Indirect and wider economic impacts of High-Speed Rail

The paper considers the development of high speed rail in Europe in terms of its wider impacts on economic development. It shows how the focus has changed from the more limited objectives of enhancing rail capacity to a recognition of the potential for use as an instrument for enhancing competitiveness and cohesion within the context of the development of the trans-European networks. This however raises a number of complex issues relating to the way transport interacts with the economy, particularly in a world of imperfect competition. The paper reviews attempts to evaluate these benefits from simple accessibility and potential models to recent developments of computable general equilibrium models. Introduction Indirect impacts, especially on regional economic performance, are often used as a justification for additional benefits from high-speed rail (HSR) projects – sometimes to make the difference between a project which is not justifiable on a strict user benefits basis and viability. Regional authorities have been especially vocal in using this argument in favour of projects, for example regional claims for the priorities in the development of the Schéma Directeur LGV in France and those made for both the Madrid-Sevilla and Madrid-BarcelonaFrench border lines in Spain. Similar arguments have been made in Italy and have surfaced in the recent discussion for a new north-south line in the UK. The usual assumption is that the improvements in accessibility which the creation of an HSR link makes will enlarge market areas and increase the implicit competitiveness and productivity of firms in a newly connected region. In addition the time savings made by all travellers are equivalent to an increase in productivity. These assumptions tend to lead to three outcomes: an overestimation of potential demand which inflates predicted user benefits; an overestimation of non-user or induced benefits; an assumption that all potential non-user benefits accrue to one region, ignoring any redistribution which the changing pattern of accessibility brings about. More recent theories of the role of infrastructure and transport improvements in regional development have stressed the way that transport costs (and hence accessibility) interact with other determinants of economic development, particularly scale economies and the size of market areas, in an imperfectly competitive world. Improvements in transport may thus benefit firms in more developed core regions more than those in less developed peripheral regions. Thus transport improvements to (and within) core city regions not only provide a 1 This paper has been prepared as part of a project on An Economic Analysis of European High Speed Rail Infrastructures, coordinated by Gines de Rus and funded by the BBVA Foundation. First draft – not for quotation 2 direct benefit in terms of the enhanced productivity of existing workers and an increase in employment, but through agglomeration effects they raise the productivity differential of the core city relative to the rest of the economy. This reflects the positive relationship between city size and productivity. Transport improvements may thus be as likely to lead to an increase in regional disparities as they do to increasing cohesion. This is not a universal or inevitable outcome, it will depend on the specific situation of the region, the initial levels of accessibility and the change in them and the existence of other policy measures which may accompany the transport improvement. Most analysis tends to have been undertaken of individual links of HSR developments, or at most of what are in most cases simple national networks. As the networks have developed they have begun, in both north-west Europe and southern Europe, to form international networks. This poses new issues for analysis and appraisal. In this paper we begin to address these issues by looking at the evidence on the impacts on development the emerging European HSR network has had. We look in particular at evidence on the links between changes in accessibility and changes in regional economic activity for a selection of regions which have benefited from the introduction of HSR services. In doing this we identify some of the limitations of existing modelling approaches. A particular focus is on the relationship between HSR networks and regional and local transport networks and the role of accompanying policies towards economic development. We also identify the way that some intermediate regions may suffer from the introduction of HSR services which form a corridor through the region with little or no benefit and often considerable costs. The broad conclusion from the paper is that HSR can be an element in improving the economic performance of regions, but there is no guarantee that all the impacts on any one region are positive. This leads to some suggestions for improvements to the techniques of appraisal for HSR projects. The Emerging European HSR Network High-speed rail in