USING ACCESSIBILITY INDICATORS AND GIS TO ASSESS SPATIAL SPILLOVERS

Report as inappropriate Your report has been received

Shared by: qnl49935

Tags

spatial analysis, accessibility indicators, case study, spatial data analysis, luc anselin, spatial data, spatial autocorrelation, land use, spatial econometrics, climate change, transport infrastructure, spatial interaction, the case, spatial heterogeneity, social science

Stats

views:: 58
posted:: 5/25/2010
language:: English
pages:: 17

Public Domain

Document Sample

USING ACCESSIBILITY INDICATORS AND GIS
TO ASSESS SPATIAL SPILLOVERS
OF ROAD PRICING POLICY

Javier Gutiérrez, Universidad Complutense de Madrid, Madrid, Spain,
Juan Carlos García-Palomares, Universidad Complutense de Madrid, Madrid,
Spain,
Ana Condeço, Universidad de Alcalá, Madrid, Spain.

Abstract.- This paper analyzes the impact of road pricing on regional accessibility, using
accessibility indicators and GIS. The Spanish road system is used as a case of study to apply our
methodology. First we analyze the global impact that road pricing policy will have on the
regional accessibility, comparing the accessibility before and after the implementation of the
road pricing policy. Then the spillover effects of the pricing policy will be measured for each
Spanish autonomous region. The methodology applied to measure the spillovers effects is based
on accessibility indicators and calculated using GIS databases. This methodology analyzes to
what extent the pricing policy implemented in one autonomous region penalizes the
accessibility of the other Spanish territories.
The traditional methodologies used by the economists to measure spillover effects fail to
acknowledge the important dynamic consequences that new transport policies produce in the
territory. In other words, previous studies fail to take properly into account the network effects.
Using accessibility indicators to measure the spillover effects makes possible the analysis of the
spatial interaction and the intensity of spillover effects within the study area.
Spillover effects of the implementation of a pricing policy on regional accessibility are
negative as a result of the increase of generalized transport cost, but the intensity of the effects is
very different depending on the region. The spillover effects are mapped in order to analyze
their spatial distribution.
The analysis of the spillover effects may be of great utility to study the impacts of the road
pricing in regional accessibility, especially to the planners and policy makers.

Key words: Road pricing, Spillover effects, accessibility, transport infrastructure.
1. Introduction: Road pricing policy, accessibility and spillovers

Pricing policy has been a subject of research for many decades. Recently public
debate about road pricing policies has increased because pricing measures are expected
to alleviate many currently existing traffic and transport problems– in the sustainable
transport and mobility framework -. The road pricing policy intends to internalize the
external costs of road transport by requiring charging a toll. Thus it is expected that the
use of private vehicle will decrease, so that external costs of transport will decrease too.
However, road pricing policies will reduce spatial accessibility in terms of generalized
transport costs.

We cannot forget that the accessibility is the main "product" of the actions on the
transport system (Schürman, Spiekermann and Wegener, 1997). If we consider the
accessibility as the facility to reach the destinations that are important for us from one
specific location (Morris, Dumble and Wigan, 1978; Linneker and Spence, 1992),
constructing a new transport infrastructure or improving the existing infrastructure will
vary this facility and have immediate consequences on the distribution of territorial
accessibility. High accessibility leads to competitive advantages of one region compared
to the others. Furthermore, infrastructure improvements will reduce the transportation
costs and lead to higher travel demands. These new demands mean a greater
relationship (goods flows, business trips, tourist flows …) between the regions served
by the new infrastructure. Therefore, the accessibility is frequently associated with
regional development (Vickerman, Spiekermann and Wegener, 1999): regions with
good accessibility are more attractive for investment and generally more competitive,
facilitating specialization benefits and economies of scale (Forslund and Johnson,
1995). However, the relationship between transport infrastructure and regional
development is complex: the improvement of transport infrastructure can promote
economic development if the region has a favourable economic environment (positive
externalities) (Beuthe, 2002; Banister and Berechman, 2003). Besides, the effects of
transport infrastructure improvement on regional development are weak when a high
accessibility already exists in the territory (very developed and efficient transport
networks), but greater when problems of accessibility exist in the region (due to both
congestion problems or to low transportation infrastructure connectivity and density).
Accessibility analyses identify that new conditions for interaction created by the
changes in the infrastructure, determine where territorial effects are.
Compared to traditional studies of infrastructure improvements and its positive
effects on accessibility, the implementation of a pricing policy in a transport system (as
the road network) will produce accessibility reductions that will vary according to the
implemented model characteristics. Therefore, from the accessibility point of view the
impacts are expected to be negative, since generalized transport cost will increase. If the
reduction of transport cost (as a result of the improvement of transport infrastructure)
permits that the economic system can function in a more efficient manner, the
increasing transport costs (as a result of the pricing policy) should act in the opposite
direction: greater cost for transport operators and increasing prices that are derived
towards other economic sectors, producing an increase in the consumer price index. At
the same time, people will have less capacity of expenditure and therefore a reduction of
consumption is expected. But it seems that the effects of a road pricing policy on the
economic system will not be very significant and, in any case, they will depend on the
toll fee.

