THE 23RD EUROPEAN TRANSPORT FoRuM

35 downloads 179 Views 7MB Size Report
Glenthorne House, Hammersmith Grove, London W6 OLG. Tel: +44 (0)181 ..... [Newberry, 1988], the main elernents that provoke externalities are of three types:.
Proceedings of Seminar L Y.

. -.

F .,r . r

' ' ... ,."'--• ~

. :-. ' 1..

l• '

..._

,~

-1-

T ''

T ¡

p .1· -. \

, .. _

FORUM

-

THE

23RD

EUROPEAN

TRANSPORT FoRuM

11 - 15 September 1995

PTRC P398 ISBN 0-86050-288-0

Organi sed on behalf of the PTRC lnternational A ssoc iation by PTRC Education and Research Services Ltd .

1'911. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .lllti

FINANCING TRANSPORT INFRASTRUCTURE

Proceedings of Seminar L held at the PTRC European Transport Forum University ofWarwick, England from 11 - 15 September 1995

Volume P 398

Price for Delegates:

fl8.00

Price for Non-Delegates:

f24 .00

Published by PTRC EDUCATION AND RESEARCH SERVICES LID on behalf of THE PLANNING AND TRANSPORT RESEARCH AND COMPUTATION INTERNATIONAL ASSOCIATION

1995

ACKNOWLEDGEMENT PTRC would like to thank members of the Highways and Outreach Programme Comrnittees who have so willingly given their time in helping to organise the programme for this Seminar:

Highways Comrnittee: Keith Miller, Highways Agency, UK Graham Bowskill, Highways Agency, UK Roger Duffell, University ofHertfordshire, UK Jeff Falconer, Rendel Palmer and Tritton, UK Luca Formis, Alpina SpA, Italy Mike Harris, W H Fairhurst, UK Ian Harrison, Devon County Council, UK George Hazel, Lothian Regional Council, UK Graham Iones, Staffordshire County Council, UK Tony Iones, Ove Arup and Partners, UK Pierre Joubert, SETRA, Ministry of Transport, France Malcolm Noyce, Sir Alexander Gibb and Partners, ·UK Ian Ross, The Scottish Office, UK Kevin Stobbs, McAlpine Wayss and Freytag Bachy, N David Taylor, West Sussex County Council, UK Clive Wright, East Sussex County Council, UK

Outreach Committee:

Phil Comwell (Chair), European Bank for Reconstruction and Development, UK Anan Allos, Scott Wilson Kirkpatrick and Partners, UK Paul Amos, Booz Allen and Hamilton, UK George Banjo, University ofLagos, Nigeria Andy Costain, PTRC, UK Richard Darbera, University of París, France

3

..........................

Marie-Paule Devaquet, SOFRETU, France AsifFaiz, The World Bank, USA Geoff Gardner, Transport Research Laboratory, UK Margaret Heraty, Independent Consultant, UK Jeremy Lane, The World Bank, USA Henning Lauridsen, Institute ofTransport Econornics, Norway Yui Motomura, P AD ECO Co Ltd, Japan Lars Nordin, European Bank for Reconstruction and Development, UK Dick Robinson, Rende! Palmer and Tritton, UK Alan Ross, Ross Silcock Partnership, UK Luis Willumsen, Steer Davies Gleave, UK Zofia Duszynska (Programme Secretary), PTRC, UK

(e) PTRC and Contributors

Ali rights reserved. No part ofthis publication may be reproduced, stored in a retrieval system, or transmitted in any form or by any means, electronic, mechanical, photocopying, recording, or otherwise without prior perrnission of the publisher.

