The effects of tourism on HSR: Spanish empirical evidence derived from a multi-criteria corridor selection methodology

Journal of Transport Geography 47 (2015) 37–46

Contents lists available at ScienceDirect

Journal of Transport Geography journal homepage: www.elsevier.com/locate/jtrangeo

The effects of tourism on HSR: Spanish empirical evidence derived from a multi-criteria corridor selection methodology Begoña Guirao ⇑, Juan Luis Campa Departamento de Ingeniería Civil: Transportes, ETSI Caminos, Canales y Puertos, Universidad Politécnica de Madrid, Avda. Profesor Aranguren, s/n, 28040 Madrid, Spain

a r t i c l e

i n f o

Article history: Received 11 February 2015 Revised 7 July 2015 Accepted 28 July 2015

Keywords: Multi-Criteria Analysis (MCA) High-speed rail Tourism Transport planning

a b s t r a c t The exorbitant cost of new High-Speed Rail (HSR) lines requires a selection methodology to define which HSR corridors within a network should be built first, and the most suitable evaluation tool appears to be the multi-criteria approach. In any corridor-ranking methodology, and especially in countries with high tourism attractiveness, tourism impacts on HSR should be considered as a variable. In addition to economic geography and destination choice models, the current literature on tourism demand is dominated by econometric models using a single-equation time-series based approach. However little research has been done so far on methodologies to rank HSR corridors taking into account the tourism variable. In 2014, a ranking methodology developed by Todorovich and Hagler was validated using the current Spanish HSR network. Twelve variables were used to create an index to assign scores to the city pairs, but tourism was not included as a variable. The findings showed the consistency of the model for ranking pairs mainly in the top O–D relations; however the tool failed to discriminate clearly between secondary groups of corridors. The aim of this paper is to assess empirically the positive effect of tourism on HSR and to enhance the abovementioned ranking tool with a tourism database. The new methodology is tested by application to 1176 city pairs in Spain, and the results clearly show that the implementation of a tourism variable helps discriminate between secondary groups of corridors and offers a more effective approach for determining the implications of tourism on HSR. Ó 2015 Elsevier Ltd. All rights reserved.

1. Introduction The exorbitant cost of building a new HSR line requires empirical evidence of its economic and social efficiency. Projects must be assessed to define which corridors should be built first, considering all the economic activities reinforced by HSR, among other variables. Tourism is the most important activity in many countries, and tends to increase HSR demand. Recent decades have seen a significant evolution in the methodologies for appraising transport projects and the variables considered, as described below. Project assessment in the European Union is generally seen as a tool to assist the process of planning transport systems: it provides relevant information for decision-makers, but does not actually ‘‘make’’ decisions (Bristow and Nellthorp, 2000). Since transport infrastructure tends to be publicly owned in most European member states, assessment of infrastructure investment is often concerned with achieving a range of social (as opposed to private) ⇑ Corresponding author. E-mail addresses: [email protected] (B. Guirao), jl.delacampa@alumnos. upm.es (J.L. Campa). http://dx.doi.org/10.1016/j.jtrangeo.2015.07.010 0966-6923/Ó 2015 Elsevier Ltd. All rights reserved.

objectives. The types of decision supported by assessment include whether or not particular projects offer good social value for money, prioritizing projects within a programme, choosing between alternative solutions to a common problem, and selecting the optimal time to make an investment. Good social value for money includes minimal environmental damage, and other objectives such as equity, accessibility, long-term cash-flow implications for the state, and the implementation of regional development policies. These types of conditioning factors are usually quite difficult to tackle within a project assessment. This paper focuses on prioritizing HSR corridors in a new network, and considers the implications of the tourism variable in the planning process. This type of approach is also linked to transport project evaluation methodologies, and public sector investment assessment in the literature review follows two main frameworks: broad CBA (Cost Benefit Analysis) and Multi-Criteria Analysis (MCA). The CBA is the recommended method in the USA, although other impacts are added for large projects, and MCA variations are used at the regional level (Hayashy and Morisugi, 2000). EU countries have a clear tradition of using CBA for assessing public

38

B. Guirao, J.L. Campa / Journal of Transport Geography 47 (2015) 37–46

sector transport infrastructure projects, although two main problems emerge from the literature review. First, although it is commonly held that indirect socioeconomic impacts (such as regional development) are important to policy, there is very little agreement on how to include these impacts in the CBA and under what circumstances they are additional to direct impacts. Second, the role of CBA is also influenced by more general institutional issues with a fixed strategic position, and the decision process involves an ever-increasing number of participants. Experiences in France and Germany shed light on these two main issues. According to Quinet (2000), transport project evaluation methodologies in France have focused on CBA (Cost Benefit Analysis), although in practice this has ranged between the strict application of economic theory based on surplus analysis, and a rather loose use of multi-criteria analysis. Project evaluation and CBA, relying on the strict principles of economic theory, have recently received a boost, although some new processes (such as hearings, enquiries and public consultations) are gaining in importance in the final decisions, meaning that CBA in France is becoming more open to democratic debate. In Germany, the evaluation method in the Federal Traffic Infrastructure Plan (BVWP) is based on a CBA analysis for investment projects in the transport sector, and was specially designed to allow a comparative assessment of projects. However there is a scientific argument (Rothengatter, 2000) that a ‘‘strategic system view’’ is still required, and a system approach assesses the strategic impacts on a network or even at the intermodal scale. The latest developments in individual countries and the EU also show a move towards the development of comprehensive multi-modal appraisal methodologies (Bristow and Nellthorp, 2000). In Spain, as in France, CBA is the most widely-used evaluation methodology for transportation projects, although the assessment of externalities (pollution, noise, accidents and congestion) is a major source of conflict when applied to ex-ante and ex-post HSR analysis (Coto-Millan et al., 2007). The ex-post CBA applied in Spain to any kind of transport project – even HSR – rarely considers regional development impacts, equity or prioritizing criteria within a network, and there is a general feeling that CBA is ‘‘not enough’’. This general impression is evident in the recent literature (Macharis and Bernardin, 2015), which shows that MCA has become more widespread as an evaluation method for transport projects. Macharis and Bernardini retrieved 276 publications concerning the application or the applicability of MCA for transport projects, and detected an increased use of MCA for transport project assessment, especially over the last ten years. The analysis reveals that – after a more general category – passenger transport (like HSR rail) and notably mobility management are the most common types of transport project decision problems handled by MCA. In this context, the authors of this paper propose a multi-criteria approach to analyse the influence of tourism in the process of prioritizing new HSR corridors. Although demand forecast is not the only criteria for ranking corridors, it has become a key factor in HSR project assessment (particularly in CBA methods). Our proposed multi-criteria approach is not a demand model, as the main expected outcome is a corridor ranking and not a demand estimate. In terms of the tools used in traditional CBA to estimate future HSR demand, it should be noted that it is always difficult to estimate the traffic generated by a new transport infrastructure by traditional modelling (Ortúzar and Willumsen, 2001) due to the percentage of ‘‘induced passengers’’ – new passengers and new trips that are not transferred from a previous mode of transportation in a corridor. As noted by Ortúzar and Willumsen (2001), the sequential four-step model reveals a clear flaw in the generation stage, as it is virtually inflexible in response to any change in the network (offer). Attempts have been made to introduce a new concept in this stage by adding an accessibility