The appearance of differential effects on the territory is also expected. Pricing
policies that penalize the flows on high capacity roads will specially damage regions
with high motorway density (usually the most developed regions). Besides, this effect
will be accentuated if the fee is higher in the surroundings metropolitan areas (that
normally have greater levels of income than the rest of the country).From the urban-
rural perspective, it is expected that the pricing policy will affect to a greater extent the
most developed territories (with a greater density of high capacity network), than the
rural areas. Therefore, we can foresee distributive effects in the territory, since the
pricing policy will have a greater impact in the most developed regions and in urban
areas. Comparatively this would increase the attraction for investment in less developed
regions and in less densely populated areas. This means that, it would produce spread
effects and will positively influence the regional equity. Obviously, the intensity of
these effects seems to be low and in any case will depend on the pricing system
adopted.

In order to measure the territorial impacts of road pricing, the scenarios before and
after the pricing action are compared, then the resulting effects are measured both in
terms of effectiveness and regional equity. From the regional equity point of view it is
also interesting to analysis the spatial spillovers of the changes in the road system:
which impacts from transport changes in one region are produced on its neighbouring
regions (Pereira and Sagalés, 2003). These effects can be positive (constructions of new
infrastructures) or negative (increase of generalized transport costs, as in the case of a
pricing policy). Measuring spillover effects is important because while the revenues of a
pricing policy are produced in one region, its negative effects are 'exported' also to other
regions.

Geographic Information Systems (GIS) are very useful in order to perform
accessibility analysis. GIS can store spatial disaggregated information about the
distribution of the population and the economic activities, they allow to perform the
necessary network analysis to calculate the accessibility indicators and besides to
represent the results cartographically. With a GIS it is possible to measure spatial
spillovers adequately, determining where and with what intensity the changes and the
accessibility spillovers are produced. The traditional methodologies that measure the
spillovers are limited because of its regional analysis framework. They fail to
understand that some parts of a region do not receive any effects (positive or negative)
while in others these effects can be very intensive.

This paper is part of a research project called META (Spanish Road Pricing
Model), financed by the Spanish Ministerio de Fomento. In this project it is intended to
propose and evaluate a pricing system for the high capacity road network. One of the
objectives of this project is the study of road pricing policy implications on
accessibility, and in consequence in regional economic potential and in regional equity.
Here we present the methodology carried out in a GIS to evaluate the changes in
accessibility and the spillover effects of those road pricing policies. It will be also
presented some of the results obtained and finally we conclude with some final remarks.

2. Methodology: Measuring accessibility effects of road pricing

In this section we present a GIS based methodology to measure the impacts of a
possible road pricing policy in the Spanish high capacity road network. The selection of
the accessibility indicator, the data, the selected measures of cost and tolls used for the
calculation of the spillovers are explained.
a) Selection of the accessibility indicator

In literature there are a great variety of indicators to measure the accessibility. Most
combine the transport cost to and the capacity of attraction of the different economic
centres (Geertman and Ritsema van Eck, 1995). The transport cost is a measure of the
impedance or friction of distance. It can be expressed as units of distance (Keeble et al.,
1988), time (Lutter et al., 1992; Bruinsma and Rietveld, 1993; Dundon-Smith and Gibb,
1994; Geertman and Ritsema van Eck, 1995; Gutiérrez and Urbano, 1996), or
generalized transport costs (Linneker and Spence, 1992; Spence and Lineker, 1994). On
the other hand, the destination capacity of attraction is related to its volume of economic
activity. Depending on the available data, different attraction variables can be utilized,
like population (Bruinsma and Rietveld, 1993), employment (Linneker and Spence,
1992; Spence and Lineker, 1994) or GDP (Keeble et al., 1988; Gutiérrez and Urbano,
1996; Gutiérrez et al., 1996).