Available from PTRC Education and Research Services Ltd. Glenthorne House, Hammersmith Grove, London W6 OLG Tel: +44 (0)181 741 1516 Fax: +44 (0)181 741 5993 e-mail: [email protected] 4

a

CONTENTS

FINAN CE: PRIVA TE/PUBLIC PARTNERSHIPS BELTRANDI F, Steer Davies Gleave, UK Benefit-cost analysis and the private finance initiative

9

GANNON M J and BROWN N C, London Underground Ltd, UK Financial modelling for public/private sector joint ventures

19

CARBONARO G, European Investment Bank, Luxembourg Transport modelling for private financing : a view from a financing institution

31

BUCHANAN C P, Halcrow Fox, UK Using the private sector to provide public transport infrastructure - a case study from Santa Fe de Bogota, Colombia

41

PROJECT FINANCE: INTERNATIONAL EXPERIENCE MARTINOVSKA H, PHARE multi-country transport programme of the EU Commission, Czechoslovakia PHARE multi-country transport programme - aims and coordination

51

GERARDIN B, Independent Consultant, France Financing infrastructure in Eastem Europe

53

NOUL TON J, Eurotunnel, UK/France Public infrastructre at no public cost

69

OPPORTUNITIES AND RISKS IN INFRASTRUCTURE FINANCE VALENTINE W H , The Scottish Office, National Roads Directorate, UK Design Build Finance and Operate - Skye Bridge: the clients role as Trunk Road Authority

77

WILSON R, Miller Construction, UK Design Build Finance and Operate (DBFO) - Skye Bridge: the contractor' s v1ew

97

CORNWELL P, European Bank for Reconstruction and Development BOT Concessions and beyond

5

109

Objectives

The PTRC European Transport Forum

PTRC - Planning and Transport Research and Computation - is the primary European organisation dedicated to the support of professionals concerned with transport, highways and spatial planning, through the dissemination and discussion of rece nt developments in both policy and practice, and of research findings.

This multi-stream conference provides a unique international forum for the presentation and evaluation of new techniques in planning and transport.

Founded in 1966, PTRC consists of two organisations - the PTRC lnternational Association and PTRC Education and Research Services Ltd - linked through the PTRC Advisory Council.

Mid Career Training Courses Approximately 50 mid-career training courses on planning, economic development, transport, highways and computing are organised each year and provide intensive technical tuition on established advanced techniques. PTRC courses are also organised on an in-house basis for local authorities and other client organisations.

The PTRC Advisory Council is a voluntary body drawing its membership from eminent European practitioners and researchers, which oversees the technical quality of PTRC Conferences, Seminars and Training Programmes.

-The Professional Development Programme

The PTRC lnternational Association is a not for profit association with over 200 members drawn from the private and public sectors throughout Europe as well as many other parts of the world . Membership is open to any organisation or individual with an interest in transportation, highways and spatial planning . Details can be obtained from the Secretary, Th e PTRC lnternational Association, 88a High Street, Hadleigh, Essex SS7 2PB. Telephone +44 (0)1702 557255, Fax +44 (0)1702 551900.

This service is designed for planners and engineers from overseas who wish to widen or up-date their professional expertise, but who do not have sufficient time to attend a full post-graduate degree or diploma course. The programme provides short intensive training periods, consisting of short co urses and conferences, study visits, working experience, professional tuition and English Language refresher courses as required.

PTRC Education and Research Services manages the PTRC European Transport Forum and Training Programme, as well as special seminars and conferences which are managed on behalf of the Association . lt also organises a range of other services and activities for professionals concerned with transportation, highways and spatial planning . Details of the fu ll programme of PTRC events can be obtained from PTRC Education and Research Services Ltd, Glenthorne House, Hammersmith Grove, London W6 OLG . Tel ephone +44 (0)181 74 1 1516, Fax +44 (0) 181 741 5993. e-mail : ptrc@cityscape .co. uk

The papers presented at the European Tran sport Forum are published in separate volumes for each seminar and may be purchased individually or in sets through a library subscription scheme. Details of thi s scheme and a full list of publications are available on request. The complete set of proceedings for the 1995 Forum is listed below.

