The demand for public transport:a practical guide

R Balcombe, TRL Limited (Editor)R Mackett, Centre for Transport Studies, University College LondonN Paulley, TRL LimitedJ Preston, Transport Studies Unit, University of OxfordJ Shires, Institute for Transport Studies, University of LeedsH Titheridge, Centre for Transport Studies, University College LondonM Wardman, Institute for Transport Studies, University of LeedsP White, Transport Studies Group, University of Westminster

TRL Report TRL593

ii

First Published 2004ISSN 0968-4107Copyright TRL Limited 2004.

This report has been produced by the contributory authors andpublished by TRL Limited as part of a project funded by EPSRC(Grants No GR/R18550/01, GR/R18567/01 and GR/R18574/01)and also supported by a number of other institutions as listed on theacknowledgements page. The views expressed are those of theauthors and not necessarily those of the supporting and fundingorganisations

TRL is committed to optimising energy efficiency, reducingwaste and promoting recycling and re-use. In support of theseenvironmental goals, this report has been printed on recycledpaper, comprising 100% post-consumer waste, manufacturedusing a TCF (totally chlorine free) process.

iii

ACKNOWLEDGEMENTS

The assistance of the following organisations is gratefully acknowledged:

International Association of Public Transport (UITP)Local Government Association (LGA)National Express Group plcNexusNetwork RailRees Jeffery Road FundStagecoach Group plcStrategic Rail Authority (SRA)Transport for London (TfL)Travel West Midlands

Richard Rampton (National Express)Julie Rickard (Network Rail)Elaine Rosscraig (Stagecoach)David Simmonds (David Simmonds Consultancy)Bob Stannard (SRA)Bill Tyson (GMPTE)Ian Wallis (Booz Allen Hamilton)Richard Warwick (Arriva)Andrew Wickham (Go Ahead Group plc)Nigel Wilson (MIT)Mike Woodhouse (Arriva)Howard Wyborn (EPSRC)

ArrivaAssociation of Train Operating Companies (ATOC)Confederation of Passenger Transport (CPT)Department for Transport (DfT)Engineering and Physical Sciences ResearchCouncil (EPSRC)FirstGroup plcGo-Ahead Group plcGreater Manchester Public TransportExecutive (GMPTE)

The Working Group coordinating the project consisted of the authors and Jonathan Pugh and MatthewChivers of ATOC and David Harley, David Walmsley and Mark James of CPT.

The study was overseen by a Steering Group consisting of the members of the Working Group and thefollowing people whose contribution is also gratefully acknowledged:

Chairman: Mike Walsh (DfT)Vince Christie (LGA)John Dodgson (NERA)Malcolm Fairhurst (TfL)Neil Fleming (SRA)Bernard Garner (Nexus)Phil Goodwin (UCL)Bill Harbottle (Nexus)Line Juissant (UITP)Steve Lowe (MVA)Roland Niblett (Colin Buchanan & Partners)Derek Palmer (Steer Davies Gleave)

In addition there are many people at the authors’ institutions who have contributed to the study. The authorswould particularly like to acknowledge the role of Chris Nash of the University of Leeds and MartinHigginson then of TRL in helping to design and initiate the project. Others to whom thanks are due are JoyceDargay, Phil Goodwin, Mark Hanly, Graham Parkhurst, and Emma Shane at University College London, PaulaBagchi at the University of Westminster, Tom Sanson then of the University of Leeds, Biao Huang and FionaRaje at the University of Oxford, and Claire Vance and Helen Harper at TRL.

