultural activity tends to be better integrated within the immediate social and economic environment of these museums than in those referred to in the previous category. In combination with other initiatives, they help to improve the urban, social and economic quality of their surroundings. Such initiatives tend to be either private or municipal. The examples studied include the museums on Montjuïc hill in Barcelona which offer a varied programme and attract mixed audiences from a range of origins and backgrounds (from both local citizens and tourists). The same objective is pursued, although with much more modest means, by both the Museo A Come Ambiente and Fondazione Merz in Turin. Museums that are seen as playing a secondary role in processes of change: cultural activity does not integrate into any specific development strategy. Museums act in an isolated way, to the point that their surroundings are unaffected by them. Although they provide visitors, most of them are tourists who only make short stays and generate little economic activity. Such museums tend to be state owned. In the majority of cases, their limited involvement with the activity that takes place around them is due to budget limitations (examples include the Palazzo Reale in Turin, the Botanical Gardens in Barcelona and the municipal museums in Paris). In other cases, it is a certain lack of vocation which leads local authorities to become actively involved in improving the management of this type of space (as in the case of the Museo Egizio in Turin or museums such as the Orangerie and the Jeu de Paume in Paris).
The most common strategy for incorporating new museums into the process of urban regeneration tends to be the first one mentioned. Even so, it is difficult to use museums to lead complex processes of physical, economic or social transformation of a space. This situation is repeated in the type of initiatives included in the second group, as cultural policies only tend to form part of broader initiatives when other projects are already in progress. With regard to the third point, there is the possibility of museums that fail to interact with their local populations and the economic activity in their immediate areas going into decline. In spite of this, there are still institutions that are solely addressed to occasional audiences, especially tourists, such as the Museu del Modernisme Català in Barcelona or the Musée Maxim's in Paris. This limit the possibilities of making a wider and longer lasting impact as far as aspects such as audience loyalty, the interrelation between stakeholders and changing the mentality of the local population are concerned.

Problems with museums in peripheral locations
In theory, cultural facilities can be located anywhere in the city, but as we have already seen, consolidated urban centres tend to attract the largest number of new museums. Our findings coincided with those reported for Spain by García et al. (2003), who observed a high degree of territorial concentration of cultural institutions in a relatively limited number of areas. The reasons for this are that, despite the potential benefits of peripheral locationswhich have already been mentionedit is impossible to ignore the fact that such areas also tend to present greater challenges than those usually encountered in the most central areas.
First of all, and as other authors have already observed, spatial dilemmas often result from the tensions between the centre and periphery (Bianchini andParkinson 1993, p. 201, Bios 2001). In order to avoid these rivalries, it is assumed that the best solution would be to place museums in central locations. There is a tendency to believe that under such circumstances accessibility would then be equal for everyone as urban centres tend to have better transport infrastructure. This homogeneity of access would, however, only applicable in terms of space; it would not apply to social, economic or cultural aspects. As Nelson (2001) warned, placing museums in central locations, which should supposedly prevent marginalisation, has led to a concentration and homogenisation of cultural initiatives at a loss of diversity in more peripheral areas. A study conducted by the Council of Europe reinforced this perception (Council of Europe and Ericarts 2002), demonstrating that the location of cultural facilities was still mainly based on the distribution of population. This situation tends to reinforce the importance of certain spaces in cities and, at the same time, tends to restrict the progressive scattering of institutions throughout the wider territory. In spite of their theoretical potential, there is therefore still considerable reluctance to establish cultural facilities in more peripheral areas. This is particularly applicable in the case of non-municipal public administrations.
A second aspect to take into consideration is accessibility to different points. Facilities located far from the centre are generally less accessible. This situation can, however, be improved by better signage and, above all, by the creation of public transport. We observed a number of successful experiences, such as that of CosmoCaixa in Barcelona. However, as we have already noted, in general terms, we observed a lack of interaction between those responsible for the museums and responsible for transport. A better coordination of these areas would be desirable for the planning and management of museums and would facilitate access to them.
A third aspect is the existence of a certain number of complementary offers. Occupying a central location offers more possibilities of generating synergies with the surrounding area than a peripheral one; most of the shopping, restaurants, leisure spaces and even research centres tend to be located in urban centres. A greater museum offer can attract more visitors and may encourage tourists to extend the duration of their visits. In contrast, museums located on the periphery are usually isolated and need to create (or help to create in the long term) complementary offers that are compatible with visits made to them. The contrasting cases of Turin and Barcelona are quite enlightening. While the Polo Reale in Turin, 12 which was given a central location, only coordinates the existing museum offer, Montjuïc in Barcelona, a green area with access difficulties due to its topography, needs to overcome the handicap of a lack of complementary offers in its immediate vicinity.