Publications

PRICE FOR NON-DELEGATES

CODE

TITLE

P388

Pan-European Transport lssues

f24.00

P389

Planning for Sustainability

f32.00

P390

Transport Policy and its lmplementation

f32.00

P391

Public Transport Planning and Operations

f32.00

P392

Transportation Pla~ning Methods

f32.00

P393

Models and Applications

f16 .00

P394

Traffic Management and Road Safety

f32.00

P395

Personal Access and Mobility

f16 .00

P396

Airport Planning lssues

f16 .00

P397

Roads Provision and Operation

f16.00

P398

Financi ng Transport lnfrastructure

f24 .00

P399

Rail

f16 .00

P400

Geographic lnformation Systems

f16 .00

Published by PTRC Education and Research Services Ltd on behalf of The PTRC lnternational Association, Glenthorne House, Hammersmith ,Grove, London W6 OLG, UK Telephone : +44 (0)181 741 1516

Fax: +44 (0)181 741 5993

e-mail : [email protected]

THE ALLOCATION AND CHARGING OF INTERURBAN ROADINFRASTRUCTUREINSPAIN P.Cantos 1, R. Pereira2 , A. Kiosseoglou 3 , J. Poveda3 l. Departamento de Análisis Económico. Universidad de Valencia 2. Departamento de Análisis Económico. Universidad de Vigo 3. LISITT, Departamento de Informática y Electrónica. Universidad de Valencia. C/Hugo de Moneada 4, Valencia, Tel.: +34 6 360 44 84, Fax: +34 6 361 61 98, E-mail: [email protected] Spain

ABSTRACT

This paper considers a methodology of allocation for the interurban road costs among the different types of vehicles in Spain. Valuing road track costs and externalities there are no reasons to reduce the current road tax levels. On the contrast, these taxes should be increased to the majority of vehicles. However we think that future charge systems must translate the emphasis from average costs to marginal costs. The new philosophy would be to pay the right price in the right place at the right time. In this line automatic road pricing is an important advance in this direction. Although automatic road pricing might be more easily justified in the scenarios with high externalities and congestion problems (as the urban context), we think that a gradual introduction of road pricing in the interurban network could in future be an instrument to improve the demand management and the efficiency of the road systern. Keywords: Interurban road cost, allocation and charging, automatic road pricing, road

taxes, road payment systems, smart card as road payment system. l. INTRODUCTION

The question of optima! road user charges is a complicated one, since a number of cost categories which are very difficult to calculate are involved. In addition, the wide diversity of vehicles using the roads increases the problems of costs allocation. Basically road users impo se various types of social costs: pavement costs, congestion costs, accident costs and the costs derived from environmental damage. 1n Spain only a few attempts have provided an exhaustive account of the whole problem [Esteban, A and Sanz, A ( 1994)]. The main issue was to determine if the current user charges generate sufficient revenue to pay for the road costs. The growing investments in new infrastructure in Spain in the last years have increased the debate about the suitability of the current tax system and about the introduction of new charging systems based on telematic toll technologies. In this line it is necessary to analyse in depth the problems and difficulties of the current road charges system. The objective of this paper is to define a methodology that allocates the different comporients of interurban road costs among the different types of vehicles. Hence ratios wi ll be stated describing the relation between revenue and costs by types of vehicles. From these results, some proposals, as automatic road pricing, will be briefly analysed in

2

the measure of its contribution to an improvement of the efficiency of the interurban road system In this a11icle we have not introduced the traffic in the private motorways, where in contrast to public motorways, there are to lis to pay in the exits of the motorway. Firstly, we could not obtain di ssagregated data in order to allocate road costs to this traffic. Secondl y, the traffic in this type of ways is not important, approximately 7% of the total traffic Sectio n 2 provides a theoretical review of the main issues and elements in the road charging theory that must be taken into account in the following sections. Section 3 analyses the road pricing from the allocation of road track costs (investment and maintenance costs) between the different types of vehicles. Then Section 4 introduces the extern aliti es as an additional component of road charging. Ratios indicative of the rel ati o n between road revenue (through taxes) and road costs will be presented in both sectio ns. Section 5 provides sorne propo sals and recommendations based on the introducti on of automatic road pricing. Besides, a brief description of the systems will be presented . Finally, Section 6 will provide the principal conclusions of our article .