variable, although the experience so far has proven ineffective – at least for aggregated models – partly due to the difficulty in establishing an adequate accessibility indicator (Ortúzar et al., 2000). As a result, demand forecast for some new HSR lines in Europe (Ni et al., 1994) has been based on an ad-hoc model in which the induced traffic generated by a new infrastructure can be interpreted as a joint gravity model (using the generalised cost of each mode together with a modal split model). After highlighting the difficulties of estimating HSR induced demand, and in view of the fact that the main aim of this paper is to analyse the effects of tourism on HSR using a multi-criteria corridor selection methodology, we can say that this research contributes added value as it is the first assessment of a ranking model including tourism as a variable. This article is divided into the following parts: first, introduction in Section 1, and literature review on tourism and HSR in Section 2; a description of the ranking model, its terms and equation, together with its application to the Spanish case (Section 3); the modelling process and results, including a tourism variable for the Spanish case in Section 4; and finally, the main conclusions (Section 5).

2. High speed rail and tourism: An overview There is practically no literature on empirical methodologies to assess the cross effects between tourism and HSR lines, even in Europe. Tourism supply is a complex phenomenon (Sinclair and Stabler, 1997) in terms of both the nature of the product and the process of delivery. It involves a number of elements that constitute tourism supply; there is a strong dependence on natural resources and historical and cultural sites; tourism activity (often) has a seasonal character; and consumers need to move towards the product. Some authors see tourism as a compound product (Caccomo and Solonandrasana, 2001) or a form of complementary demand for which the main components are transport, food, and accommodation (Morley, 1992). Despite its complexity, tourism is an economic activity and can be analysed using the new economic geography models. These models integrate the factor of the location of economic activities and take account of the costs of transport and the role of spatial competition. The core-periphery model introduced by Krugman (1991) showed that a change in transport cost induces a change in the intensity of spatial competition, which in turn influences the location of firms. Using Krugman’s core-periphery model, Masson and Petiot (2009) discussed the influence of the southern European high speed rail (HSR) between Perpignan (France) and Barcelona (Spain) on both tourism activity and economic development. Through a theoretical (not empirical) discussion using Krugman’s theory, they argued that HSR can facilitate the development of tourism activities, and particularly business and urban tourism. Another type of economic geography models uses gravitational and iso-tourist line models to evaluate the scope and levels of the tourism market. Crampon (1966) first introduced the gravitational model into tourism research, which was then adjusted and refined by others (Bao and Chu, 1999). The only application of this approach to a HSR scenario took place in 2012: the projected effects (not validated) of the Chinese HSR on tourism were examined by applying the iso-tourist line from a time–space replacement concept (Wang et al., 2012). They forecast the effects of the HSR network on three aspects: the redistribution and transformation of the tourist market, market competition on a larger scale, and reallocation of the urban tourism centre. As can be seen from this review of the state of the art, the link between tourism and transports cost has been widely studied. Apart from economic geography models, the choice destination approach, based on Rugg’s model, has played an important role.

B. Guirao, J.L. Campa / Journal of Transport Geography 47 (2015) 37–46

Rugg (1973) was the first to introduce a time constraint, the modification of the budget due to the transportation cost between alternative destinations, and the modification of the time constraint resulting from including the time cost between alternative destinations. The tourist – like any traveller – could be assumed to want to maximise utility. This maximisation considers budget and time constraints, as well as the time and monetary costs of transportation towards a principal destination (and between destinations). Assuming that tourists usually have a fixed holiday budget, Prideaux (2000) developed a model to demonstrate the dynamic relationship between holiday expenditure categories and the tourists’ point of origin. The model is based on a total holiday expenditure function which depends on the discretionary spending and accommodation costs at the destination, and the cost of transport to access the destination. The first approach to study the effects of HSR on tourist destination choice was first developed by Delaplace et al. (2014) in Paris and Rome. On the basis of a survey of tourists in Paris and Rome, data collected from the two surveys were used for a quantitative analysis using regression models and results demonstrated that HSR influences destination choice in different ways in the two cities, being in Paris the tourism more dependent on HSR. This interesting contribution was completed by Pagliara et al. (2015a) using Madrid case study, where a revealed preference survey was also conducted. Preliminary results showed that the Spanish HSR system seems to have a significant effect on the tourists’ choice to visit other cities close to Madrid, but the choice of Madrid as a tourist destination (like in Rome) is not influenced by the presence of HSR. Afterwards, a detailed and valuable comparison between Paris and Madrid case studies was developed by Pagliara et al. (2015b). Although these results are quite clarifying, the notion that destination choice may be influenced by HSR is no proof that the construction of a new line will automatically reinforce a tourist destination, increase tourist-sector revenues in this city (accommodation, restaurants, museums and so on) or even increase HSR demand. It is therefore important to quantify the increase in tourism demand in order to assess the influence of the new infrastructure. The existing literature on tourism demand is dominated by econometric models that tend to follow a single-equation time-series approach (Lim, 1997; Song and Li, 2008; Song and Wong, 2003), along with a few advanced studies of demand systems (O’Hagan and Harrison, 1984). This type of model is fairly dependent on the existence of a sound database. The only approach of this type applied to HSR corridors was developed by Chen and Haynes (2012). Through a multivariate panel analysis, they investigated the impact of Chinese high-speed rail systems on the tourism industry, selecting only the numbers of incoming foreign tourists and tourism revenue as the dependent variables (as data on domestic tourism demand was not publicly available in China). The results of this research confirmed that during the period between 1999 and 2010, the emerging high-speed rail services had significantly boosted tourism in China and provinces with high-speed rail services were likely to have approximately 20% more foreign arrivals and 25% higher tourism revenues than provinces without these systems. This model is fairly difficult to apply to European countries with HSR networks for two main reasons. First, it does not consider domestic tourists, and this percentage is quite high in Europe. Second, there is not always a database available on tourists in each city in Europe. Moreover, the cross impacts between HSR and tourism in a country like China, where the interurban transport network is less developed than in Europe, are probably of a higher magnitude and easier to detect by an econometric demand model. Apart from the abovementioned studies on the link between HSR and tourism (based on economic geography models and