Finally, the last case in which a peripheral location constitutes a disadvantage is in relation to obtaining resources. Institutions that occupy central locations tend to enjoy greater visibility, which is the main element that sponsors seek in order to make their investments profitable (Strom 2002, p. 10).
There are therefore a number of elements that should be taken into consideration before deciding to set-up a museum in a peripheral location. The potential difficulties should not, however, be automatically seen as conditions that prevent the location of museum facilities in such areas. Although there may be various difficulties, the potential benefits in terms of urban, economic and/or social repercussions may be too important to ignore, so the potential of other more peripheral locations must also be analysed.

7.
In conclusion: what needs to be considered when planning the location of a new museum As we have already seen, the locating museums may carry with it a series of advantages and disadvantages. It is therefore essential to appropriately calculate the repercussions that these processes will have. A central location may offer a series of advantages. Even so, this does not mean that other locations could not also offer positive repercussions. The question that we have to ask concerns which objectives we should follow and what the best location would be in order to obtain desired repercussions.
As we have shown, in recent years we have observed a growing tendency to assign some specific museums to peripheral locations within cities. Only a few institutions continue to have reservations about locating their museums in peripheral areas. Even so, the diversity of the ambitions, budgets and objectives of the different facilities considered in the study prevents us from reaching an invariable set of conclusions about which type of location, whether central or peripheral, would constitute the best investment. It is, however, beyond doubt that the location of some emblematic museums in fully consolidated areas has had almost no impact on the regeneration of the surrounding urban territory. This has certainly been the case of the provisional location of the DHUB in Barcelona: just opposite the Museu Picasso, the second most visited museum in the city. It is highly probable that a museum of these characteristics would have been profitable in a peripheral location even though this would probably affect the type of visitors that it would receive and could force it to make a greater communicative and programmatic effort in order to make itself better known to the public.
Of course, we need to be aware of the fact that museums alone cannot be the solution to all types of problems. Many processes cannot only be approached from the perspective of cultural infrastructure, but need firmer initiatives on various fronts, to which culture can be added as another factor.
It is therefore desirable to carefully consider the locations of the different museums. Although it is true that some initiatives can follow a given logic in a consolidated historic city centre, this should not lead us to discard other possible locations from the very onset. Choosing an easy solution like a central location has short-term image-related benefits, but in the medium term, there can be an opportunity cost to consider deriving from the lost opportunity to intervene in areas where this type of initiative would be more useful from a social, economic and urban planning perspective; three fundamental aspects for the progress of a city.
In general terms, we can therefore state that for new or renovated museums to have the biggest possible influence on the urban area there are a number of important questions that must be considered.
First, a wide consensus is needed before the location of a museum is decided; this applies not only to the political arena, but also with respect to social organisations. Such a requirement is relatively easy to obtain, particularly if the project is eye-catching (large, with a nice design and star architect) and offers thematically 'neutral content' 13 (as people, media and politicians rarely oppose the creation of a new museum). It is also essential to obtain the full support of the local citizens, which should go beyond simply silencing any vociferous criticism. Citizens should be encouraged to directly participate in all aspects of the process of establishing the new museum. It is then recommendable to have local citizens actively involved in the management of the museum once it opens. This is, however, quite a complex process, which means that it is rarely achieved.
Secondly, the effectsboth positive and negativeshould affect citizens over a large area. All too often, initiatives only benefit a very small part of the population, whether in social or territorial terms (Harvey 1989, McGuigan 1996. For example, if all of museums are located in the same area, this may create greater dynamism and increase tourist flows, but it can also create resentment in other parts of the city which will demand similar types of investment. 14 Such situations can be avoided if the cultural interventions within the city are planned from the outset.
Finally, one last danger lies in the justification of emergency actions which do not form part of city plans or provide solutions that promise optimum development. It is therefore possible to justify actions undertaken without taking into account such questions as transport, signage and/or complementary offers, all of which are essential to the success of a museum, wherever it is located. Often, only the most central areas, with the best infrastructure are sought after, or those which already have good potential for development.