In this paper the economic potential indicator was selected. This indicator belongs
to the family of the gravity models, since the weight given to the relationships among
economic centres decrease with the increase of distance. The economic potential is
without doubt the most utilized measure in the accessibility studies (see for example
Harris, 1954; Hansen, 1959; Clark, Wilson y Bradley, 1969; Keeble, Offord y Walker,
1988; Smith y Gibb, 1993; Spence y Lineker, 1994; Dundom-Smith y Gibb, 1994;).

Its equation is commonly known as:

(1)

Where Pi is the potential accessibility of centre i, Mj is the mass (population) of the
economic destination centre, Cij is the cost through the network between centre i and j.
Finally, α is a parameter representing the friction of distance: a high parameter values
give a greater weight to relations established in a short distances.

Analyzing the accessibility in conditions of road pricing requires the consideration
of generalized transport cost as a measure of impedance between the places of origin
and destination of the trips. Implementing the cost function means including both
internal costs (consumption, maintenance, vehicle depreciation) and external cost
associated with the trips (environmental, social, …). It is assumed in our model that the
proposed fee represents the external costs of the trips, so that external costs are
internalized via fees. On the other hand, the congestion reductions expected as a result
of road pricing involve a certain decrease in travel costs and consequently benefits in
the levels of accessibility (MuConsult, 2000 Tillema et al, 2003).

b) Data

The accessibility analysis in a GIS requires introducing a digital road network to
simulate transport flows along the network. For this study a digital road network was
built to include all the Spanish national roads and the main regional roads. All the
motorways and the main road network in Portugal and the South of France have also
been included in the network. The baseline year is 2005. It is a dense network (2200
arcs), with large territorial coverage, to guarantee that the spatial variability of the
effects of the pricing policy is well estimated and to interpolate the results precisely.

Each arc of the network contains information about the type of road (high capacity
roads, conventional roads, urban roads), length, speed and travel time, this one
calculated as a function of length and speed parameters. The speed on the roads varies
from a maximum of 120 Km/h, in the case of motorways, to a minimum of 30 km/h for
the case of urban roads.

In order to represent the places where the trips have their origin and destination,
815 transportation zones have been defined, and the respective centroides calculated. In
Spain, the transportation zones were built from the automatic allocation of
municipalities to the closest nodes of the network, considering the distance through the
network. Subsequently this automatic allocation was adjusted with respect to various
homogeneity and coherence criteria like: size and form (avoiding vast or sinuous zones,
to assure a homogeneous accessibility to the network), continuity and considering the
natural barriers as transportation zones limits. The transportation zones present an
average of 60,000 inhabitants and an area of approximately 730,000 km2.

In the neighbouring countries, Portugal and France, where the required precision
was not so detailed, the transportation zones coincides with the concelhos and
departments centroids, respectively.
For each Spanish zone the population data in year 2005 was stored. The population
serves as a weighting factor in the accessibility indicator, attributing greater importance
to the relations established with the most populated zones. In this way, the population
will be the variable that serves as a proxy of the economic activity volume of the
different transportation zones. The population data were obtained from the respective
national statistical institutes (INE, from Spain; INE from Portugal and INEES from
France).

c) Scenarios definition

To measure the impacts of a pricing policy in Spanish accessibility it has been
considered two scenarios. A reference scenario, that represents the road situation in the
year 2005 (T0) and a road pricing scenario (T1) in which fees in all Spanish high
capacity roads are introduced. The accessibility is measured in each one of these
scenarios and subsequently the results are compared to determine the impact of the
pricing model in the regional accessibility, according to:

(2)

Being, Ii the accessibility impact in zone i; AiT0 the accessibility value calculated for
the zone i in referense scenario and AiT1 the accessibility value calculated for the zone i
in the road pricing scenario. In the reference scenario, without the pricing policy, the
transportation costs were calculated converting to economic values the costs of travel
time. In this cost the only value considered was the time value of the driver, that was
estimated in 10 euro/hour. On the other hand, the cost related to consumption of fuel
and the vehicles maintenance where aded. We consider a value of mean consumption of
fuel equal to 6 liters/100 km, and a price of 1 euro per liter of fuel. With respect to the
car maintenance costs, they were estimated at 0,024 euro/km, based on the report the
Market Observatory of Freight Transportation by Roads of the Spanish Ministerio de
Fomento. The generalized transport costs were then associated to each arc of the
network.