iv

v

CONTENTS

Page

Executive Summary 1

1 Introduction 3

1.1 The need for a new report on public transport demand 3

1.2 Scope of the report 4

1.3 Structure of the report 4

2 Setting the scene 5

2.1 Scope of the study 5

2.2 Transport modes 5

2.3 Demand for different forms of public transport 7

2.4 Variations in demand 7

2.5 Trends in public transport demand and provision 11

2.6 Concluding observations 15

3 Summary of findings 15

3.1 Effects of fares 15

3.2 Effects of quality of service 19

3.3 Demand interactions 22

3.4 Effects of income and car ownership 23

3.5 Relationships between land-use and public transport 24

3.6 New public transport modes and services 26

3.7 Effects of other transport policies 28

3.8 Application of elasticity measures and modelling 30

4 Data sources and methodology 31

4.1 Principal data sources on public transport ridership 31

4.2 Measures of aggregate demand 32

4.3 Issues in the use of operator-based data 33

4.4 Use of survey data 34

4.5 Other concepts in market analysis 35

4.6 Data on factors affecting demand 36

4.7 Modelling and elasticities of demand 38

5 Demand functions and elasticities 39

5.1 Introduction 39

5.2 The demand function concept 39

5.3 The elasticity concept 40

5.4 Dynamics and public transport 42

5.5 Effects of demand interactions 45

5.6 The ratio of elasticities approach 46

vi

Page

5.7 An example of the use of elasticities 46

5.8 Some guidelines on the practical use of elasticities 47

5.9 Incorporation of revealed preferences changes notreflect in 'elasticity' values 48

5.10 Concluding observations 49

6 Effects of fares 49

6.1 Introduction 49

6.2 Types of fares 50

6.3 Elasticity of bus travel 50

6.4 Elasticity of rail travel 51

6.5 Effect of methodology 53

6.6 Effect of types of fare change 55

6.7 Variation of elasticity with type of area 56

6.8 Fare elasticities for different trip purposes 58

6.9 Elasticities for different types of traveller 60

6.10 Elasticity by distance travelled 61

6.11 Effect of ticket types and fare systems 62

6.12 Zero fares 64

6.13 Effect of concessionary fares 64

6.14 Meta-analysis of British fare elasticities 68

6.15 Comparison with the analysis in the 1980 version of theDemand for Public Transport and other major studies 68

6.16 Concluding remarks 69

7 Effects of quality of service: time factors 69

7.1 Introduction 69

7.2 Travel time 69

7.3 Effect of access time to boarding point and egresstime from alighting point 72

7.4 Effect of service intervals 73

7.5 Effect of changes in time spent on board the vehicle 78

7.6 Inferring elasticities from attribute valuations 82

7.7 Conclusions 83

8 Effects of quality of service: other factors 83

8.1 Introduction 83

8.2 Effect of the waiting environment 84

8.3 Staff and security 85

8.4 Effect of vehicle or rolling stock characteristics 86

8.5 Effect of interchanges between modes 89

8.6 Reliability 90

8.7 Effect of information provision and promotional activity 91

vii

Page

8.8 Impact of marketing campaigns and service quality 94

8.9 Bus specific factors 96

8.10 Conclusions 101

9 Effects of demand interactions 101

9.1 Introduction 101

9.2 Competition between modes 101

9.3 Within mode competition 108

9.4 Concluding remarks 111

10 Effects of income and car ownership 111

10.1 Introduction 111

10.2 The expected effects of income and car ownershipon public transport demand 112

10.3 The effect of income on travel expenditure anddistance travelled 113

10.4 The effect of income on the demand for publictransport 114

10.5 The effect of car ownership on the demand forpublic transport 115

10.6 Joint effects of income and car ownership on thedemand for public transport 117

10.7 Posible variations in income elasticity over time 121

10.8 Conclusions and recommendations 122

11 The relationship between land-use and public transport 122

11.1 Introduction 122

11.2 The effects of land-use on public transport demand 123

11.3 The use of land-use policy to increase the demandfor public transport 129

11.4 The effects of public transport on economic growthand development 132

11.5 Public transport as an instrument of planning policy 135

11.6 Conclusions 138

12 New public transport modes 139

12.1 Introduction 139

12.2 Light rail 139

12.3 Guided busways 147

12.4 Park and Ride 149

12.5 Forecasting demand for new services 152

13 Effects of other transport policies 153

13.1 The objectives of transport policies 153

13.2 Infrastructure management 156

viii

Page

13.3 Employer subsidies 160

13.4 Congestion charging 160

13.5 Parking 163

13.6 Land use planning 165

13.7 Transport policy integration 165

13.8 Concluding remarks 165

Bibiliography 168

Notes 188

Glossary of Terms 189

Appendix to Chapter 4 192

Appendicies to Chapter 6 194

Preface 194

Appendix to Section 6.1 195

Appendix to Section 6.2 196

Appendix to Section 6.3 197

Appendix to Section 6.4 202

Appendix to Section 6.7.3 207

Appendix to Section 6.7.4 209

Appendix to Section 6.8 211

Appendix to Section 6.14 215

Appendix to Chapter 7 223

Appendix to Chapter 9 225

Apppedix to Chapter 10 226

Abstract 237

Related publications 237

1

Executive Summary

This document reports on the outcome of a collaborativestudy undertaken by the Universities of Leeds, Oxford andWestminster, University College London and TRL. Theobjective of the study was to produce an up-to-dateguidance manual for use by public transport operators andplanning authorities, and for academics and otherresearchers. The context of the study was principally thatof urban surface transport in Great Britain, but extensiveuse is made of international sources and examples.

The study was co-ordinated by a working groupconsisting of researchers from the aforementionedorganisations, and officials of bodies representing thepassenger transport industry. The overall direction of theproject was the responsibility of a steering group whichincluded other researchers, transport consultants,representatives from local and central government, bus andrail operators as well as members of the working group.

The overall objectives of the study were to:

� undertake analysis and research by using primary andsecondary data sources on the factors influencing thedemand for public transport;

� produce quantitative indications of how these factorsinfluence the demand for public transport;

� provide accessible information on such factors for keystakeholders such as public transport operators andcentral and local government;

� produce a document that assists in identifying cost-effective schemes for improving services.

In 1980 the then Transport and Road ResearchLaboratory, now TRL Limited, published a collaborativereport The Demand for Public Transport, which becamewidely known as ‘The Black Book’. The report has beenthe seminal piece of work on demand evaluation for manyyears, but in the succeeding two decades a great deal ofchange has taken place. The values of many of theparameters under consideration have changed, newmethodologies and concepts have emerged and theinstitutional, socio-economic, environmental and legalframeworks are substantially different.

While such changes have not invalidated the generalconclusions of the Black Book, they will have reduced therelevance to modern conditions of much of the quantitativeanalysis. The concerns of policy makers and planners noware less with the problems of maintaining public transport,on which the mobility of a sizeable minority of peopledepends, but with increasing its attractiveness to car users.Effecting significant shifts from car to public transporttravel would reduce congestion and improve efficiency ofnecessarily road-based transport operations, as well assecuring important environmental benefits. An improvedunderstanding of the determinants of public transportdemand will help to inform those involved in this processand this was the aim of the new study.

The study has re-examined the evidence on the factorsaffecting the demand for public transport, and has

extended the coverage from that of the 1980 study toreflect the changing sociological and policy background.

The most widely estimated parameters have been priceelasticities of demand and, in particular, public transportfare elasticities. Evidence collected during the studysuggests that short-term elasticities, relating to changes indemand measured soon after changes in fares, may besubstantially different from long-term elasticities, based onmeasurements made several years after fare changes.Broadly speaking: bus fare elasticity averages around -0.4in the short run, -0.56 in the medium run and -1.0 in thelong run; metro fare elasticities average around -0.3 in theshort run and -0.6 in the long run, and local suburban railaround -0.6 in the short run. These results appear toindicate a significant change from those reported in the1980 study.

The examination of quality of service identifies sevencategories of attributes of transport services thatcollectively determine quality, and examines evidence asto how these components of quality affect demand. Thefindings are presented either in the form of elasticities, oras weights to be given to the various quality componentswhen incorporating them in generalised costs for purposesof modelling. There is limited evidence on elasticities withrespect to in-vehicle time (IVT). The available evidencesuggests that IVT elasticities for urban buses appear to beroughly in the range -0.4 to -0.6, while those for urban orregional rail range between -0.4 and -0.9.

Attribute values have been derived for various aspectsof bus shelters, seats, lighting, staff presence, closed-circuit TV and bus service information. Estimates forindividual attributes of the waiting environment range upto 6p per trip (subject to a limiting cap of around 26p onthe total), or up to 2 minutes of in-vehicle time per trip.

Regarding the effect of income on public transportdemand, the bus income elasticity, which includes the carownership effect, appears to be quite substantial, in a rangebetween -0.5 and -1.0 in the long run, although somewhatsmaller in the short run. Evidence of the effect of anotherkey influence on public transport demand, car ownership,indicates that in Great Britain, a person in a car-owninghousehold is likely to make considerably fewer trips byboth bus (66% fewer) and rail (25% fewer) per week thana person in a non car-owning household.

While the Guide examines the influence of fares, qualityof service and income and car ownership, it also considersnew transport modes such as guided busways, therelationship between land use and public transport supplyand demand and the impacts of transport policies generallyon public transport. It looks at the influence ofdevelopments in transport and technology over the pasttwo decades, such as innovations in pricing, changes invehicle size, environmental controls on emissions, anddevelopments in ticketing and information provisionfacilitated by advances in computing.

2

The main objective of this Guide is to provide practicalguidance on demand estimation for those involved inplanning and operating public transport services. It istherefore written in a modular form so that readers mayfind the information and guidance they require withouthaving to read the whole document. The derivation ofconclusions from the large body of research and sources ofdata considered is presented in order to establish thereliability of the advice presented, and to serve as a sourcebook for future research.

3

1 Introduction

This book is the result of a new collaborative studyundertaken by the Universities of Leeds, Oxford andWestminster, University College London and TRLLimited. The objective of this study was to produce an up-to-date guidance manual for use by public transportoperators, planning authorities, academics and otherresearchers. The context of the study is principally that ofsurface transport in Great Britain, but extensive use ismade of international sources and examples.

The study was co-ordinated by a working groupconsisting of researchers from the aforementionedorganisations, and officials of bodies representing thepassenger transport industry. The overall direction of theproject was the responsibility of a steering group whichincluded other researchers, transport consultants,representatives from local and central government, bus andrail operators as well as members of the working group.