39

econometric models of destination choice and tourism demand), there are other research works with interesting results worth mentioning in the literature. These studies describe specific HSR experiences (study cases) but do not lead to a general conclusion. Guirao and Soler (2008) examined the impacts of Toledo tourism on the new HSR link Madrid–Toledo. Through a survey approach, they found that tourism accounts for over thirty percent of weekday HSR ridership, and that this type of user finds it hard to obtain tickets due to the massive presence of commuters. Bazin et al. (2011) studied the impact of HSR on urban and business tourism in French cities near Paris such as Lyon. In the Paris–Lyon relation, the daily expenditure of business tourists may be up to four times greater than that of leisure travellers. Business tourism is a key strategic orientation for cities and large urban areas, although the possibility of returning within the same day reduced the average length of the visitors’ stay. Although this topic is the subject of great scientific interest, the literature review reveals that there are so far very few studies that empirically show the impact of tourism on HSR. This is an issue that needs to be addressed, and may be especially relevant at the planning stage, before HSR lines are built. Given that the construction of new HSR lines in any country should follow a prioritisation process and that this issue (as an indirect impact) is rarely taken into account by traditional CBA, the authors of this paper propose – as an initial approach – the inclusion of a tourism variable in the only existing HSR ranking model, a tool designed by Hagler and Todorovich (2009) for application in the US. Although the implementation of HSR in the United States is controversial and the scenario is totally different to Europe or Asia (Ryder, 2012), this multi-criteria methodology is the only ex-ante prioritizing tool applied to a HSR network in the world. This model has recently been used in Spain (Guirao and Campa, 2014a) and in Malaysia (Saat and Serrano, 2015) with interesting and consistent results, and in both these studies the original formulation was neither changed nor improved. Unlike Malaysia, in the case of the Spanish application the model was validated using current HSR traffic data, and it may initially appear surprising to use a prioritization method based on existing data. The main reason for applying an ex-ante tool to an ex-post scenario is that this is the only way to validate a planning tool. Transportation planners at present rarely check that their ex-ante theories and conditions for the future are met, and much can be learned from this kind of comparison to avoid repeating certain assumptions in future ex-ante evaluations. In the case of Spain, some deficiencies were detected in the formulation, (as the ex-ante results could be compared to the current situation), and the tourism criterion was clearly lacking. Spain has more than 20 years’ HSR experience, and operates the longest HSR network in Europe (2900 km). It offers a good scenario for this model improvement and validation, as the tourism sector represents 10.2% of its gross domestic product (GDP). This initial approach at the network level is the first part of an ongoing and more ambitious research project developed by UPM (Universidad Politécnica de Madrid), in which certain implications for tourism revealed by the ranking models are also being analysed at the local level. This paper contributes to the limited existing literature by developing the analysis of the impact of tourism on a HSR network. The added value of this research lies in the first assessment of a ranking model using current HSR traffic data and taking account of tourism as a variable. This ranking tool is designed to be used before the construction of a new HSR for an ex-ante analysis, but has been validated in Spain with current HSR traffic data. Section 3 below describes the original model of Hagler and Todorovich (terms and equation) and presents the application to the Spanish case without any changes to the original equation.

40

B. Guirao, J.L. Campa / Journal of Transport Geography 47 (2015) 37–46

Immediately after, the same section contains a demonstration and discussion of the modelling process and the results for the Spanish case when a tourism variable is included. 3. The ranking model applied to the Spanish HSR network The first attempt to develop a HSR prioritisation tool was made by two American urban planners in 2009 (Hagler and Todorovich, 2009). The proposed methodology was based on the hypothesis that five main categories of variables determine the value of the Ranking Index (RI) when scoring corridors in order to evaluate their potential HSR demand: population size, urban transit connections, origin–destination distance, economic vitality and congestion. These five categories were weighted and then added in an equation (Eq. (1)) that allocated scores to 27,000 city pairs in the US, with New York–Washington coming top of the ranking. The top city pairs appeared to be consistent from a potential demand approach, although the model was not validated with real data, as the US has no HSR network. The equation was applied only to American cities of over 50,000 inhabitants, and this process included approximately 600 cities and towns. The city pairs were created using a geographic information system (GIS), connecting each city to all other cities located between 100 and 500 miles (160 km and 800 km) from the origin city. This yielded approximately 27,000 city pairs across the nation on which to base the analysis. Table 1 gives an explanation of the meaning of the variables used by Hagler and Todorovich in this equation, and the values assigned to each variable by the authors can be found in the same paper (reference).

RI ¼ ðCRÞ þ 0:5ðLRÞ þ 0:5ðS LR Len IÞ þ 0:5ðHRTÞ þ 0:5ðS HR Len IÞ þ ðMet PopÞ þ 10ðMetro MainÞ þ ðCity popÞ þ ðMegaÞ þ ðCR 1Þ þ 0:5ðLR 1Þ þ 0:5ðE LR Len IÞ þ 0:5ðHRT 1Þ þ 0:5ðE HR Len IÞ þ ðMet Pop 1Þ þ 10ðMetro Ma 1Þ þ ðCity pop 1Þ þ ðMega 1Þ þ ðC LengthÞ þ ðG GDP ScalÞ þ ðTTI IndÞ

ð1Þ

Eq. (1) shows that the model is fairly dependent on the weight allocated to each variable. The values of the variables range from 0 to 3.0, and the authors logically give the maximum weight (10) to the variable Metro_Main (or Metro_Ma_1), which reflects whether

Table 1 Variables of the prioritisation model (Hagler and Todorovich, 2009). Variable

Meaning

CR CR_1 LR LR_1 S_LR_Len_I E_HR_Len_I HRT HRT_1 S_HR_Len_I E_HR_Len_I Met_Pop Met_Pop_1 Metro_Main Metro_Ma_1 City_pop City_pop_1 Mega Mega_1 C_Length C_GDP_Scal

Commuter Rail at Origin City Commuter Rail at Destination City Light Rail at Origin City Light Rail at Destination City Origin City Light Rail System Mileage Destination City Light Rail System Mileage Heavy Rail Transit Origin City Heavy Rail Transit Destination City Origin City Heavy Rail System Mileage Destination City Heavy Rail System Mileage Metropolitan Area Population of Origin City Metropolitan Area Population of Destination City Is the origin city the largest in the metropolitan area? Is the destination city the largest in the metropolitan area? Population Origin City Population Destination City Is the origin city located in a megaregion? Is the destination city located in a megaregion? Corridor Length (miles) Geometric mean of per capita GDP of the two metro regions (dollars) Combined TTI index of the two cities in city pair