In the road pricing scenario two tolls were defined: a toll of 0,03 euro/km for not
congested roads network, and a higher value (0,06 euro/km) for congested roads
situated around the large cities. So, in this scenario, we add the tolls costs existing in
high capacity road network to the cost function of the reference scenario.

d) Calculation of the accessibility indicator

To obtain the economic potential indicator we used a GIS to calculate the necessary
cost to cover the distance between the places of origin and destination, using an
algorithm of shortest network paths. The internal relations in transport zones were
estimated. For urban zones (more than 75000 inhabitants) the effect of congestion was
calculated proportionally to its population, applying the following formula:

tii = 3.4336 Ln (pi) – 0.8476 (3)

Being tii the internal time of the zone i and pi the population of that zone. This
formula is the result of a function obtained from internal time and the population known
for several Spanish cities. In the rest of the transportation zones internal travel time was
established on 10 minutes.

Finally, a penalty in both origin and destination was added to the travel cost. This
penalty tries to represent the delays caused by exiting and accessing the cities centres.
The total transportation cost for each one of the relations is:

CGTij = po + CGTrij + pd (4)

Where, CGTtij is the generalized transport cost between the node of origin i and the
centroid of destination j, po is half of internal generalized transport const of the origin
zone, CGTrij is the shortest generalized transport cost trough the network between i and
j, pd is half of internal generalized transport cost of the destination zone.

Once the travel time origin - destination matrix was obtained, the economic
potential indicator was calculated. Being a gravity model, this indicator tries to
represent the decrease of the spatial interaction with the increase of the distance. In this
study a value of 1 for the α parameter of the equation (1) was assumed.
e) Calculation of spillover effects

The methodology that evaluate the transportation networks spillovers based on
accessibility indicators has been placed in some recent works (Gutiérrez and Condeço,
2006a and 2006b; López et al, 2006; López, 2007; o Gutiérrez et al, in press). In the
case of road pricing spillovers the methodology is also based in the comparison of two
scenarios: the reference scenario is the situation with tolls in all the Spanish high
capacity roads; the comparison scenario is a situation without tolls in the case study
region and with tolls in the rest of high capacity roads. The spillovers are obtained
comparing the two previous scenarios, according to the following formula:

ED xi = AiT0 − AiT x1 (5)

where, Edxi is the spillover effect in node i produced by road pricing in region x;
AiTx1 is the accessibility of the node i calculated in the scenario with tolls in all the
region except in the region x; AiT0 is the accessibility of the node i calculated in the
reference scenario setting with road pricing.

3. Results

Figure 1 represents the accessibility conditions before and after the road pricing
implementation. The central image symbolizes the decrease of accessibility due to the
increase of generalized transport cost as a result of the toll incorporation in the road
system. A general decrease of accessibility in Spain can be noted, but also in the
neighbour countries – Portugal and France. The reduction of accessibility in the
neighbour countries, due to the Spanish road pricing policy is a good example of
spillover effects.
Figure 1 Accessibility effects of road pricing policy

Big cities and their surroundings present the highest values of economic potential
(Figure 1), because they can access to more important destinations with less
transportation cost. Madrid, Barcelona, Valencia and the two bigger cities in Portugal –
Lisbon and Porto – have higher accessibility values, both for their self-potential as
because they have a high capacity road network that allow better relations with the main
urban areas. On the other hand, some peripheral regions in the Iberian Peninsula and
also most of the territories at both sides of the Spain – Portuguese border, stay at the
worst position in the accessibility ranking. Table 1 shows the results aggregated by
autonomous communities.
Table 1 - Accessibility impacts of road pricing in Spanish autonomous communities

Economic Potential Average
Autonomous
Before road After road Accessibility
Communities
pricing Pricing change (%)

Andalusia 7778 7255
6.75
Aragón 8572 8148
4.49
Asturias 7380 6969
5.52
Cantabria 7931 7442
6
Castile-La Mancha 9207 8519
6.88
Castile and León 8449 7985
5.44
Catalonia 10532 10099
3.77
Extremadura 7675 7283
4.8
Galicia 7189 6856
4.79
La Rioja 8280 7986
4.04
Community of Madrid 13908 12845
8.59
Region of Murcia 8188 7561
7.6
Navarra 8313 8013
3.67
Basque Country 8630 8207
5.16
Valencian Community 9107 8488
6.11
National average 9442 8870 5,5