1.1 The need for a new report on public transportdemand

In 1980 the then Transport and Road Research Laboratory,now TRL Limited, published a collaborative report: TheDemand for Public Transport (Webster and Bly, 1980).This report, which became widely known as ‘The BlackBook’, identified many factors which influence demandand where possible, given the limitations of the data thatwere available for analysis, quantified their effects. TheBlack Book subsequently proved to be of great value topublic transport operators and transport planners andpolicy makers. However, in the following 20 years therehas been a great deal of change in the organisation of thepassenger transport industry, the legislative frameworkunder which it operates, in technology, in the incomes,life-styles and aspirations of the travelling public, in carownership levels, and in the attitudes of policy makers.

While these changes have not invalidated the generalconclusions of the Black Book, they will have reduced therelevance to modern conditions of much of the quantitativeanalysis. There is therefore a need for a revised versionwhich can take into account another 20 years’ worth ofpublic transport information, and recent advances intransport research techniques.

The Black Book was written at a time when demand forpublic transport was falling very rapidly (Figure 1.1), andoperators’ options for maintaining profitability - fareincreases, reductions in service levels and networkcoverage - seemed counterproductive. It was predicted thatever-increasing levels of subsidy would be needed just topreserve current public services.

Some 20 years on the demand for bus travel in GreatBritain appears virtually to have stabilised, arguably at ahigher level than would have been predicted by extrapolationof the trend from 1970 to 1980. More vehicle km wereoperated in 2000/01 than at any time since 1970, following adecline of 21% between 1970 and 1985/86. Publicexpenditure1 on bus services has fallen by about 16% inreal terms since 1985/86, from £1637m (in 2001/02 prices)to £1367m in 2001/02. So two objectives of the TransportAct 1985, which abolished quantity control of local busservices and led to privatisation of most publicly ownedbus operators, were achieved, at least in part. The failure toreverse the trend in passenger numbers was adisappointment, at least to authors of the policy.

The resurgence of rail travel since about 1995 isremarkable in view of recent financial difficulties facingthe industry and (possibly exaggerated) public concernover safety and service reliability. Recent growth may belargely attributable to economic growth, constraints on caruse, service improvements and the fact that rail fares(unlike bus fares) have been subject to price controls.

The concerns of policy makers and planners now areless with the problems of maintaining public transport, onwhich the mobility of a sizeable minority of people

0

1000

2000

3000

4000

5000

6000

7000

8000

9000

10000

1970 1975 1980 1985 1990 1995 2000

Year

Pas

seng

er tr

ips

(mill

ion)

Local bus

National rail

London underground

Figure 1.1 Trends in public transport demand in Great Britain 1970-2000

4

depends, but with increasing its attractiveness to car users.Effecting significant shifts from car to public transporttravel would reduce congestion and improve efficiency ofnecessarily road-based transport operations, as well assecuring important environmental benefits. This objectivewill not be achieved easily, but there appears to be a strongpolitical will to pursue it. An improved understanding ofthe determinants of public transport demand will help toinform those involved in this process and this book isdesigned to provide it.

1.2 Scope of the report

There can be little doubt that a wide range of factorsinfluences the demand for public transport. There is plentyof empirical evidence as to what the relevant factors are,and which of them may be more important than others, indifferent circumstances. But devising useful definitionsand measures of these factors can be a formidable task.Even with that achieved, the remaining problems ofexplaining observed demand as a complex function of allthe relevant factors, in order to develop models of howdemand is likely to be affected by changes in any or all ofthem, may be even more difficult. That is not to say thatimperfect models which do not entirely reflect all thecomplications of the real world are without value: animperfect model may be more useful as a planning orpolicy-making tool than a series of well informed guesses,but it must always be recognised that the results may besubject to a considerable degree of uncertainty.

The key issues addressed in this book are theidentification of factors influencing demand andassessment of their impact on trip generation and modalsplit. Research outputs are synthesised, including thoserelating to setting fare levels, devising marketingstrategies, determining supply strategies, and assistingstrategies to reduce car dependency culture and theassociated environmental disbenefits this causes.

The overall objectives of the study are to:

� undertake analysis and research by using primary andsecondary data sources on the factors influencing thedemand for public transport;

� produce quantitative indications of how these factorsinfluence the demand for public transport;

� provide accessible information on such factors for keystakeholders such as public transport operators andcentral and local government.

� produce a document that assists in identifying cost-effective schemes for improving services.

The new report presents evidence on factors influencingthe demand for public transport drawn from three keyareas:

� fundamental principles relating to transport demand;

� evidence from new factors and research carried outsince publication of the 1980 report.

� empirical results for a range of modes.

The study also considers the influence of developmentsin transport and technology over the past two decades,

such as innovations in pricing, changes in vehicle size,environmental controls on emissions, and developments inticketing and information provision facilitated by advancesin computing.

1.3 Structure of the report

The main objective of this book is to provide practicalguidance on demand estimation for those involved inplanning and operating public transport services. It istherefore written in a modular form so that readers mayfind the information and guidance they require withouthaving to read the whole document from beginning to end.The derivation of conclusions from the large body ofresearch and sources of data considered is presented inorder to establish the reliability of the advice presented,and to serve as a source book for future research.

The arrangement of the Chapters is as follows:

Chapter 2 sets the scene for the study, discussing recentdevelopments in public transport operation, current trendsin demand, and the various factors which may influence it.Much of this material will be familiar to well informedreaders, who may skip this chapter.

Chapter 3 is a summary of all the principal findings of theresearch. For ease of reading it is presented in as non-technical a manner as possible, without details of theevidence and argument supporting the conclusions: this isto be found in the subsequent technical chapters, to whichreferences are made. This chapter is recommended to allreaders, to gain an overall understanding of all the issuesraised, and to point them towards those parts of thetechnical evidence and argument which they need for theirown purposes.

Chapter 4 outlines the public transport data sources whichare available for analysis, their strengths and weaknesses,and their use in demand modelling.

Chapter 5 is a mainly mathematical exposition of theconcepts of demand functions and elasticities, whichunderlie the findings of subsequent chapters. It is notrequired reading for non-specialists.

Chapter 6 deals with the effects of pricing (publictransport fares) on demand.

Chapter 7 deals with aspects of public transport serviceswith a time dimension, principally walking times to andfrom stops and stations, waiting times and in-vehicle times.

Chapter 8 considers other service quality factors, includingthe waiting environment, comfort and safety.

Chapter 9 considers effects of changes in alternative publictransport modes and in costs and times of journeys by car.

Chapter 10 discusses the influence of incomes and carownership on public transport demand.

Chapter 11 analyses interactions between public transportdemand and land use patterns.

Chapter 12 discusses the impact of new public transportsystems, and methods for forecasting demand for them.

Chapter 13 examines the effects of other transport policies.

5

2 Setting the scene

The purpose of this chapter is to put the study into contextby providing background information on public transport,its users, and non-users. In particular the followingquestions are raised, for more detailed quantitativediscussion in the following chapters:

� What is public transport?

� How is public transport developing?

� What is the demand for public transport?

� How does demand vary between different areas, types oftransport services, types of people, types of journey,journey purposes?

� How do external factors affect demand?

� How is demand changing?