TTI_IND

the origin city (or destination) is the largest in the metropolitan area. A metropolitan area is usually associated with an area of interactions between a core or cores (which can be defined using morphological criteria such as population or employment thresholds) and its hinterland of neighbouring municipalities showing a significant relationship with the core (usually estimated from travel-to-work commuting flows). The ranking index also aims to take into account urban form and population density by determining whether a city is located in a megaregion (also called megalopolis or the megapolitan area). Megaregions as a concept (Gottman, 1961) are defined as networks of metropolitan regions with shared economies, infrastructure and natural resource systems, stretching over distances of roughly 300–600 miles in length. This term is rarely used in Europe, as in most cases each European country can be functionally considered as an independent megaregion. As can be seen, the number of variables associated to the features of each city (urban structure, transit connection and population size) is greater than the combined variables associated to the corridor itself: distance, combined economic variable and combined congestion index. Distance is an indirect way to measure travel time and, in relation to this variable, Hagler and Todorovich established that distances below 100 miles were better suited for private car and commuter rail networks whereas distances greater than 500 miles were more efficiently travelled by air. This index weighted the distance criteria such that it peaked between 200 and 300 miles and decreased to zero after 500 miles. The value begins at 2 for corridor lengths of 100 miles, increases linearly and peaks at 2.5 for corridor lengths between 150 and 300 miles, decreases linearly to 2 at lengths of 350, then decreases to 1.5 and continues decreasing linearly to a value of 0 for lengths of 500 miles. One important point is that this combined variable ignores the accessibility time to the HSR station in each city in the Origin–Destination pair. This accessibility time, added to the HSR travel time (dependent on speed and distance), gives the total travel time and is one of the key variables that conditions HSR demand. The model proposed by Hagler and Todorovich was applied to the US, and the final scores for the 27,000 city pairs ranged from 3.9 to 44.9. To aid understanding the scores of the final city pairs are expressed as a percentage of the top score (100). The results were consistent with an intuitive a priori assessment: high population density US regions would lead the top of the ranking pairs and, as expected, the top 50 city pairs identified were primarily concentrated in the Northeast, California, and the Midwest. As there is no HSR in the US, there is no ranking based on current HSR traffic data for comparison and therefore no possibility for a proper discussion of the results. Spain offers a good opportunity for the application of the Hagler and Todorovich model, and Guirao and Campa (2014a) validated and discussed this multi-criteria model in Spain, where current HSR traffic allows the detection of potential deficiencies. The direct application of the ranking methodology to the Spanish case raises several difficulties, but the only way to validate and discuss the model, in the first stage, was by attempting its direct application. In order to use similar criteria to the US model, Spanish cities with over 50,000 inhabitants were selected, and city pairs were created by connecting each city to every other city located between 100 and 500 miles (160 km and 800 km) from the origin city. Spain is administratively divided into 17 regions and 50 provinces, and all the capitals of the provinces were considered (after removing cities on Spanish islands). In Spain, two cities that were not capitals of provinces were included in the database because their population was greater than the capital itself: Jerez de la Frontera and Gijón. This selection process yielded 49 cities and 1176 city pairs across Spain on which the analysis was based. Adopting the criteria and definition of Gottman, Guirao and Campa

B. Guirao, J.L. Campa / Journal of Transport Geography 47 (2015) 37–46

considered Spain, due to its size and functionality, as a single megaregion. And finally, as the model was not devised for use with metric units, Spanish distances were converted into miles, and the values in the GDP variable were converted into 2001 dollars. Table 2 shows the top 50 Spanish city pairs obtained by applying the Hagler and Todovich model with no changes in the equation or the values of the variables adopted. The connections between the three most populated cities (Madrid, Barcelona and Valencia) appear in the top ten of the ranking, and their scores show a considerable difference (up to 3.0) compared to the following city pairs (the average difference between subsequent city-pair scores is 0.41). Fig. 1a shows the current Spanish HSR lines in operation and their corresponding opening date, while Fig. 1b shows a proposal for a theoretical phasing of construction based on the modelling results. The present network covers the top ten city pairs with the exception of four important missing links: Barcelona–Valencia, Madrid–Bilbao, Barcelona–Bilbao and Madrid–Murcia. First, the model was validated by comparing these results with the ranking derived from the current HSR network operation (Ministerio de Fomento, 2013). This ranking was obtained by recording traffic in each city pair with a HSR link, and only considering pairs in the top 50 theoretical ranking. Table 3 shows the city pairs according to their position in the model ranking, indicating distance, travel time and annual traffic recorded in 2011 (2013 traffic has also been included). Traffic can be seen to decrease progressively down the ranking, and Madrid–Barcelona continues to be the top origin–destination pair with more than 2.5 million passengers. In general terms, the results can be assumed to be consistent with recorded traffic, and the proposed model – which focuses mainly on the metropolitan areas size and transit offer – could be used as a tool in a HSR network planning process. This approach offers other advantages such as simplicity and ease of application, and although it requires the availability of a detailed database, this can usually be obtained from the National Statistics Institutes in

Table 2 Top 50 HSR city pairs in Spain according to the results obtained using Todorovich and Hagler’s model. Rank

City pair

Score

Rank

City Pair

Score

1 2 3 4 5 6 7 8 9 10

Madrid–Barcelona Barcelona–Valencia Madrid–Valencia Madrid–Bilbao Madrid–Sevilla Madrid–Zaragoza Barcelona–Bilbao Madrid–Murcia Madrid–Malaga Barcelona– Zaragoza Madrid–Gijón Barcelona–Murcia Madrid–Alicante Barcelona–Alicante Madrid–Vitoria Madrid–San Sebastián Barcelona–San Sebastián Madrid–Santander Barcelona–Vitoria Valencia–Zaragoza Barcelona– Pamplona Madrid–Pamplona Valencia–Bilbao Madrid–Burgos Madrid–Valladolid

100.00 96.73 96.73 93.01 92.81 90.50 90.44 89.81 89.53 89.44

26 27 28 29 30 31 32 33 34 35

Valencia–Murcia Valencia–Sevilla Madrid–A Coruña Salamanca–Madrid Madrid–Granada Madrid–Almería Barcelona–Castellón Madrid–Castellón Madrid–Cordoba Madrid–Lleida

83.95 83.78 82.90 82.82 82.51 82.14 82.14 82.14 82.14 82.14

88.21 88.04 87.84 87.41 86.93 86.77

36 37 38 39 40 41

Madrid–Logroño Barcelona–Santander Barcelona–Burgos Madrid–Tarragona Valencia–Málaga Madrid–Albacete

82.14 81.86 81.62 81.46 81.23 81.07

86.16

42

Barcelona–Logroño

81.07

85.71 85.24 84.64 84.27

43 44 45 46

81.07 80.98 80.13 80.13

84.27 84.23 84.21 84.00

47 48 49 50

Madrid–León Valencia–Alicante Barcelona–Valladolid Madrid–Jerez de la Fontera Madrid–Badajoz Madrid–Huesca Madrid–Teruel Sevilla–Malaga