2. THEORETICAL BACKGROUND

The highways produce two pmducts:

1. Capacity, in the form of a number of traffic volume( traffic is measured in passager car units, PCU, where one PCU is the congestive effect ora representative vehicle) . 2. Durability, in the form of thick pavement to facilitate trips in heavy motor vehicles. In this c¡uestion an efficient u ser charge would approximate the social cost of damege that a vehi cle inflicts on highways. It is well stablished that highway damage depends upon individual axles weigh, varying roughly with their 4th power [Small and Winston, 1986]. Current taxes in Spain on heavy vehicles are unsatisfactory for two reasons: they are based on total weight rather axle weight and they do not vary nearly steeply enough [Small et al , 1989] and [Hau, 1992]. On the other hand, users of infrastructure impose costs on themselves and others by co ntributing to co ngestion, which increases travel time. They also impose costs by wearing out the infrastructure, which needs maintenance expenditure to repair pavement and vehicles, and finall y they impose accident costs and environmental costs. Putting importance into order, the infraestructures cost supposes the third or fourth component of the transport total cost [Jansson, 1993]. The user time costs in the case of personal transport and the capital and operating costs of the transport vehicles are considerably greater than infrastructure costs. This means, for all kinds of transport infrastructure, that in optimization of facility design, maintenance policy, etc., the user costs pl ay a deci sive role, as seen from the point of view of the transport infrastructure owner. · A narrow limitation to the producer costs could be very misleading. The eco nomic justification for the overcharging of road users is that both the externa! costs of accidents and the cost of pollution caused by road traffic are of magnitude well above total road expenditure.

3

The basic economic assumption for efficient demand management is to set a price equal to the marginal costs of road use. The Total Costs (TC) of road traffic can be seen as the producer's cost (investment and maintenance), the user's costs (travel time, vehicle costs) and other externa! costs such as those for environmental damage and accidents. The Marginal Cost (MC) will be the sum of the differentiated total cost according to traffic volume and consists of the marginal cost for the producer, the marginal costs for the users and the marginal cost of environmental damage and accidents. The marginal cost for the users consists of the last user's own costs when entering the system (the average cost, AC) and the costs that the last user imposes on ali others (congestion). The unpriced user will take a decision based on his own costs, i. e., the average user cost. The relevant price to charge (P) is then the difference between the society's marginal cost and the user's perceived costs. The price will give the consumer the correct information about the trne cost of his decision.

TC = rc rroducer + TC users + TC extemalities MC

= MC producer

+ AC users + Congestion + MC extemalities

p

= MC producer + Congestion + MC extemalities

p

= MC11 roducer + Q

()AC users) (

· ()Q

+ MC extemalities

Therefore, for maximum net social benefit of the ex1stmg pieces of transport infrastrncture, a necessary condition is that price is set equal to the sum of the short-rnn marginal cost of the producer of transport infrastrncture services, the cost imposed on fellow users and the cost imposed on outsiders by an additional user. The first item consists mainly of use-dependent wear and tear and other annualised costs. The second item is usually called the congestion cost component. This and the third item, which contains a variety of negative externalities (environmental damage and accidents), are the majar components in the pricing of relevant costs. Particularly total accident costs in road transport is approximately half the total costs of travel time. But the calculations have big difficulties. For example, the treatment of capital cost is complicated. Two methods of capital costs were suggested. The first, normal commercial approach would be to require current traffic levels pay interest on entire capital stock of roads [Newberry, 1988]. The second approach, termed pay as you go, whereby capitaloutlays for each year would be treated simply as a current cost. If both the rate of traffic growth and the leve) of capital spending were expected to show long rnn stability, this aproach migth not be to much of a distorsion. In the face of fluctuation in traffic growth and the leve) of capital expediture, it is capable of producing illogical fluctuations in the capital cost assesed per unit of output. For instance, a cutback in road investment, leading more congestion, would result in lower charges.