2.1 Scope of the study

The main concern of this book is with the demand forpublic transport in Great Britain. It seems likely howeverthat many of the findings will also be applicable to othercountries in broadly similar states of socio-economicdevelopment and be of practical use there. Indeed, we havecast our net as widely as possible in a search for relevantinformation on public transport demand and researchstudies. We have concentrated mostly on high per capitaincome ‘western’ countries, primarily those in WesternEurope, North America and Australasia. Conditions inlower-income developing countries are often so different,in terms of private vehicle ownership for example, that fewuseful generalisations can be made. Data availability alsotends to inhibit the level of analysis that can be undertaken.However, major industrialised centres in Asia also displaycharacteristics similar to those in western Europe and offerexperience of intensive public transport use and provisionwhich is relevant to this study. Examples include Japan,Singapore, and the Hong Kong region within China. Inaddition, the conditions faced by urban metro systems inlarge cities are often similar, and in this case experiencefrom a wider range of countries may be relevant (forexample, including some cities in South America).

Within countries considered of relevance, the emphasisin this study is on urban and regional markets, i.e. thosedominated by short-distance travel, and fairly highfrequencies of movement (such as home-to-workcommuting). The long-distance, air and tourist markets arenot explicitly considered, although they do form part of thenational aggregate transport demand.

In terms of drawing a distinction, a useful example isthat used in the British National Travel Survey(hereinafter, NTS), in which ‘long distance’ is defined asthose trips above 50 miles (approximately 80 km) oneway. This would include all intra-urban trips and the vastmajority of commuting travel (apart from some long-distance commuting into very large cities such as London),together with some shorter inter-urban trips.

Sources of data, problems arising from differences in theways in which they are collected and reported, andsuggestions for resolving these problems are discussed in

some detail in Chapter 3. Suffice it to say here that wehave obtained and analysed data from public transportoperators relating to:

� trips made on urban networks (bus, tram light rail,metro);

� trips on regional bus systems (which tend to beconcentrated within the urban areas they serve, or madefrom surrounding rural areas).

The main ambiguity arises in respect of national railsystems which also carry substantial flows into large cities.Separate patronage statistics may not be available, and evenwhere the administrative structure has been split up, it doesnot necessarily match urban hinterlands (for example, theprivatised rail system in Britain includes some self-contained urban networks, such as that on Merseyside, butalso companies handling a mix of interurban and long-distance work, such as South West Trains, serving theregion within, and to the south-west of, London).

We have also made extensive use of NTS informationcollected over a number of years, and have analysedresults of ad hoc surveys of passenger demand, reported innumerous research papers.

2.2 Transport modes

The main emphasis in this study is on public transportmodes, but it is also vital to take into account privatetransport modes. These can generally be regarded as incompetition with public transport (this applies especially toprivate cars) but may also be complementary (e.g. walkingto bus stops, driving to railway stations).

The transport modes included in the broad analysesincorporated in this study are described briefly as follows.

2.2.1 Public transport modesBuses and coaches

The largest element in public transport provision, being themost ubiquitous mode. A distinction may be drawnbetween ‘local’ and ‘other’ services. Local services areavailable to the general public on demand (generallyserving all stops along a route, with cash payment onboard, or at stops, permitted). Other services includelonger-distance scheduled services (such as intercityexpress coach) which are also available to the generalpublic. Contract school services are not open to the generalpublic. Some buses and coaches are available for privatehire by organisations or individuals.

In the British case, there is a legal distinction betweenregistered ‘local’ services (eligible for Bus Service OperatorGrant2) and ‘other’ (the latter comprising about one third ofthe operating industry turnover and vehicle-km run).

However, the ‘other’ category may contribute asubstantial element of urban and regional public transportprovision, especially where large numbers of schoolchildren are carried. Their use may be detected inhousehold surveys such as the NTS, where sufficientlyaccurate definitions are applied.

In recent years there have been various developmentswith the aim of making bus services more attractive to

6

passengers. These include bus priority schemes, designedto reduce bus journey times and make services morereliable by isolating buses from general traffic congestion.While some such schemes have been successful othershave achieved few benefits (Daugherty et al., 1999), oftenbecause of difficulties in circumventing physical obstacleswhere priority measures are most needed.

Guided bus schemes are a variation on conventional buspriority measures, imbuing bus services with some of thefeatures of light rail systems (including more effectiveexclusion of non-priority traffic), but with the addedadvantage of greater flexibility at the ends of guidewaysections. Low floor buses are becoming more common,enabling easier access to elderly and infirm passengers,parents with young children, as well as to wheelchairusers. Off-bus ticketing systems are improving passengerconvenience, and reducing boarding times, with benefitsfor journey times and service reliability. Bus locationsystems can contribute to bus priority measures, and to realtime information systems for passengers.

Taxis and private hire vehiclesThese are classified as public transport since they areavailable for public use, and may have a rolecomplementary to, or competing with, that of buses andrailways. In Great Britain their provision and use hasgrown rapidly since 1985, and is substantial in periodssuch as late evenings when conventional publictransport services may be limited in scope. A ‘taxi’ maybe defined as a vehicle available for hire on demand onstreet and at designated ranks. Fare scales are normallyprescribed by licensing authorities, and are incorporatedin the settings of taximeters. A ‘private hire vehicle’(PHV) is typically a saloon car which may be hired bypre-arrangement (such as telephone booking), the farebeing determined by agreement with the passengerrather than according to a fixed scale. In some largecities, such as London, the roles of taxis and PHVs maydiffer markedly, but in smaller centres they oftenperform a similar role.

Together, they account for some 10% of all publictransport trips in Great Britain. The average trip length issimilar to that by local bus, but less for public transport asa whole, thus comprising about 6% total of publictransport passenger-km (see Table 2.1). Fares (per km) aregenerally higher than for other modes. Consequently taxisand PHV receive a disproportionately high share (21%) ofuser expenditure on public transport – almost as much asthat for non-local buses and coaches (25%).

It is not the purpose of this study to examine the taxi/PHV market in detail, nor produce forecasting models.However, this mode may have a substantial role as anexplanatory factor in the market for bus and rail services.

Tramways and light rail

This mode includes traditional street tramways (such asBlackpool or Amsterdam), many of which have beenexpanded and upgraded through reserved track extensionsand priority measures (such as Gothenburg). Entirely new

Table 2.1 Demand for surface public transport inGB (2001)

Passenger Passengerjourneys Passenger km revenue

(M) (billions) (£M)

Local bus 4310 2889 46

Non local bus/coach 200 1531National railways 960 39.1 3548London underground 950 7.45 1151Glasgow underground 10 0.04 10Light rail 130 0.8 98Taxis/PHVs 700 5.7 2320

All public transport 7260 99 11547

Private cars 37070 500

Most of the statistics in this table are taken from Transport StatisticsGreat Britain: 2002 Edition (Department for Transport, 2002c) derivedfrom operators’ reports and statistical returns.

The estimated number of passenger journeys by non-local bus and coachis very uncertain (±100M). It is based on an average annual trip rate oftwo per person (subject to rounding errors) derived from NTS datashown in the same publication, together with an allowance for trips foreducational purposes by ‘other private transport’, much of which mayactually be ‘non-local bus/coach’.It is not possible to derive separateestimates of passenger kilometres for local buses and non-local busesand coaches.