11 12 13 14 15 16 17 18 19 20 21 22 23 24 25

80.01 80.01 80.01 79.93

41

each country Nevertheless, Table 3 also reveals some deficiencies in the ranking list that require explanation. Madrid–Valencia is second in the ranking, but the recorded traffic in 2011 was lower than for Madrid–Seville (position 4); this may be for two main reasons. First, the Madrid–Seville line opened in 1992, and Madrid– Valencia in 2010; this latter connection had probably not yet reached its ‘‘maturity’’. Seville also has a considerable tourism attraction factor, and the model only considers (for each metropolitan region) population, transit and per capita GDP. These conclusions can be extended to another poorly-scoring connection, Madrid–Cordoba (ranking position 34), with 800,679 passengers. Tourism is clearly a trip attractor variable, and particularly in countries where tourism is one of the main contributions to national GDP (over 10% in Spain). Improving accessibility to a tourist city can be made a priority in the process of planning a new HSR network. It is important to identify the predominant economic activity of a metropolitan area (not simply its per capita GDP) as a demand attractor. Tourism is a clear example in Spain and Toledo and Segovia (in Spain) are good examples of this tourism-based approach. Madrid–Toledo was not considered by the model mainly due to its proximity to Madrid (less than 160 km), although the size of the city (almost 78,000 inhabitants) would also condition its ranking position. However, HSR traffic in 2011 was 1,497,660 passengers, a higher figure than for city pairs in positions above 15 in the ranking. Toledo is a mid-sized city in central Spain, 70 km south of Madrid, and was declared a World Heritage site by the UNESCO in 1986 for its extensive cultural and monumental heritage. The case of Toledo highlights another deficiency in the model: cities located within a 160 km radius from the centre of a major metropolitan area are not included in the database. In the Hagler and Todorovich model, city pairs were created using a geographic information system (GIS), connecting each city to all other cities located between 100 and 500 miles (160 km. and 800 km.) from the origin city. Distances of less than 200 km are equivalent, in terms of HSR, to approximately 1 h or less travel time and this scenario enables its use by commuters. Spain has many examples of HSR services that are used as commuter rail (commercially known as HSR Medium-Distance services or AVANT services), and they have high passenger demand (Ureña, 2012). One important feature of these medium-distance services is the use of slower and cheaper HSR shuttles (260 km/h maximum speed trains) running frequently – with between 6 and 12 services per day – at lower fares. Single ticket fares for these medium-distance services are 0.10 euros/km (2010), almost half that of long-distance pure HSR service, and similar to medium-distance conventional rail fares. Although their frequency is lower than conventional suburban rail services, they offer similar discounts of between 20% and 50% for frequent travellers. Thanks to these advantageous operating conditions the AVANT services are often used for commuting. The consequences of eliminating city pairs in the database that are less than 160 km apart should be analysed in the case of Spain, as some of these pairs operate with HSR Medium-Distance services. In conclusion, the first findings of Guirao and Campa (2014a) show the consistency of the model as a preliminary approach to ranking pairs, mainly for the top O–D relations; however (as shown above) the model fails to discriminate clearly between secondary groups of corridors, and needs improving. These deficiencies are chiefly due to the type of variables used by the model, which focus on metropolitan population size and local transit in each city and overlook other factors that affect HSR demand. To support these arguments, Guirao and Campa (2014b) conducted a sensitivity analysis, changing the weights of the variables under the four hypotheses and seeking to avoid some of the model’s weaknesses. In the first hypothesis, all the variables are given the same weight (1.0) to determine whether this modifies the top-ranked pairs. In

42

B. Guirao, J.L. Campa / Journal of Transport Geography 47 (2015) 37–46

Fig. 1. Current Spanish HSR lines in operation with their year of opening (above, Fig. 1a). Proposal for a theoretical construction phasing based on the model results (below, Fig. 1b).

43

B. Guirao, J.L. Campa / Journal of Transport Geography 47 (2015) 37–46 Table 3 Long-distance HSR traffic for the only city pairs in operation in the top 50 in 2011. Source: Ministerio de Fomento (2013). Origin

Destination

Ranking position

Distance (km)

Year service opened

HSR travel time (min)

Passengers 2011

Passengers 2013

Madrid Madrid Madrid Madrid Madrid Barcelona Madrid Madrid Madrid Madrid Madrid Seville

Barcelona Valencia Seville Zaragoza Malaga Zaragoza Cordoba Valladolid Lérida Tarragona Albacete Malaga

1 2 4 6 9 10 34 25 35 39 41 50

621 391 471 306 513 260 345 180 442 521 322 270

2008 2010 1992 2003 2007 2008 1992 2007 2003 2006 2010 2008

150 100 150 75 150 90 105 56 125 150 90 110

2,545,907 1,836,500 2,137,026 1,175,053 1,433,361 600,511 800,679 1,083,590 238,754 294,702 248,992 104,317

3,070,184 1,858,436 2,175,808 1,176,841 1,533,363 623,555 757,673 1,212,632 231,582 300,918 238,495 96,480

the second, more emphasis is given to the combined variables, while the third hypothesis awards the same importance to all the transit modes. Lastly, in the fourth hypothesis all the weights are modified in order to obtain the same level of importance per block of variables (transit, population and combined variables). This weighting analysis, under various hypotheses applied to the Spanish network, revealed that population variables exert an excessive and even redundant influence on the results if we consider the evident relation of dependence between the size of a population and the length of its transit network. It is true that the election of the variable weightings in the original model (proposed by Hagler and Todorovich) appears controversial, and that the ideal and more rational scenario is one in where the weighting number of each variable is determined by an expert panel evaluation approach. It is difficult to find results for this type of evaluation in the literature, and we should note that despite its weaknesses, this tool offers other advantages such as simplicity, ease of application and consistency of the results. The Hagler and Todorovich methodology has been recently extrapolated to other countries apart from Spain, such as Malaysia (Saat and Serrano, 2015) as part of the new HSR network planning process. Another point to consider is whether the model should be validated using panel data and not only 2011 data. In the case of Spain, the use of panel data would entail a major problem: the HSR network was built in sequence and there are only a few years with complete HSR traffic data on all the corridors. Table 4 now shows the evolution of the ranking of the HSR corridors from 2005 to 2013, and the stability of this ranking from the year 2011. Working with panel data prior to 2011 would have involved taking into account conventional rail traffic for some corridors, adding complexity to the interpretation of the results. We should also consider that this equation ranks pairs, and compares city pairs inside the same country; the evolution of some model

variables such as population or GDP in these cities usually shows the same growth rate. The points mentioned above have encouraged the authors of this paper to continue their research work on this ranking tool and improve the original equation. The sensitivity analysis also enabled to present some specific suggestions for the generalised extrapolation of the methodology, including the definition and introduction of new variables such as tourism. Significant progress has been made with this new approach, which is described in the section below. 4. Effects of the tourism variable on the ranking model The type of tourism variable needed in the model is relatively dependent on the availability of a tourism database in each country. The ranking model structure requires a variable that represents the tourism value of each city considered in our research (at the provincial capital level) and – as tourism is a compound product – one that should consider both the attraction capacity exerted by the city’s offering of leisure and business tourism products, along with the urban environment and local life. In Spain, the most comprehensive study on the tourism competiveness of Spanish cities is UrbanTUR 2012 (Exceltur, 2013). This document is the first to show the 20 most important Spanish urban destinations, obtained by assessing 57 city indicators classified into six categories (see Fig. 2, with the explanation of the ranking position of the city of Toledo): – Category 1: Attraction capacity offered by leisure products – Category 2: Attraction capacity offered by business tourism products – Category 3: Competitive conditions offered by the urban environment and local life