4

3. THE LONG-RUN MARGINAL COST PRJCING The costs treated in this section consist mainly of new construction and maintenance of interurban road network . These public costs fall usually upon public administrations and methods must be stated in order to allocate this type of costs. The usual practice in Spain has recommended that fuel taxes must cover the costs of new infrastructure, and that another type of taxes (particularly, the annual vehicle tax) must cover the current road costs (maintenance, repairing and policy costs) As a first approxirnation, there are sorne calcu lations [Estevan, A. y Sanz, E , 1994] where revenue obtained from the diverse road taxes (fuel tax, vehicle tax, tax on car purchase) are compared with the total road costs (new construction, rnaintenance and repairing, police, administrative costs). The results in 1992 indicate a modest deficit of 18 .000 million pesetas compared with the enormous deficits of other public transport modes which transport a rnuch lower quantity of passengers and goods . Firstly, a method must be defined in order to allocate the road track costs between the different types of vehicles. Given that there is no clear allocation methodology in Spain we have opted to use a similar method that was used by the British Department of Transport. Then fifteen per cent of capital costs are allocated to heavy vehicles and buses depending on the gross vehicle weight-kilometers (gvw-km), and the rest are divided between ali the types of vehicles depending on their PCU-kilometers. Maintenance and repairing costs are divided between vehicles according to a mixture of gvw-km and standard axle-krn, and finally policy and adrninistrative costs were splited according to the runned vehicle-krns. Using these criteria of allocation the costs obtained to each type of vehicles to 1993 are the following Table l. Allocation of road tracl< costs (millionsm of IJtas.)

Capital costs

Motorcycles Cars Buses 1 LGV HGV Total

844, 1 298279,3 15258, 7 47266, 1 213301 ,6 574950

Maintenance and repairing costs 39,3 14058,2 7984, 1 3704,2 131981 , 1 157767

Police and administration

Total costs

1537,5 108662, 1 1591,8 11479,2 19464,9 142735

2421 ,0 420999,6 24834,7 62449,6 364747,6 875452,6

On the other hand, the revenue obtained from each vehicle has been calculated on a basis of the kms runned on interurban roads and to the annual tax vehicle. Given that we take into account only the interurban costs, the revenue obtained from annual tax vehicle were reduced proportionally on a basis of the interurban runned kms per vehicle with respect to the total runned krns. The revenue by type of vehicle were the following :

1

LGV : lorries with load capacity lower than 3 tons . HGV : lorri es with load capacity higher than 3 tons .

5

. 1e T a bl e 2. R evenue b y t¡ype of ve h 1c

Motorcycles Cars Buses and coaches LGV HGV

Fuel tax 2769 44299 436196 32253 287374

Tax vehicle 1674 10270 20234 2865 22578

Total 4443 54570 456430 35119 309952

Therefore we can easily obtain a ratio revenue/cost per type of vehicle: Table 3. Ratio revenue/cost

Motorcycles Cars Buses and coaches LGV HGV

Revenue 4443 54570 456431 35119 309953

Cost 1893 31322 528083 33107 429691

Ratio 2,34 1,74 0,86 1,06 0,72

Nowadays we can observe that the total revenue are higher than the total road costs. Mainly this difference is very big in the case of motorcycles and cars. In this line the complaints of sorne groups of users of these vehicles on the high road taxes would be justified. Table 5 provides an allocation between the different types of heavy good vehicles (differentiated by number of axles and average gross vehicle). With the described all ocation criteria and the revenue obtained by each vehicle the results were the fo ll owing Fuel taxes