The increasing costs due to the toll implementatio in the high capacity road network
entail a generalized reduction of accessibility. However, this reduction is bigger in some
territories. The Spanish accessibility reduces an average of 5,5 % compared to the
reference scenario. Logically, road pricing most affected regions coincide with those
with a more dense high capacity road network. The new accessibility distribution in the
territory is more homogeneous due to the road pricing measure. This is show by the
dispersion indicators (Table 2) and it can be interpreted as a increase in spatial cohesion.
Table 2 - Statistic summaries of accessibility situation and accessibility changes

Without road
With road pricing Changes
pricing
B A – B (%)
A
N 1171 1171 0.00
Range 11438.62 9549.24 -16.52
Minimum 4949.93 4766.21 -3.71
Maximum 16388.55 14315.45 -12.65
Mean 7982.01 7476.74 -6.33
Standard Deviation 1188.73 1038.26 -12.66
Variance 1413078.97 1077984.32 -23.71
Coefficient of Variance 14.89 13.89 -6.72
Skewness 1.58 1.45 -8.23
Standard error of skewness 0.07 0.07 0.00
Kurtosis 7.56 6.96 -7.94
Standard error of kurtosis 0.14 0.14 0.00

Due to space constraints, in this paper we present only the spillover effects of two
autonomous regions. Two regions with opposite characteristics (in relation to their
geographical location, spatial extension, density of high capacity road network) were
selected. Madrid (a region located in the centre of the Iberian Peninsula, small and with
a high motorway density) and Extremadura (peripheral and with a low density of
motorways) have been considered as case study.

Figure 2 represents the spillover effects produced by toll introduction in the road
system network in Madrid, while figure 3 shows the spillovers produced in the case of
Extremadura. In both maps, the region analyzed appears in grey color because what we
pretend to measure and map are the spatial spillovers in the neighbor regions.

The implementation of a road pricing policy in Madrid produces very relevant
spillovers. The most penalized areas are the closest territories, especially those located
just near the toll motorways. These areas depend in a great extent on Madrid transport
infrastructure, both in the relationships to Madrid, as in the relationships to other
regions. The central position of Madrid in the Iberian Peninsula determines that the
implementation of a road pricing system in this region will produce important spillover
effects on many others. The reductions in accessibility due to Madrid road pricing
system are situated in about 4%.

More peripheral regions like Extremadura produce more reduced spillover effects
basically because there are fewer trips in its territory and generally have less attraction
capacity. Furthermore, in 2005 the only motorways in Extremadura were the A5 and
part of the A66. The introduction of toll on these motorways will have little accessibility
reductions outside its territory. The economical potential decreases are more
accentuated in the proximities of the A5 and hardly exceed the 2%, specially in the
areas close to the Portuguese border. This situation is due to the existence of more
important destinations in the Spanish territory that cause more important accessibility
reductions on the Portuguese side.

To conclude, the spillover effects produced by the road pricing policy in the
Spanish regions vary with the following factors:

1) The geographical position of the region: the transport infrastructure of a
central region has a greater importance on interregional trips. The
implementation of toll in this infrastructure will penalize more trips than it
would do if the toll was to be introduced in a peripheral region.

2) The density of high capacity road network: autonomous regions with a
greater density of motorways will produce more intense spillovers.

3) The proximity to the region: the spillovers are higher near the region
where the pricing policy is implemented, due to the gravity component of the
economic potential indicator that gives more weight to relationships over short
distances.
Figure 2 - Spillover effects of road pricing in Community of Madrid

Figure 3 - Spillover effects of road pricing in Extremadura
4. Conclusions

This paper presents a recent research in evaluating the accessibility spillovers
effects of a road pricing policy. We use a commercial GIS (ArcGis) to carry out the
accessibility analysis and calculate the road pricing impacts. This is possible through the
comparison of two scenarios: with and without the pricing policy. Finally we measure
the regional spillovers of road pricing.

The first results show generalized reductions of accessibility. More penalized
regions are those located near the high capacity road network, especially the bigger city
surroundings, where the density of toll motorways is greater. The most penalized
regions are the Community of Madrid, Castile-La Mancha, Andalusia and Region of
Murcia, because of their high density of motorways.

On the other hand, the Spanish road pricing policy will produce effects outside the
national territory, especially in Portugal and France. This is due to the existence of
spillover effects. In order to analyze these effects we present a methodology based on
accessibility. Compared to the traditional spillovers methodologies, our methodology
has the advantage to represent correctly the real spatial extent of the impacts and also to
capture the different effect intensities on the territory.