Table 4 Evolution of the corridor ranking in Spain according to real long-distance HSR traffic data (2005–2013). Source: compiled from data from the Ministry of Public Works (2013). Origin

Destination

2005

2006

2007

2008

2009

2010

2011

2012

2013

Madrid Madrid Madrid Madrid Madrid Madrid Madrid Barcelona Madrid Madrid Madrid Madrid

Barcelona Seville Valencia Málaga Zaragoza Córdoba Alicante Zaragoza Valladolid Tarragona Albacete Lérida

6 1 5 7 2 3 4 10 11 12 9 8

6 1 5 7 2 3 4 10 12 11 9 8

4 1 6 7 2 3 5 11 12 8 10 9

2 1 6 4 3 5 8 9 7 10 12 11

1 2 6 3 4 5 7 8 9 10 12 11

1 2 6 3 4 5 7 8 9 10 12 11

1 2 3 4 5 6 7 8 9 10 11 12

1 2 3 4 5 6 7 8 9 10 11 12

1 2 3 4 5 7 6 8 9 10 11 12

Notes: The italicised and bolded cells indicate the year the HSR service opened. Prior to this, the traffic used to calculate the ranking corresponds to the previous conventional rail line.

44

B. Guirao, J.L. Campa / Journal of Transport Geography 47 (2015) 37–46

Fig. 2. Example of the basis (categories) for Toledo’s position in the URBANTUR ranking.

Table 5 UrbanTUR ranking for the top 20 urban Spanish destinations. City

Ranking 2012

Score

Ciudad

Ranking 2012

Score

Barcelona Madrid Valencia Sevilla San Sebastián Málaga Bilbao S. de Compostela Zaragoza Granada

1 2 3 4 5 6 7 8 9 10

141.4 139.1 111.0 104.3 103.1 98.3 97.9 97.3 96.6 95.9

Salamanca Gijón Córdoba La Coruña Alicante Santander Toledo Burgos Oviedo León

11 12 13 14 15 16 17 18 19 20

95.7 94.2 93.9 92.4 91.9 91.1 90.8 88.0 86.4 85.7

even redundant influence on the results if we consider the marked relationship of dependence between the size of a population and the length of its transit network. In view of this, the tourism variable (T) has been given a weight of 10 in the model (see Eq. (2)) in order to balance the importance assigned to population variables. The maximum score for the tourism variable – 3.0 (similar to the maximum score for the rest of the variables) – was assigned to Barcelona (see Table 4). The last city in the ranking – León – was assigned 1.0; and all other Spanish cities that do not appear in the UrbanTUR ranking were assigned a value of 0.0. Cities on the UrbanTUR list have been given a linear value between the maximum 2.0 and the minimum 1.0, according to their score in the UrbanTUR ranking.

RI ¼ ðCRÞ þ 0:5ðLRÞ þ 0:5ðS LR Len IÞ þ 0:5ðHRTÞ – Category 4: Accessibility and mobility – Category 5: Administration and strategic management – Category 6: Economic and social results Table 5 shows the final ranking obtained for the 20 previously selected cities based on the scores for each category in each city. One of the drawbacks of the UrbanTUR study is that it only considers 20 Spanish cities, two of which are not provincial capitals (Santiago de Compostela and Gijón). Our prioritization model works with Spanish provincial capitals as origins and destinations; the capital of a province is usually the most populated and important city. In Spain, two cities that are not provincial capitals were included in the database, as their population was greater than the capital itself: Jerez de la Frontera and Gijón (see Section 3). To adapt the tourism ranking to our model database, and as Santiago de Compostela is only 55 km from La Coruña – an insignificant distance for a HSR link – we have assigned La Coruña (included in the model database as a provincial capital) the same tourist score as Santiago de Compostela. Our aim was to assign a tourist score to each city in the model database. Gijón was not a problem, as it was included in the original model database from the start, and both cities in the province (Gijón and Oviedo, the provincial capital) are part of the UrbanTUR ranking. Another important issue to discuss is the weight given the tourism variable to be added to the model and the values assigned to it. In the sensitivity study conducted for the initial model, Guirao and Campa (2014b) calculated the percentage of influence of each variable on the RI (Ranking Index) and on the maximum score for each block of variables (population, transit or combined variables). They reported that population variables can account for up to 70% of the RI value, while transit variables and combined variables (length, combined GDP and combined congestion) account for only 18%, and 12% of the total RI respectively. From this point of view, the level of influence of the combined variables is very low, and their evident insignificance does not seriously affect the final results. However the population variables have an excessive and

þ 0:5ðS HR Len IÞ þ ðMet PopÞ þ 10ðMetro MainÞ þ ðCity popÞ þ ðMegaÞ þ 10ðTÞ þ ðCR 1Þ þ 0:5ðLR 1Þ þ 0:5ðE LR Len IÞ þ 0:5ðHRT 1Þ þ 0:5ðE HR Len IÞ þ ðMet Pop 1Þ þ 10ðMetro Ma 1Þ þ ðCity pop 1Þ þ ðMega 1Þ þ 10ðT 1Þ þ ðC LengthÞ þ ðG GDP ScalÞ þ ðTTI IndÞ

ð2Þ

One of the deficiencies detected in the original model when applied to the Spanish case, was that corridors below 160 km. were ignored. As discussed in Section 4, this fact was important in case of the existence AVANT services (high frequency and low prices) in which commuter traffic can arise. According to Ureña (2012), these Medium-Distance HSR services connect a few regional city systems (type 1) and some small and medium cities with metropolitan areas (type 2). Today in Spain there are only six services of this profile using approximately half the HSR network. The first to be established was metropolitan (type 2): Madrid–Ciudad Real– Puertollano. The second was also this type (Madrid–Toledo). The third was regional, in Andalusia (Seville–Cordoba–Puente Genil– Antequera–Malaga), and the other three are mixed: Madrid–Sego via–Valladolid, Madrid–Catalayud–Zaragoza, Barcelona–Tarragon a–Lérida. The Spanish HSR experience shows that cities located within a 160 km radius of the centre of a major metropolitan area should be included in the database and appear in the ranking result. There are also a number of tourist cities in this situation (Toledo, Salamanca, Burgos or Segovia) and in order to consider them, corridors with lengths of below 160 km have been introduced in the model database, and their distance variable has been given the value of 0.0. Table 6 shows the main results of the model after applying these changes to the original equation: the top 50 HSR city pairs in Spain according to the results obtained using a tourism variable and considering pairs O–D separated by less than 160 km. If we compare these results with current HSR network traffic (see Table 3), the tourism variable has clearly corrected some previous