6153 7.85 25244 .73 41702.21 8 12.3808 70406.34 43597 .77 2152.809 Total 245454. 1

SU2 SU3 SU4 CS3 CS4 CS5 TT4

Table 2 4. Balance to HGV (millions of utas.) Vehicle Total Capital Mainten. Poi ice tax Reveo ue costs costs and adm

5276.349 1967.747 268 1.709 52 .24 108 4527.561 2803.605 104.4822 17413 .69

66814.19 27212.48 44383 .92 864.6219 74933 .9 46401.37 2257.29 1 262867 .8

44075.52 2195 1.98 36880.7 561.7082 64076.25 44194 .02 1561.421 213301 .6

25085.03 13551.41 23262.56 320.476 40542.78 28283.25 935 .6176 13198 l.l

Total costs

5897.854 75058.4 2199.53 37702.92 2997.589 63140.85 58.39459 940.5786 5060.865 109679.9 3 133 .843 75611.1 116.7892 2613 .828 19464 .86 364747.6

Ratio Rev./cost

0.89 0.72 0.70 0.91 0.68 0.61 0.86 0.72

However there are sorne important arguments that modify the previous conclusions. Firstly, capital costs considered as the spending in the current year is a controversia! way of treating capital costs. Newberry (1988) argues that a more rational approach would be to require traffic levels to pay interest on the entire capital stock. This approach obtains [following to Fowkes, Nash and Tweddle, 1992] capital costs per unit of the double of 2

SU2 : single unit two axles. SU3: single unit three axles. SU4 : single unit axles. CS3: conventional se mithree axles. CS.+: conventiona l semifour axles. CS5: conventional semifive axles. TT4: truck trailer íour axles .

6

the obtained by the approach of spreading actual capital costs over total traffic in the year in question. Another argurnent suggests to rise the capital costs allocated to heavy vehicles from 15 to 25% [Echenique, 1987]. A third impo11ant aspect is the inclusion of the externalities, rnainly accidents and environrnent that not are considered in a long run context. And, finally, a more efficient way of costs allocation would lead to rise the taxes to heavy vehicles. This argurnent has problems with respect to its political acceptance, and besides this type of rneasures would be able to be contradictory with other measures related with the regeneration of industry, exports, etc.

4. THE TREATMENT OF EXTERNALITIES 4.1. The Costs of Accidents The first elernents to analyse are the externalities provoked by road accidents. Sorne recent works in the Transport Ministry have estirnated the value of the different types of casualty. Besides, the different cornponents of each casualty have been splite arnong loss of productive capacity, medica! costs, human costs and others. The results of this valuation to deaths, serious and slight injuries were the following: Table 5. Valuation of thc component of road accidents (in 1>tas of 1993)

Fatal Serious Slight

Lost output

Medical costs and others

Pain, suffering

25266275 532654 94618

87785 857138 344962

10350069 319655

o

Source: Estudio sobre los costes de la accidentalidad en los diferentes modos de transporte. MOPTMA. Marzo (1994) and COST 313 (1995).

The following step is to define clearly the accident externality costs. Following [Newberry, 1988], the main elernents that provoke externalities are of three types: The relationship between traffic flow and accident rates. Newberry argues that the number of accidents may be independent frorn the vehicle-kms runned, proportional or may rise more than proportionally. The evidence on this point is very controversial. Given the difficulty to know if increases in traffic rise proportionally the accident rates, we opted to the rnost rnoderate assurnption, and hence no externalities were considered. However, even in the case that traffic levels and accident rates are independent, the traffic cornposition will create an externality in the rneasure that the accidents provoked by different types of vehicles have very different consequences. However its rnagnitude could not be calculated given the scarcity of inforrnation to the Spanish case. The following externality is caused by the fact that accident costs are not covered totally by road users . One of the usual assurnptions has been to apply a quarter of the accident costs as comrnunity costs on the basis that the rnajority of these costs have been covered by road users in the forrn of insurance. Besides it will be reasonable to think that the