Europe has had a 25 year period of development from the inauguration of the first French TGV service between Paris and Lyon in September 1981. From slow beginnings national networks have emerged in France, Germany, Italy and Spain. An international network bringing in Belgium, UK and Netherlands to form the so-called PBKAL (Paris-Brussels-Köln-Amsterdam-London) will be complete in 2007 and further international links are planned or under construction to links the French network to Germany (via TGV Est), Italy (via Lyon-Turin) and Spain (via Perpignan-Barcelona). This network (and the rationale for it) has emerged a little haphazardly. The original lines in France, Germany and Italy were seen largely as a means of overcoming bottlenecks on the original classic rail network. Such bottlenecks limited capacity, caused conflicts between types of traffic and imposed unpredictability. Increased speed was in many respects an accidental by-product of improved reliability. Constraints on improving existing, often highly curved and circuitous routes through mountains or along river valleys led to the often cheaper 2 This refers to what is now conventionally thought of as HSR, operating on dedicated newly constructed track with operating speeds of 250 km/h or above. The first Japanese Shinkansen was introduced in 1964 with speeds of around 200 km/h, speeds also achieved in both France and the UK by conventional trains on conventional track. First draft – not for quotation 3 option of new construction. New construction could take more direct routes – the rail distance between Paris and Lyon was reduced by about 120 km or 20 per cent and take advantage of more efficient new traction and rolling stock to employ steeper gradients and sharper curves – there are no tunnels on the more than 400 km of LGV between Paris and Lyon. Taking more direct routes could mean that traditional centres along the route were no longer served. This applied to cities such as Dijon on the TGV-Sud Est and later similar problems arose in connection with such towns as Orléans (on TGV-Atlantique), Amiens (on TGVNord) and Metz and Nancy (on TGV-Est). Attempts to use the new lines to generate new access points to the high level networks through a range of stations such as Montchanin-Le Creusot in Bourgogne, TGV Picardie (between Amiens and St Quentin) and TGV Lorraine have not thus far been a success either in generating traffic or in providing a focus for economic development. On the other hand TGV-Est is linking Reims more closely into the high-level network and the redevelopment of Lyon La Part Dieu and Lille Euralille shows how accompanying regeneration policies can be effective. TGV Lorraine is also interesting in that although as far as 27 km from Metz and 37 km from Nancy, it is relatively close to the regional airport for these two cities. Furthermore, the use of a technology which enables HSTs to travel at normal speeds on conventional rail enables connection-free services to penetrate to large parts of the rail network. In Germany and Italy, where the lack of a dominant metropolitan focus has led to a more diffuse development of HSR to overcome bottlenecks without creating a national network, this focus on the major centres has been less strong until recently. This the first German NBS (Neubaustrecke) provided parallel routes over the links between Göttingen and Würzburg and Mannheim and Stuttgart avoiding particularly slow and circuitous routes. Similarly the first Italian links sought to improve journey times on the core Milano-Bologna-Firenze-Roma route where again the traditional route suffered from speed penalties due to the topography. Second and third generation developments have led to a greater concentration on key intercity routes rather than by-passing natural barriers, for example the Frankfurt-Köln line and improvements to lines serving Berlin in Germany and the eventual moves to completing a Tshaped network in Italy serving all the major cities. The emphasis has now shifted to more detailed improvements to the network in terms of smaller ABS (Ausbaustrecke) The success of HSR in generating traffic and in switching traffic from both air and road led to an increasing desire for regions not to be excluded from the emerging network. Although there had a been an overall national network for France, through a developing Schéma Directeur, and ones began to emerge in Germany, Italy, and more lately in Spain, this was largely to provide a ranking of projects to determine the likely order of construction according to rates of return. The early projects carried sufficiently large estimated rates of return on a financial basis to make them worthwhile, later projects would increasingly rely on the need to identify a wider social benefit to justify the use of public funds. Moreover as a network began to emerge, it became more difficult to identify a precise added value for each new link as the network effects became more complex. Experience with the first HSR in France, TGV-Sud Est, showed that for distances