B. Guirao, J.L. Campa / Journal of Transport Geography 47 (2015) 37–46 Table 6 Top 50 HSR city pairs in Spain according to the results obtained using a tourism variable and considering O–D pairs separated by less than 160 km. Rank

City pair

RI

Rank

City pair

RI

1 2 3 4 5 6

Madrid–Barcelona Barcelona–Valencia Madrid–Valencia Madrid–Seville Madrid–Bilbao Barcelona–Bilbao

106.31 97.21 96.59 92.95 91.32 90.73

26 27 28 29 30 31

81.12 80.30 80.18 79.63 79.47 78.59

7

Barcelona–San Sebastián Barcelona–Zaragoza

90.12

32

Barcelona–Toledo Madrid–Toledo León–Barcelona Valencia–Bilbao Zaragoza–Valencia Valencia–San Sebastián Valencia–Málaga

89.91

33

76.49

Madrid–Málaga Madrid–San Sebastián Madrid–Zaragoza Madrid–Gijón

89.79 89.79

34 35

Valencia–Alicante/ Alacant Oviedo–Madrid Sevilla–Málaga

89.79 88.07

36 37

75.31 75.30

Barcelona–Alicante/ Alacant Madrid–Alicante/ Alacant Madrid–Coruña, A Madrid–Santander Madrid–Salamanca Madrid–Granada Madrid–Córdoba Barcelona– Santander Madrid–Burgos

87.69

38

Valencia–Granada Valencia– Salamanca Gijón–Valencia

87.28

39

Valencia–Córdoba

74.43

86.41 86.06 85.94 85.85 85.14 84.87

40 41 42 43 44 45

74.06 74.02 73.70 72.49 72.23 72.17

84.52

46

83.92 83.91

47 48

24

Burgos–Barcelona Salamanca– Barcelona Madrid–León

Valencia–Santander Valencia–Burgos Zaragoza–Bilbao Valencia–Toledo Sevilla–Granada Zaragoza–San Sebastián Sevilla–Alicante/ Alacant Sevilla–Salamanca Gijón–Bilbao

82.43

49

71.70

25

Valencia–Seville

81.14

50

San Sebastián– Bilbao Gijón–Sevilla

8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23

78.33

75.97 75.91

75.26

72.09 72.00 71.97

71.55

deficiencies, although the three top pairs remain the same: Madrid–Barcelona, Barcelona–Valencia, Madrid–Valencia. The fourth pair has changed. When the tourism variable is considered, Madrid–Bilbao is replaced by Madrid–Seville, as Seville has a higher tourism score than Barcelona in the UrbanTUR ranking. Madrid–Zaragoza goes from 6th to 11th position in the ranking, while Madrid–Málaga remains the same (9th position), and traffic validates this fact (Madrid–Malaga has higher HSR traffic than Madrid–Zaragoza). Other cities are clearly disadvantaged by the UrbanTUR study (like Vitoria or Murcia): pairs such as Madrid– Murcia, Barcelona–Murcia or Madrid–Vitoria are relegated from the top HSR ranking. Other cities like Toledo are included on the list for the first time, and traffic figures validate this inclusion. Madrid–Cordoba also improves its position in the ranking list, as do other cities like Santander and Burgos. As can be seen, this model explains the relationship between HSR and tourism more effectively: tourism is a positive HSR demand factor and should be taken into account. One of the determining factors for using this approach and extrapolating this methodology to other countries is the availability of a database on the tourism competitiveness of cities; even the Spanish UrbanTUR has its limitations, and only 20 Spanish cities are included in the study. This database does not allow the model to establish any differences between the cities outside this top ranking (like Segovia, Cuenca or Gerona), which are also attractive to tourists. Returning to the specific case of Toledo, the authors wish to highlight some other considerations. The Madrid–Toledo corridor is successful in terms of traffic because of its HSR link to Madrid and its tourist attractiveness, and also due to the type of service provided by the operating company. Since 2005, Toledo has had

45

AVANT services with over ten daily HSR shuttles (30 min travel time), and is also favoured by the availability of monthly tickets which are economically very advantageous compared to ordinary one-way tickets. If the schedules are compatible with working timetables (as is in fact the case), this frequency enables their use for commuting. The high traffic in Toledo is due mainly to a mix of commuting and tourism, but the commuting variable is not considered by the current model equation. Over 60% of the demand between Madrid and Toledo is from commuters and not tourists, and there are also other important HSR commuter links with high traffic such as Madrid–Valladolid or Madrid–Ciudad Real. As a line of future research, the authors consider it necessary to introduce a commuting variable in the model (for cities less than 200 km from major metropolitan areas). This variable will help separate two types of travel demand in the model: tourism and commuting. With the current equation, the model only considers the first, and some medium-distance HSR pairs with high traffic are missing (like Madrid–Ciudad Real). However, the improvements to the model are evident; it has been empirically demonstrated that tourism conditions HSR demand, and that the implications for tourism should be studied in detail and at the local level. 5. Conclusions The lack of empirical methodologies to establish the crossover effects between HSR and tourism has become an important issue in countries that already have HSR systems and – for financial reasons – also for countries that do not. After a discussion of the different approaches to project assessment, a multi-criteria selection methodology was designed to define which HSR corridors should be built first in a new network, taking into account the tourism variable. Apart from the economic geography and destination choice models, the existing literature on tourism demand is dominated by econometric models which tend to follow a single-equation time-series approach. However since the construction of new HSR lines there has been little research into methodologies to rank HSR corridors (origin–destination O–D pairs) taking into account a tourism variable. The authors of this paper have used an existing prioritisation mode to include a tourism variable, and the results have been validated with the current Spanish HSR network. In the original model equation, corridors with lengths under 160 km were ignored, and the number of variables associated to the features of each city (urban structure, transit connection and population size) prevailed over the combined variables associated to the corridor itself: distance, combined economic variable and combined congestion index. This approach stressed the functional structure of the two cities over the analysis of their interaction, with the result that the first modelling results failed to discriminate clearly between secondary groups of corridors (although not the top-ranking pairs). With the introduction of a tourism variable, one of the model’s weaknesses – the identification of the predominant economic activity in a metropolitan area (not only its per capita GDP) – is amended, as tourism is a demand attracting factor. It was not easy to select the appropriate tourism variable to use in the model, as it depends substantially on the availability of the tourism database in each country. After analysing all the tourism databases available, the authors chose the UrbanTUR study, where the twenty top Spanish urban destinations are ranked using scores obtained by assessing 57 city indicators, following the idea that tourism is a compound concept. Using the Spanish UrbanTUR ranking and including corridors of under 160 km, the results obtained from the model clearly distinguished between secondary groups of corridors. This is validated by HSR traffic figures, and encourages the authors to study the tourism phenomenon and its transport implications in greater depth.