7

costs of pedestrian accidents have not been borne by private insurance, and then these will be an externality provoked by the road accidents On the other hand , [Fowkes, Nash and Twedle, 1992] reviewed in Britain the propo11ions borne by the community from data obtained from the Department of Environment. Basically this new allocation rises the proportion borne by the society until an average of 4 7% of the total accident costs. Particularly the new rates were the following T·1ble 6 Percentaoe of accident costs borne by the community

'



Lost output Fatal Serious Sli2ht

50 50 50

Medical costs 100 100 100

Pain and sufferin2 80 20

o

Poli ce

Property

100 100 100

o o o

Then we can assemble two alternative estimations for the accident externality cost. The low estimation includes the cost of pedestrian accidents besides the 25% of the cost of accidents to other non-pedestrians . The high estimation includes the cost of pedestrian accidents together with the allocations of the cost of accidents to non-pedestrians obtained from the percentages of the table. In thi s line the estimations of accident externality costs would be the following : Table 7. Valuation of accident externalities Cotas. 1>er veh.-km.)

Pedestr. costs Motorcycles Cars Buses LGV HGV

1.914 0.251 0.519 0.325 0. 121

Low estimat. (nonpedestr.) 4.638 0.286 0. 199 0.278 0.063

High estimat. (nonpedestr.) 16.28 1.007 0.783 1.003 0.216

Total Low estimation

Total High estimation

6.552 0.537 0.719 0.604 o 184

18.200 1.259 1.303 1.328 0.337

With these results we can obtain a new table with the tax/cost ratios where accident externalities have been included. T a bl e 8 R at10 . revenue/ cost .me 1u d.mg acc1 ent externa 1 1es

Motorcvcles Cars Buses and coaches LGV HGV

Ratio (low estimation) 0.391 1.513 0.825 0.947 0.714

Ratio (hi2h estimation) 0. 186 1.327 0.802 0.831 0.685

8

4.2. The Congestion Costs In an interurban context, this problem is much more important in urban areas where it is very usual that vehicles tend to delay others to some extent. Some calculations [Newberry, 1988] have estimated average values relatively very low for the marginal time costs of congestion in the case of motorways, trunk, main and other interurban roads . AJso the appropriate valuation of these costs would require to know the kilometers runned by each vehicle and theutility leve! of capacity . However there are some points where periodically there are important congestion problems. This is the case of some access motorways to the principal cities as Madrid, Barcelona and other coast cities, as well as other motorways as the Tunnel of Cadí in the skiing periods. Therefore it is usual that periodically (mainly in weekends and holidays in Spain) there are congestion problems, though in average this problem is small.

4.3. The Environmental Costs The major environmental effects of traffic are noise, visual intrusion and air pollution. Some methods based on reveled preferences (the hedonic price method or the travel cost method) have tried to value quality of the environment or the loss of ecological sites. More directly, stated preferences try to find out how much the society would be prepared to pay to avoid a determined environmental damages. One of the more used techniques has been the estimation of environmental bad from changes in the price of the houses identified with changes in the quality of environment [Pennington, G .; Topham, N. and Ward, R., 1990]. [Pearce and Markdandya, 1989] collect the results of some studies in North American cities indicating the impact of air pollution on property values . More recent investigations [Small, K and Kazimi, C., 1995] have estimated the costs of air pollution in the region of Los Angeles. The appraisal methodology consists on the evaluation of the damage provoked by air emissions. The major damages are of three types: mortality from particulate, morbidity from particulate and morbidity from ozone. The results of this study provided figures in cents for vehicle-km of 3 .28 for gasoline car, 7. 79 for light-duty diese! truck, and 52. 7 for heavy-duty diese) truck. However, these results were obtained for an urban area as Los Angeles, where the mountain barriers favour the maintenance of high pollution levels. Therefore it would not be correct to extrapolate these results (based on the relation between pollution and its health effects) to the interurban context where our work is defined. In an interurban context, the pollution and noise costs are much lower that in the urban roads. So the principal problem would consist on the contribution of the emissions of interurban traffic to the global warming. Also, Lindberg [Johanson, 1995] has valued the em1ss1ons of pollutants from the abatement cost or the cost of introduction of cleaning technology such as a catalytic conve1ter. High values are obtained for the marginal environmental costs. So, for