For taxis and PHVs, estimates have been made using annual trip ratesand average annual distance travelled by each mode, derived from NTSdata shown in the same publication. Passenger revenue is based on anestimated average fare of £3.33 per person trip (Department of theEnvironment, Transport and the Regions, 2001c).

Light rail includes: Docklands Light Railway, Tyne & Wear Metro,Manchester Metrolink, Sheffield Supertram, West Midlands Metro,Croydon Tramlink.

}

systems have been developed from the 1970s, eitherlargely segregated (such as Calgary) or reintroducing thestreet tramway in a more modern form (such as Nantes).Average trip length is generally short, and the role filledmay be similar to that of buses, but with greater potentialfor attracting car users due to higher speeds and servicequality. Wholly automated systems often fall within thiscategory in terms of density and trip length.

Most of the light rail systems listed in Table 2.1 havebeen introduced in recent years, and have achievedsubstantial growth, albeit from a very low base. Whetherthe new patronage represents mainly transfers from caruse, or from other public transport modes, is a question tobe discussed in Chapter 9.

‘Heavy’ urban railThis mode comprises underground and metro systemsdesigned for high capacity, and fully segregated fromsurface traffic (such as those in London, New York andParis). Station spacing tends to be somewhat greater thanfor tramway/light rail, and average trip length longer. Usetends to be concentrated on radial work trips to/from citycentres. The degree of short-run substitution with othermodes is often less than for bus or light rail.

7

2.2.2 Private transport modesWalkingThis mode continues to play an important role, both in itsown right, and as a feeder mode to public transport. InBritain it comprises about 25% of all trips, when short tripsare included, being particularly important for shopping,personal business, and home-to-school trips by youngerchildren (5-10 age range). Much of the apparent growth inoverall trip-making in recent years is in fact a growth inmotorised trip-making, much of which derives from atransfer from walking. In some cases this involves shiftsbetween walking and public transport : strong substitutionmay exist between bus use and walking for trips around 1to 2 km. This is reflected in higher price elasticities forexample (see Chapter 6, Section 10).

Cycling (non-motorised)

This mode has generally declined in recent years,comprising only about 2% of all trips in Britain(Department for Transport, 2002b). It thus has little impacton overall demand, but may form an alternative to publictransport for trips up to about 10 km. In some countriesfavourable climate, topography and priority measuresmake this mode much more important. In the Netherlands,for example, it accounts for about 29% of all trips (CentralBureau of Statistics for the Netherlands, 1995), and maytake a substantial share of the short-distance market thatwould be handled by bus in Britain. In some instances,cycling acts a feeder to public transport, where secureracks are provided at rail stations or accompaniment on-train is encouraged (as in Copenhagen), thus enlarging theeffective catchment areas of public transport networks.

Cycling (motorised)This mode includes all powered two-wheelers(motorcycles, scooters, etc.). In most developed countriesit now plays a very small role, but remains important inlow-income countries as the first stage of privatemotorised transport, thus competing with public transport.

Private carWithin this mode, a distinction may be drawn between‘driver’ and ‘passenger’ use. Passengers may be moreinclined to switch to public transport, for example when itoffers greater flexibility of trip timing than afforded by thedriver. However, if reducing vehicular movement is thepolicy aim (as for instance in Park and Ride (P&R)schemes) it is necessary to persuade drivers to switch topublic transport.

A further distinction may be drawn between ‘main’and ‘other’ drivers (for example, in the NTS), the formerbeing the driver undertaking the greater distance in asurvey period. In many countries, the most commoncategory of household car ownership is the one-carhousehold (46% of all households in Britain in 2001, forexample). A complementary pattern of car and publictransport use may exist within such households, and the‘other’ driver may make only limited use of the car. Asthe proportion of households with two or more cars rises

(22% in Britain in 2001 - Department for Transport,2002c) some of the driver trips made in the second carare diverted from the first car. Hence, the impact onpublic transport use tends to be lower than the initial shiftfrom zero to one car ownership.

In considering competition between public transport andprivate car, average occupancy levels in the latter modemay be important, since direct perceived costs (such asfuel, and/or parking charges) may be divided by thenumber of occupants. Where public transport use involvesseparate tickets for each person, relative costs may appearmuch greater, for example where family members aretravelling together. Surveys such as the NTS may be usedto derive average values (which may also be assessedthrough specific local studies).

The overall average car occupancy level in Britain in1999-2001 was 1.6 (for all purposes and trip lengths), butlevels are often much higher for non-work purposes - in1999-2001 averaging about 1.9 for leisure, and 1.8 forshopping, reaching 2.5 for holidays or day trips. Hence forthese purposes, perceived cost per person by publictransport may compare particularly unfavourably. Ticketssuch as the ‘family railcard’ or ‘family travelcard’ (inLondon) can be seen as response to this.

2.3 Demand for different forms of public transport

Current levels of demand for the public transport modesdescribed in Section 2.2.1 are summarised in Table 2.1.Demand measures are shown in three ways: annual numbersof passenger trips (one-way); annual passenger km; andannual passenger revenue.

Two caveats should be borne in mind when consideringTable 2.1. The first is that the possibility of some doublecounting may arise from information supplied by operatorswhen trips necessitate interchange within or betweenmodes (see Chapter 4 for more detailed discussion). Thesecond is that national rail and non-local bus and coachstatistics include a considerable proportion of long-distance trips which do not comply with the definition oflocal and regional trips adopted for this study.

Despite these minor complications Table 2.1 gives areasonable impression of the relative sizes of the demandsfor different modes. In particular, it illustrates thedominance of car travel over public transport.

2.4 Variations in demand

Despite their importance national statistics give a rathersuperficial impression of the complex subject of demandfor public transport. Public transport is used for a range ofpurposes by people of both sexes and of all ages, withdifferent levels of income and car ownership, living indifferent types of area. In this section we briefly discusshow these factors affect demand in order to illustrate theissues to be explored in later chapters.

2.4.1 Variation by person typeThe most advanced public transport ticketing systems canprovide information on trip ends, trip lengths, time and dayof travel. They can also provide separate information for

8

passengers enjoying various kinds of concessions, orusing season tickets or travelcards. But for a detailedanalysis of person types and journey purposes and triprates it is necessary to resort to survey data, especiallythose produced by the National Travel Survey (NTS).These data enable us to indicate variations in publictransport use by age and sex, both in terms of absolutetrip rates and market shares.

Trip rates are highest in the ‘working age’ groups,from 17 to 59. Bus and coach use tends to beconcentrated at each end of the age spectrum,representing about 6% of all trips, but around 12% eachin the age groups 17-20 and 70+. It is lowest in the agegroups 30-59, at only 3 to 4%. This is associated with caravailability, the youngest groups not yet being able toown cars, and many of the oldest group never havingdone so. Conversely, rail use shows much less variation,its highest share being in the 21-29 age group (at 4%).Taxi and private hire car use is fairly well spread over theage groups (an average share of 1%, highest at 3% in the17-20 age group) (Table 2.2).