of around 400-450 km and a two hour typical journal time, HSR was not just competitive with air, it 3 It was estimated that the cost of the new Paris-Lyon LGV would be about 40% of the cost of upgrading the classic route, even where this was feasible First draft – not for quotation 4 would completely dominate the market. Even at the 600-800 km range, HSR could be highly competitive given access times, frequencies and reliability. HSR thus started to be seen not just as a competitor to air for inter-regional journeys, but as a complement for longer international and inter-continental journeys. Thus interconnection of the HSR network with airports became seen as a core design feature. The French network serves Paris Charles de Gaulle and Lyon St Exupéry (Satolas) directly and a number of smaller regional airports by close proximity. The Frankfurt-Köln line serves both Frankfurt (Main) and Köln-Bonn airports. At a European level, the emergence of the concept of trans-European networks (TENs) during the 1990s led to the designation of an HSR network independent of that for classic rail. To some extent this was simply an amalgamation of the four main national networks already referred to, but it went further in two main areas: the designation of Northern and Southern European HSR networks. The Northern network is the star shaped link of the PBKAL; the Southern network provides links between the Spanish, French and Italian networks. These have (or still are) being built on a link by link basis with little or no regard for the overall network effects and the by-passing and redistributive impacts are significant. The major problems of the TENs are the duplication between networks and the desire of regions to ensure they are on each network in the (mistaken) belief that there is a financial advantage. In fact the amount of central EU funding for each TEN link is relatively small and although dedication can give rise to eligibility for Structural and Cohesion Funds, national and regional governments (and the private sector) have to find the bulk of the finance and this can prove a long-term burden on the public sector’s commitments. Because the various HSR projects were in a reasonably advanced state of planning, and because the promotion of a new mode of transport which could provide a serious competitive challenge to the environmentally less attractive road and air modes (with both of which HSR competes) was an attractive proposition, HSR featured strongly in the initial list of priority projects determined at Essen in 1994. The extension of the priority list to around 30 projects following the Van Miert report of 2003 reveals that there are now six HSR TEN priority projects of which only three will have been completed by 2007 (Table 1). Of these the LyonTorino-Venezia project has been broadened to include multiple uses and arguably the West Coast Main Line in the UK is not a true HSR project as it involves upgrading rather than new build and is for maximum speeds well below 300 km/h. It will be clear from Table 1 that the priority TENs build on national links to provide international networks. In some cases this is because the international linkage underwrites the viability of the national project and vice versa. For example TGV Nord only makes full economic sense in the context of Paris-Brussels and Paris-London links, although each of these would be difficult to justify in its own right without the domestic Paris-Lille service and the ability to link Brussels-London services on the same infrastructure. Some similar issues arise in the routing of TGV-Est and in the various elements of HSR South West. The problem has been that the intentional impact of HSR has so far been much less dramatic in traffic generation terms than domestic HSR services. Eurostar services between Paris/Brussels and London have only reached about one-third of the original forecast passenger numbers after 4 Paris-London via Lille is a longer route than the direct route from Paris via Amiens to the Channel Tunnel. The latter is listed as a future project in the French Schéma Directeur, but it is likely to be some time before it becomes viable. The failure to serve Amiens has remained a persistent point of contention. First draft – not for quotation 5 more than ten years of operation although the rail share of the air plus rail market has reached more than 70 per cent. Table 1: HSR Projects in Priority TENs Projects or sections of projects Date for start of operation A. Essen List Projects PP2 High Speed Railway Paris-Brussels-CologneAmsterdam London 2007 PP3 Southern TGV Madrid-Barcelona Barcelona-Figueres-Perpignan Madrid Vitoria-Hendaya 2005 2008 2010 PP4 TGV Est Paris-Baudrecourt Metz-Luxembourg Saarbrücken-Mannheim 2007 2007 2007 PP6 Lyon-Torino-Trieste-Torino-Venezia 2010 PP14 West Coast Main Line (UK) 2007 B. New 2003 Projects High Speed Railway, South-West – Lisboa/Porto – Madrid – Perpignan – Montpellier – Montpellier Nîmes – Irún – Dax (cross border) – Dax Bordeaux – Bordeaux – Tours 2011 2015 201