The spatial distribution of spillover effects will depend on different factors which
determine its magnitude and its geographical extension. Spillovers will be greater near
the toll motorways, but will also depend on the motorway orientation, the geographical
position of the region and on the toll network density in the region.

Acknowledgment

This research has been financed by the project “META” (CEDEX, Ministerio de
Fomento)
5. Bibliography

Banister, D. and Berechman, Y. (2001). Transport investment and the promotion of
economic growth. Journal of Transport Geography, 9(3), 209-218.

Beuthe, M. (2002): Transport evaluation methods: from cost-benefit analysis to
multicriterio analysis and the decision framework. In Project and Policy Evaluation in
Transport, L. Giorgi and A. Pearman, (Eds.), Ashgate, Aldershot.

Bruinsma, F.R. and Rietveld, P. (1993): "Urban agglomerations in European
infrastructure networks". Urban Studies, 30, pp.919-934.

Clark C, Wilson F, Bradley J (1969) Industrial location and economic potential in
Western Europe. Regional Studies 3:197-212

Dundon-Smith, D.M. and Gibb, R.A. (1994): "The Channel Tunnel and regional
economic development". Journal of Transport Geography, 2(3), pp.178-189.

Forslund, U.M. and Johansson, B. (1995): Assessing road investments: accessibility
changes, cost benefit and production effects. The Annals of Regional Science, 29,
pp.155-174.

Geertman, S.C.M. and Ritsema van Eck, J.R. (1995): "GIS and models of accessibility
potential: an application in planning". International Journal of Geographical
Information Systems, 9(1), pp.67-80.

Gutiérrez, J. y Urbano, P. (1996): "Accessibility in the European Union: the impact of
the Transeuropean road network". Journal of Transport Geography, 4(1), pp.15-25.

Gutiérrez, J., González, R. y Gómez, G. (1996): "The European high-speed train
network: predicted effects on accessbility patterns. Journal of Transport Geography,
4(4), pp.227-238.

Hansen, W.G (1959): How accessibility shapes land-use. Journal of American Institute
of Planners, 25(2), pp.73-76.

Harris, C.D. (1954): "The market as a factor in the localisation of industry in the United
States". Annals of the Association of American Geographers, 44, pp. 315-348.
Keeble, D., Offord, J. and Walker, S. (1988): Peripheral regions in a community of
twelve. Brussels/Luxembourg: Office for Official Publications of the European
Communities.

Linneker, B. and Spence, N.A. (1992): "Accessibility measures compared in an analysis
of the impact of the M25 London Orbital Motorway on Britain". Environment and
Planning A, 24, pp. 1137-1154.

Lutter, H., Pütz, T. and Spangenberg, M. (1992): Accessibility and peripherality of
Community regions: the role of road-. long-distance railway- and airport networks.
Commission of the European Communities, Brussels.

Morris, J.M., Dumble, P.L. and Wigan, M.R. (1978): "Accessibility indicators for
transport planning". Transportation Research, 13A, pp. 91-109.

MuConsult (2000). "Ruimtelijke Effecten Prijsbeleid in Verkeer en Vervoer
(Achtergronddocument)". MuConsult: 1-93, Amersfoort.

Pereira MA and Sagalés OR (2003): Spillover effects of public capital formation:
evidence from the Spanish regions. Journal of Urban Economics, 53:238-256.

Schürmann, C., Spiekermann, K., and Wegener, M. (1997). Accessibility indicators.
Berichte aus dem Institüt for Raumplanung 39, IRPUD, Dortmund.

Smith, D.M. and Gibb, R.A. (1993): "The regional impact of the Channel Tunnel. A
return to potential analysis". Geoforum, 24(2), pp.183-192.

Spence, N. and Linneker, B. (1994): "Evolution of the motorway network and changing
levels of accessibility in Great Britain". Journal of Transport Geography, 2(4), pp.247-
264.

Tillema, T., van Wee, B. y de Jong, T. (2003). Road pricing from a geographical
perspective: a literatura review and implications for research into accessibility. Paper
presented at the 43rd ERSA Congreso, Jyväskylä (Finland)

Vickerman, R.W., Spiekermann, K. and Wegener, M. (1999): Accessibility and
economic development in Europe. Regional Studies, 33 (1), 1-15.