Females tend to make greater use of public transportthan males, their average bus and coach share being 7%(compared with 5% for males), with a similar distributionby age category. Rail use is marginally lower amongfemales than males, but taxi and private hire car usesimilar. A more noteworthy difference between males andfemales is the split between car driver and car passengeruse, 48% of all trips by males are made as car drivers and17% as car passengers. For females these proportions are33% and 28% respectively. The differences are much lessmarked in the youngest groups.

For the public transport market, there are someimportant implications. While rail and taxi use is fairlywell spread by age and sex, bus use is clearly associatedwith lack of access to cars. In the case of the older groups,its level of use is also associated with lower fares due toprovision of concessionary travel. In future, people inolder age groups are more likely to have developed habitsof car use when younger and to maintain them. This maymake it necessary to ‘market’ the concept of concessionarytravel to such groups. West Yorkshire PTE for example

undertook such a campaign in Summer 2002, with anintroductory offer of one month’s free travel (theconcessionary pass in its area normally denoting eligibilityfor a cash flat fare).

There may, however, be some prospect of retaining andincreasing public transport use among the younger agegroups, provided that an acceptable quality of service andprice can be offered. School and education travel may beoffered at concessionary fares (or free travel in somecases), but in many instances the full adult fare may bepayable from the age of about 16 (dependent uponoperator policy).

2.4.2 Variation by time of day, and day of weekThe internal structure of the public transport market mayalso be examined in terms of trip length distribution, andsplit by time of day and day of week. Within the Mondayto Friday ‘working day’, work and education trips tend tobe concentrated at peak periods (around 0800-0930, and1600-1730). However, they do not usually coincide in bothpeaks, since the school day is generally shorter than theadult working day. Where service industry employmentpredominates, working hours are typically around 0900-1700, causing the morning school and work peaks tocoincide, but with a spread in the late afternoon, as schoolsfinish around 1530-1600.

In many areas, it is the school peak which causes almostthe entire additional peak vehicle demand above a ‘base’level from 0800 to 1800. This is evident in almost allsmaller towns, and in most cities up to about 250,000population, such as Plymouth and Southampton. Althoughjourneys to work by public transport are substantial, theydo not necessarily require more vehicular capacity (giventhe higher load factors accepted in the peak) than forshopping, and other trips between the peaks. Even in thelargest conurbation bus networks, it is only on the radialroutes to the central area that journeys to work create sharppeaks, school travel causing the peak within suburbanareas.

Rail networks display a very different peaking ratio,however, being oriented almost entirely to the centres oflarge cities, and thus the adult work journey.

Table 2.2 Percentage of trips by age and main mode (1998-2000)

All ages

9

The ratio of peak to base demand may be somewhatgreater in terms of passengerkm than passenger trips, sincethe journeys to work are often much longer than localshopping and personal business trips within the suburbanareas. The 1999-2001 NTS (Department for Transport,2002b) shows that for all modes, work commuting trips tendto be longer (at 13.7 km average) than those for education(4.8 km) or shopping (6.7 km), with an ‘all purposes’average of 10.68 km. Conversely, in smaller townsemployment and shopping may show a similar degree ofconcentration, leading to similar trip lengths, and hence agood balance of demand during the base period.

In many areas, a greater decline than average hasoccurred in early morning, evening and Sunday publictransport use. Car availability to the household as a wholeis much greater in the latter two periods, and the first hasbeen affected by loss of work journeys and changes inworking hours (although changes in activity patterns onSundays have led to increases in public transport usewhere fairly comprehensive networks are provided, as inLondon). Evening travel has also been affected by thelong-term drop in cinema attendance (albeit recentlyreversed), and a reluctance among residents of some areasto go out in the dark for fear of assault. However, very lateevening and allnight bus travel has grown rapidly inLondon following the revamped network introducedduring the 1980s and subsequently expanded.

Personal security has emerged as major issue in recentyears, and a number of operators have introduced measuresto counteract passenger perceptions, for example in Swedenand Belgium (Warden, 2002, Detroz et al., 2002).

Table 2.3 shows variation by day of week (averagedover the whole year) for selected modes. On an index of100, distance travelled per day is typically higher onMonday to Friday, local bus use peaking on Thursdays,and rail use on Friday when commuting, business andweekend travel coincide. Saturday use is slightly belowthat for Monday to Friday for buses, usually due to highlevels of shopping travel, but substantially lower for rail.Sunday travel is particularly low for bus (associated withlow levels of service provision) but less so for rail. Taxiand minicab use peaks on Fridays and Saturdays,associated with leisure travel.

Table 2.4 shows corresponding variation by month ofyear. This is somewhat greater, typically lower in February(possibly due to weather conditions), June and July forlocal bus, with a similar variation for taxi/PHV. Raildisplays a somewhat different pattern, possibly associatedwith long-distance travel.’

A study by one major bus operator group in Britain hassuggested that weather conditions may affect ridership.Like-for-like period comparisons suggest little impact oftemperature or hours of sunshine (within the rangeexperienced prior to 2000), but heavy rainfall produced areduction of about 3% in demand. Where demand changeshave been measured through sample counts or surveysover short periods (such as one day), adjustments for dayto week, month and/or weather factors, using data abovemay be desirable.

Table 2.3 Index of daily distance travelled by day of week and mode: 1999/2001

Monday Tuesday Wednesday Thursday Friday Saturday Sunday All days

Walk 101 113 110 105 110 90 73 100Car/van driver 99 102 107 108 111 91 83 100Car/van passenger 82 72 81 77 101 150 139 100Local bus 105 121 119 121 114 91 32 100Rail 111 118 110 125 130 63 46 100Taxi/minicab 94 75 84 110 130 131 78 100

All modes 96 96 100 102 111 109 97 101

Derived from National Travel Survey (unpublished data). ’All modes’ also includes cycle, motorcycle, express and tour coaches, etc. Where data areshown for a specific mode and day the sample size obtained was at least 300 in all cases.

Table 2.4 Index of annual distance travelled per person per year by selected modes and month of year: 1999/2001

AllJan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec months

Walk 95 91 104 100 104 97 103 96 109 103 108 87 100Car/van driver 87 98 100 95 101 106 111 98 95 109 110 91 100Car/van passenger 85 91 81 94 105 121 107 124 106 102 99 87 100Local bus 117 86 110 99 102 90 91 108 102 96 101 95 100Rail 76 125 86 94 94 117 101 89 79 124 132 87 100Taxi/minicab 79 61 99 168 103 82 103 106 125 91 111 75 100

All Modes 85 97 94 95 103 111 111 106 98 105 107 88 100

Derived from National Travel Survey (unpublished data). ’All modes’ also includes cycle, motorcycle, express and tour coaches, etc. Where data isshown for a specific mode and month the sample size obtained was at least 300 in all cases.

10

2.4.3 The journey to work

Table 2.5 shows the shares of the journey-to-work marketby different modes, from the 1997-99 NTS (Department ofthe Environment, Transport and the Regions, 2000b). Notethat the principal mode of transport used is shown - forexample, for a journey to central London comprising a longride by surface rail, and a shorter ride on the Underground,only surface rail will be shown as the mode used.