9

example this cost would be of 4 ptas. per km . for a car without catalytic converter, 4,8 for LGV and 22,4 for HGV . These figures confirm the importance of the problem, and hence the ratios obtained in table 9 should be modified indicating a greater difference between revenue and costs. As a measure of the extent of the problem, table 1O provides the em1ss1ons of the different types of pollutants per vehicle in the total road network. Table 9. Emissions of uollutants in TM (1992).

Motorcycles Cars Buses, coaches LGV BGV Total

co,

oc

NOX

SO,

403370 35117923 1939194 15989596 17283761 70733844

5602 429453 8974 78380 39506 561915

5378 422474 33894 257088 370366 1089200

67 11247 2817 20065 22222 56418

Source: Hacia la reconversión ecológica del transporte en España, 1994.

Respect to the global emissions of ali the modes of transport, road supposes the 88% of C0 2 _, the 98.6% ofHC, the 96 .5% ofNOX and the 85.4% of S0 2 .

5. SOME PROPOSALS AND PERFORMS From the resu lts of the previous sections so me conclusions may be obtained. Firstly there are no reasons that justify reductions in the current road taxes without a reform in the global charge system. We think that automatic road system leads towards an increase in the system efficiency, because it supposes to substitute the current average cost practice for the marginal cost theory. The expansion of these technologies in the interurban network may be carried out firstly in those motorways with important congestion problems. In this line, a new proposal would consist on the gradual introduction of automatic road pricing at the same time as the road taxes would be reduced. Besides, these technological methods permit to differentiate among types of vehicles, and hence the efficiency of the transport system would be increased in an important way with respect to the current tax system In the measure that technologic progress reduces the price of these new technologies, the advantages of this new charge system would be able to be extended to more motorways .

5. l. Description of Automatic Road Systems The electronic payment systems have a preponderant place among the road payment applications. In this section sorne general information related to this type of systems will be reviewed . The main system that we are going to describe uses as a key part of the system, smati card, a new card technology based on a microprocessor.

10

The objective of this electronic payment system is to carry out a transaction in a network . Speaking from a general point of view we can diferenciate two possibe approaches to electronic road payment system [ I.Lang, J. Macgregor, D. Hunt, 1993]:



An open system: Where the automatic account identification tag is registered as the vehicle carrying it passes the beacon installed at a point on each section ( the stretch of road between two junctions).The conventional system used to charge for bridges and tunnels, which are in effect individual sections, are "open ...



A closed system: Where the automatic account identification tag is registered as the vehicle carrying it passes the beacon installed at its point of entry to and exit from the network, and the charge calculated on the basis of the distance between the two .

There are two options for the location of the user's account: •

On vehicle The in-vehicle units reads a smart card which can be loaded with credit; this credit is reduced in response to instructions from the charging beacons. This option is very interesting due to be absolutely confidential with the information of the u ser.



Off vehicle: The beacon registers data from passing automatic account identification tags and transmit it to a processing unit which charges centrally help accounts the user would pay into the account in advance or be billed in arrears.

In all the cases the interchanged information between the vehicle and the beacon are transmited by microwave . The in-vehicle unit had two main functions : one of them to control the communication with the externa! equipment, the beacon, and the other one to identify the smart card, where is loading ali the necessary information to achieve a transaction . These type of systems can be installed to use the current installations. These systems can be also used both, for light vehicles and trucks, with a suitable performance up to speeds of250 Km/ h.

5.2. The Use of Smart Cards in Payment Systems E ven though it has a great