It is necessary to observe that when it comes to choosing the location for a museum, space is not neutral. It is possible to obtain similar repercussions in terms of image, economic and/or cultural impact, but the efforts required to achieve this vary according to the location chosen. This is something that, in turn, carries with it a significant opportunity cost. As we have seen in the cases studied, the locations nearest to historic city centres tend to have more limited repercussions. In contrast, new museums in peripheral locations help to promote change, especially in terms of image and urban transformation. If the correct choice is made, the repercussions can be important. If the chosen location is not correct, the investment made may not have the expected returns. Several questions need to be taken into consideration when taking decisions about where to locate a museum, because the repercussions for the city could vary considerably according to the location chosen.
1. It is quite difficult to define what is understood by the term 'museum', as the concept may vary from country to country. In the present work, we have avoided imposing a single definition. As a result, we shall understand a 'museum' to include any space which the local councils of the cities studied have considered to constitute a museum. For the purposes of the present work, this includes museums, collections, galleries and exhibition spaces among other facilities. The specific criteria followed in this work are outlined in greater detail in the methodology section. 2. We should underline the fact that in some cases the extensive renewal of a museum could have a similar cost to the creation of a new museum. The expansion of the Museu Picasso in Barcelona cost €6.7 million. In contrast, the construction and equipping of the new Museu Olímpic i de l'Esport in the same city cost €8 million. 3. The decision to take 2000 as the base year is explained by the desire to establish a cut off point at the beginning of the decade that would offer a certain historical perspective. The year 2000 was not, therefore, chosen because of any major change. In fact, it is difficult to establish a significant date that would be applicable to all three of the countries studied. Even so, we also examined other cut off dates in order to see if this would alter the results for 2000-2013 (for example, we compared the results for the decades 2004-2013 and 2000-2010) but they were similar. 4. We must remember that we only studied museums that were open at the time of conducting this research and not the ones that were closed during this period. Even so, the Rock Museum of Barcelona constitutes a rather special case. It opened its doors in 2011 and closed in 2012, but at the time of the study, it was still featured on the web site that was analysed. 5. Semi-structured interviews lasting approximately one hour were conducted within the museums themselves. These interviews were carried out in the local languages of the different locations. Notes were taken of the answers provided. 6. The cut off in 1940 was based on separating buildings built before and after World War II (and after the Spanish Civil War). It was thought that the war period would have corresponded to a period of almost 10 years without any major museum construction (we only recorded 7 museums opening during the 1940s). 7. Using a cut off distance of 500 m is considered normal for a renovation project in a case of urban development; a distance of more than 500 m is considered remote. This is approximately the distance that a person can walk in 10 min and, for example, the limit used by Noguera Tur et al. (2011). 8. The distance of 200 m from a hotel was based on an approximate calculation. In the cities analysed, we considered that there was a means of transport (whether bus or subway) approximately every 200 m. If a tourist staying at a hotel could take a means of transport before reaching the museum, the museum will go unnoticed. 9. For example, we could cite the fact that only 31 museums had analysed the profiles of their visitors. Only 18 provided data about their budget. As far as the number of workers was concerned, different criteria were used to count them (according to whether they were part-time or sub-contracted staff). Something similar occurred with the number of visitors. Although 58 museums provided data, there were still too many differences when it came to making comparisons (visitors, users, tickets sold). 10. The choice of the buildings that house council buildings as central points responds to the fact that these are buildings which are located in central positions in their respective cities. These buildings are highly representative and many of them have not changed their location in centuries. The building that houses the City Council of Barcelona dates from 1369, that of Paris dates from 1357 and the one in Turin is also medieval in origin. 11. We took as our reference an area within a radius of 2 km, which we considered an easy distance to cover on foot in less than an hour. This was therefore a distance that should not have conditioned mobility. 12. This initiative seeks to create in the medium term, a cultural district based on both the existing and future institutions that will share the Palazzo Reale (the former royal residence of Turin) as the main attraction. 13. By 'neutral content' we mean topics that don't generate controversy among the majority of the population. For example, it is unusual that a museum dedicated to arts becomes criticized by population or media. By the other hand, a museum dedicated to the history (for example, la Cité nationale de l'Histoire de l'Immigration in Paris), could generate more critical opinions about his contents. 14. We observed this type of criticism in the three main initiatives in each of the cities studied: in Paris about the Musée du Quai Branly, in Barcelona about DHUB and in Turin about the Museo di Arte Orientale.