Overall, buses account for about 9% of all journeys towork, and rail about 5% (or, as shares of the motorisedmarket, about 11% and 6% respectively). As one wouldexpect, the public transport mode share is greater forcentral London, with 49% of journeys to work by surfacerail or Underground, and 10% by bus. Note that thesefigures are for the whole day : during the morning peak(0700-1000), the rail share is substantially greater at about75%. The proportions vary substantially between differentparts of outer London (Croydon is well-served, forexample) and between the conurbation centres. Thegreatest share handled by bus and coach is for conurbationcentres (16%). Note that almost six times as many carcommuters travelled as drivers than passengers, giving anaverage car occupancy for this purpose of only 1.2.

2.4.4 Other journey purposes

Although public transport’s role tends to be associatedmainly with the work journey, it is evident that this is notnecessarily where bus takes the greatest share. In manycases, buses take a larger share of the shopping market,and this in turn forms a larger share of all bus trips thanwork, as Table 2.6 shows. The largest share taken by localbus is often within the education trip market. In 1999-2001, local buses (i.e. scheduled services open to thepublic) carried 23% of journeys between home and schoolby children aged 11-16, more than by car at 19%. ‘Privatebuses’ (typically, buses and coaches contracted to carrychildren on behalf of local education authorities)accounted for another 8%. This is also a sector in whichthe scheduled local bus share has been increasing, havingrisen from 20% in 1985/86 (Department for Transport2002b). As many home-to-school trips lie above walkingdistance, a major demand for public transport is created,especially in rural areas.

Conversely, the role of rail is generally small for non-work purposes (1% in the case of home to school trips by11-16 year olds, for example).

2.4.5 Variation by area typeThe interaction of all the factors discussed so far anddemographic factors and variations in public transportservice levels from place to place results in a strong overallrelationship between demand, expressed in trip rates, andsettlement size. This is illustrated in Table 2.7.

Table 2.6 Composition of the market for each mode, byjourney purpose 1993/95

Local bus(outside Surface

Purpose London)% rail%

Work (commuting) 21 51Business 1 6Education 15 7Shopping 34 9Personal business 10 7Visiting friends 13 11Sport/Entertainment 4 5Other 2 4

Total 100 100

Source: Department for Transport, 20 National Travel Survey 1993/95(Department of Transport 1996).

Table 2.7 Frequency of local bus use by size ofsettlement 1997-99

Frequency of use(percentages of respondents)

Less thanonce a Once

Once week more or twicea week than twice a year

Settlement size/type or more a year or less

Greater London 45% 28% 29%Metropolitan built-up area 38% 20% 41%Large urban area (over 250k) 33% 25% 42%Urban area (25k to 250k) 24% 15% 61%Small urban area (3k to 25k) 19% 19% 62%Rural 12% 14% 74%

Source: Special tabulations from 1997/99 NTS.

Table 2.5 Usual means of travel to work by usual place of work 1997/99

Pedal/ Londonmotor Car Car Bus or Surface Under-

Area Walk cycle driver passenger coach rail ground Other

Central London 7 3 26 4 10 29 20 –Outer London 10 6 63 6 9 4 3 –Conurbation centre 8 2 54 13 16 6 n/a 1Other urban 11 5 61 12 10 1 n/a –Not urban 13 4 65 12 4 – n/a –

Average 10 4 60 11 9 3 2 1

Percentage (rounded to nearest whole number).

Source : National Travel Survey 1997/99, (Department of the Environment, Transport and the Regions, 2000b).

11

Further NTS information for 1999/2001 shows that17% of the population travelled by local bus three ormore times a week, and a further 10% once / twice aweek. However, 43% reported use of local buses ‘once ayear or never’ and 52% made similar use of rail(Department for Transport, 2002b).

2.4.6 Changes in individuals’ travel over timeSo far, although we have disaggregated the market intocertain categories of person, we have not looked atindividual behaviour.

Individuals shift from one category to another, notsimply as their ages change, but also their status – forexample from child to student, to adult, to married personpossibly with children, to pensioner. These stages in the‘life cycle’ are associated with changes in household sizeand structure, car availability and in trip purpose. Thus thework journey is a major factor determining householdtravel behaviour for certain stages, the need to get childrento school at others.

Changes in travel behaviour are often associated withcritical events in the life cycle, such as setting up a newhome, or changing jobs. Many people may change theirmode of travel for such reasons, at least in the short run,rather than because of differences in modalcharacteristics. This leads to a high turnover in themarket, such that net changes between one year and thenext are often small compared with the gross changes thatproduce them. For example, panel surveys in Tyne andWear revealed a net reduction in the public transportshare for the journey to work of 2 percentage pointsbetween 1982 and 1983. This proved to be the net resultof 7% of respondents ceasing to be public transport users,while 5% became new users in that period. A net changeof 2% thus involved about 12% of the sample inchanging modes (Smart, 1984)

These changes are likely to be particularly noticeable ifindividual services are examined, since people maychange routes used when changing homes or jobs, whileremaining in the public transport market. Even in a zoneof apparently stable land use and total population, such asa well established residential area, constant change isoccurring. On a typical urban bus or rail route, as manyas 20% of users may have begun to use that specificservice within the last 12 months. Hence, if examiningthe impact of a recent change (such as introduction of a‘Quality Partnership’ upgraded bus service) it isimportant to distinguish users who have switched to aroute for such personal reasons, as distinct from thoseattracted by service characteristics.

A case where a high level of turnover was found arose inevaluation of the Manchester Metrolink light rail system:data from a set of ’control’ stations which continued to beserved by conventional rail services, shows that about 50%of the users were replaced by new individuals over a 3-year monitoring period, even where little change took placein service offered (Knowles, 1996).

Patterns of individual behaviour may influence tripfrequencies over a very long period. For example, basedon work in South Yorkshire and elsewhere, Goodwin and

others have suggested that trip rates developed in earlyadult life may strongly influence subsequent modal use.

This ‘turnover’ effect, also known as ‘churn’ has beeninvestigated by several researchers. A review by Chatterjee(2002) identifies ‘asymmetric churn’ in which the numbersof new and lost users over a given period are not necessarilyequal. In the case of data from the Netherlands for theperiod 1984 to 1987, some 123 ‘high users’ of publictransport in 1987 were shown to comprise only 58 who hadbeen ’high users’ in 1984, 48 previous ‘non users’ and 17previous ’low users’ (from Goodwin, 1989). Work byDargay and Hanly on the British Household Panel Survey(2003) indicates that over a nine-year period, over 50% ofcommuters change their main mode at least once. Of thosewho both move house and change employer during twoconsecutive years, 45% also change mode. Exceptionallyhigh rates may be found in areas such as central London, arecent survey in July 2002 of office workers showing thatnearly half had changed their place of work, home or meansof transport since October 2000 (Brook, 2002).

The implications of this for transport operators andplanners is that responses to changes in fares and servicequality should be assessed not only in the short run, butover long periods. Much short run change is causedprimarily by non-transport factors, but in the long runtransport characteristics will affect other choices. Forexample, individuals may be firmly committed to aspecific mode of travel for their existing home to worktrips, which may not be affected even by large changes inprice or service quality, but when relocating, will have toreconsider the routing, and perhaps mode, of those trips.