46

B. Guirao, J.L. Campa / Journal of Transport Geography 47 (2015) 37–46

However, the UrbanTUR database also has its weaknesses, such as the fact that it is limited to twenty Spanish cities. With this database, the model cannot establish any difference between non-top-ranked cities (such as Segovia, Cuenca and Gerona), which are also attractive to tourists and may be on a HSR network. Moreover, there is still an unresolved issue in the ranking model that may affect tourism results, namely the way medium distance HSR services are taken into account. As part of future research and in order to extrapolate this methodology to other countries, the authors consider it necessary to introduce a commuting variable in the model (for cities less than 200 km from major metropolitan areas). This variable will help separate two types of travel demand in the model – tourism and commuting – and be of crucial importance in empirically discriminating some corridors. Finally, recommendations for any new HSR network planning process include setting down the main targets and priorities before ranking the potential corridors, and conducting a study of the previous transportation system. For some countries and cities, tourism can represent a real priority and source of income, and the application of this validated methodology will ultimately provide authorities and policymakers with a useful tool when planning the construction of a new HSR network or assessing an existing one. References Bao, J.G., Chu, Y.F., 1999. Tourism Geography, 6. Higher Education Press, p. 152. Bazin, S., Beckerich, C., Delaplace, M., 2011. High speed railway, service innovations and urban and business tourism development. In: Sarmento, M., Matias, A. (Eds.), Economics and Management of Tourism: Trends and Recent Developments. Collecçao Manuais, Universidade Luisiada Editora, Lisboa, Portugal. Bristow, A.L., Nellthorp, J., 2000. Transport project appraisal in the European Union. Transp. Policy 7, 51–60. Caccomo, J.L., Solonandrasana, B., 2001. L’innovation dans l’industrie touristique, enjeux et strategies. L’Harmattan, Paris. Chen, Z., Haynes, K.E., 2012. Tourism industry and High Speed Rail, is there a Linkage: evidence from China’s High Speed Rail development. In: ASRDLF, 2012 Conference special session on High Speed Rail, Tourism and Territories, 9–11th July, Belfort, France. Coto-Millan, P., Inglada, V., Rey, B., 2007. Effects of network economies in highspeed rail: the Spanish case. The Ann. Reg. Sci. 41 (4), 911–925. Crampon, J.L., 1966. A new technique to analyse tourism markets. J. Marketing 30– 2, 27–31. Delaplace, M., Pagliara, F., Perrin, J., Mermet, S., 2014. Can high speed foster the choice of destination for tourism purpose? Proc.-Social Behav. Sci. 111 (2014), 166–175. Exceltur, 2013. UrbanTUR 2012. Monitor de competitividad turística de los destinos urbanos españoles. Gottman, J., 1961. Megalopolis: The Urbanized Northeastern Seaboard of the United States. Twentieth Century Fund, New York. Guirao, B., Campa, J.L., 2014a. The construction of a HSR network using a ranking methodology to prioritise corridors. Land Use Policy 38, 290–299.

Guirao, B., Campa, J.L., 2014b. A methodology for prioritising HSR corridors: from U.S. theory to Spanish practice. J. Transp. Geogr. 35, 95–106. Guirao, B., Soler, F., 2008. Impacts of the new high-speed service on small touristic cities: the case of Toledo. In: The Sustainable City V. Urban Regeneration and Sustainability. Wessex Institute of Technology, pp. 465–473. Hagler, Y., Todorovich, P., 2009. Where high speed rail works best? Report America 2050 Association. Hayashy, Y., Morisugi, H., 2000. International comparision of background concept and methodology for transport project appraisal. Transp. Policy 7, 73–88. Krugman, P., 1991. Increasing returns and economic geography. J. Political Econ. 99 (9), 483–499. Lim, C., 1997. Review of international tourism demand models. Ann. Tourism Res. 24–4, 835–849. Macharis, C., Bernardin, A., 2015. Reviewing the use of multicriteria decision analysis for the evaluation of transport projects: time for multi-actor approach. Transp. Policy 37, 177–186. Masson, S., Petiot, R., 2009. Can the high-speed reinforce tourism attractiveness? The case of the high-speed rail between Perpignan (France) and Barcelona (Spain). In: Technovation, vol. 29, pp. 611–617. Ministerio de Fomento, 2013. Observatorio del Ferrocarril en España. Fundación de los Ferrocarriles Españoles y Ministerio de Fomento. Madrid, 2013. Morley, C., 1992. A microeconomic theory of international tourism demand. Ann. Tourism Res. 19 (2), 250–267. Ni, J., Pradayrol, J.P., Arduin, J.P., 1994. Evaluation économique d´un project de ligne a grande vitese. Revue Generale de Chemins de Fer 6–7, 13–18. O’Hagan, J.W., Harrison, M.J., 1984. Market shares of US tourism demand in Europe, an econometric analysis. Appl. Econ. 16, 919–931. Ortúzar, J. de D., Martínez, F.J., Varela, F.J., 2000. Stated preferences in modelling accessibility. Int. Plan. Stud. 5, 65–85. Ortúzar, J. de D., Willumsen, L.G., 2001. Modelling Transport, third ed. Willey, New York. Pagliara, F., Delaplace, M., Vassallo, J.M., 2015a. High-speed trains and tourists: what is the link? evidence from the French and Spanish capitals. Urban Transp. XX 138, 17–28. Pagliara, F., Lapietra, A., Gomez, J., Vassalo, J.M., 2015b. High speed rail and the tourism market: evidence from the Madrid case study. Transp. Policy 37, 187– 194. Prideaux, B., 2000. The role of the transport system in destination development. Tourism Manage. 21 (1), 53–63. Quinet, E., 2000. Evaluation methodologies of transportation projects in France. Transp. Policy 7, 27–34. Rothengatter, W., 2000. Evaluation of infrastructure investments in Germany. Transp. Policy 7, 17–25. Rugg, D., 1973. The choice of journey destination: a theoretical and empirical analysis. Rev. Econ. Stat. 55 (1), 64–72. Ryder, A., 2012. Viewpoint: high speed rail. J. Transp. Geogr. 22, 303–305. Saat, M.R., Serrano, J.A., 2015. Multicriteria high-speed rail route selection: application to Malaysia’s high-speed rail corridor prioritization. Transp. Plan. Technol. 38 (2), 200–213. Sinclair, T., Stabler, M., 1997. The Economics of Tourism. Routledge, London. Song, H., Li, G., 2008. Tourism demand modeling and forecasting – a review of recent research. Int. J. Hospital. Manage. 22–24, 435–451. Song, H., Wong, K.F., 2003. Tourism demand modelling: a time varying parameter approach. Tourism Manage. 23, 475–487. Ureña, J.M., 2012. Territorial Implications of High-Speed Rail: A Spanish Perspective. Ashgate. Wang, X., Huang, S., Zou, T., Yan, H., 2012. Effects of the high speed rail network on China’s regional tourism development. Tourism Manage. Perspect. 1, 34–38.