2.5 Trends in public transport demand and provision

In this section, trends in Britain are considered in moredetail and compared with those in other European countries.

2.5.1 Overall trends by mode and regionLocal busesTable 2.8 shows the aggregate trends in the local busmarket for the period 1985/86 to 2001/02. The number ofpassenger journeys fell by 23% over this period. However,there were marked regional variations. In Greater London,in which Transport for London (TfL) now has majorresponsibility for public transport provision, bus useincreased substantially. Elsewhere bus use declined, mostseverely in the English metropolitan areas (the six areas inEngland comprising the major conurbations outsideGreater London where Passenger Transport Executives(PTEs) are responsible for planning and procuringservices). A lesser decline occurred in the rest of England,and Wales and Scotland as a whole.

The marked differences shown are influenced by thefollowing factors:

� The increase in average fare per trip (which covers allpassengers, i.e.including concessionary travel). On abasis of the long-run elasticities (discussed further inChapter 6) applicable over such a period, one wouldexpect a decline in travel of similar magnitude to thepercentage fare increase shown, attributable to this

12

cause. However, it should be noted that the fare increasein the metropolitan areas was from a low absolute base,and hence a lower elasticity may be applicable there.

� The increase in bus-km run, generally a proxy forservice frequency. Again, long-run elasticities wouldimply an expected growth of similar magnitude to thepercentage shown, attributable to this cause.

� Exogenous factors, including trends in total population,its composition, and car ownership growth.

A major part of the differences between London and themetropolitan areas may be attributed to these factors. Forexample the increase in bus-km in London has offset theeffect of higher real fares, combined with populationgrowth. In the metropolitan areas, the large increase in realfares appears to be the dominant factor. However, thedecline in the rest of Britain (primarily in urban areas, withpopulations up to around 500,000) is striking, given thatthe growth in bus-km exceeded that in real fares.

Rail

Table 2.9 shows trends for rail systems from 1991/92 to2000/01. ’National rail’ corresponds to the surface networkformerly operated by British Rail and now by 25 privatisedTrain Operating Companies (TOCs). There has beensubstantial growth both in passenger journeys andpassenger km for both national rail and the Londonunderground system. Demand for rail travel appears to bestrongly correlated with employment. Thus bothunderground and national rail use peaked in the late 1980s,declined during the following economic recession, thengrew rapidly from the mid 1990s along with employmentgrowth (especially in central London).

‘Other rail systems’ comprise the Glasgow Undergroundand the light rail systems listed in Table 2.1. The largegrowth in their number in recent years producescorrespondingly large percentage increases. A consistentdata series for train-km run equivalent to that for buses isnot available.

Further consideration of light rail developments is givenin Chapter 12.

Taxis and private hire vehiclesA notable feature of trends revealed in the NTS data isthe rapid growth in use of taxis and PHVs, despite thelabour-intensive nature of this mode. The total number oftaxis and private hire cars has grown rapidly in recentyears, following legislative changes, which have partiallyremoved restrictions on the taxi trade. Further, increasingunemployment (until the mid 1990s) stimulated entryinto the trade. The number of licensed taxis in GreatBritain grew from about 39,100 in 1985 to 69,000 in1999, or by 76% (Department of the EnvironmentTransport and the Regions, 2000a) Use of taxis andprivate hire vehicles (miles per person per year) grew byabout 125% between 1985/6 and 1999/2001 (Departmentfor Transport, 2002b).

In some respects, the roles of taxis/private hire vehiclesand other public transport could be seen ascomplementary: they are used particularly for late-nighttravel. Some of the growth since 1985/86 may have beenassociated with reduced quality of bus services sincederegulation. However, in London a high level ofconventional (bus and rail) public transport use, iscombined with the highest taxi and private hire mileageper person per year within Britain.

2.5.2 Public transport and car use As car ownership has grown, it has had a direct effect onpublic transport use. Firstly, the individual having firstchoice in use of a newly acquired car (usuallycorresponding to the ‘main driver’ in the NTS, andtypically the working head of household) will tend to useit. His or her trips will then be lost to public transport,except for occasional journeys. In addition, other membersof the household may also transfer some of their trips tothe car as passengers, for example a child being given a liftto school, or the family travelling together at weekends.

Table 2.9 Changes in demand for rail, and fare levels

National London Other railrail Underground systems

Change 1991/92 to 2001/02 (%)Passenger journeys 21.2 26.9 127.2Passenger km 20.3 26.4 140.3Passenger receipts/km 6.5 26.3 33.1

Source: Department for Transport 2002d.

Table 2.8 Changes in local bus demand, service level and fare levels

England

Non-Metro- metro-

Greater politan politan AllLondon areas areas England Scotland Wales Great Britain

Change 1985/86 to 2001/02 (%)Passenger trips 24.5 -44.8 -23.2 -21.1 -34.3 -36.2 -23.1Bus km 39.2 12.7 30.7 26.1 30.9 31.6 26.9Passenger receipts per trip 10.9 69.4 23.6 31.0 21.7 23.4 28.7

Note: Passenger receipts at 2001/02 prices; concessionary fare reimbursement excluded.Source: Department for Transport 2002d.

13

The loss of trips to public transport will thus be greaterthan those of one person alone, although this could dependupon price and quality of the service offered: if it is good,then other members of the household may be less inclinedto arrange their trips so as to travel as passengers in thehousehold car.

Cross-sectional data from the NTS may be used toindicate the principal impacts of changes in car ownership.For example, as shown in Table 2.10, persons in no-carhouseholds made 156 local bus trips per person in 1995/97,falling to 43 in one-car households. At an averagehousehold size of approximately 2.4, the drop in bus tripsper annum as a result of a household shifting from the 0 to 1car category would thus be about 270.

not have a car, but this fell to 29% in urban areas of 25,000to 250,000, and to 29% in rural areas (Department forTransport, 2002).

2.5.3 International comparisonsTrends in passenger kilometres over all 15 European Unioncountries over the last decade are shown in Table 2.11.

Table 2.10 Effect of car ownership on public transport use

Number ofhousehold Local bus Bus Rail Railcars trips/person/a km/person/a trips/person/a km/person/a

0 156 875 22 5401 43 311 17 5412+ 22 200 13 589

All 62 252 17 345

Source: National Travel Survey 1995/97 (Department of theEnvironment Transport and the Regions 1998a).

Table 2.11 Growth in public transport use: EuropeanUnion countries 1990-1999

Mode Growth in passenger km

Passenger cars 18%Buses and coaches 9%Tram and metro 5%Railway 8%Air 65%

All 19%

Source: European Commission, 2002.

Table 2.12 Trip rates and time budgets (GB – allmodes, trip lengths)

1990 2001

Hours travelled per person per annum 369 360Journeys per person per annum 1090 1018Distance travelled per person per annum 10545 km 10964 kmAverage trip length 9.7 km 10.8 kmAverage speed 28.7 km/hour 30.5 km/hour

Source: Department for Transport, 2002b.

As noted earlier, the effect is greater for the first car thanthe second, since the latter will be used in part to take tripsthat were being made as car passengers in the first (thechildren acquiring their first cars, for example).

The public transport trips made by the members ofone-car households tend to be concentrated into categoriessuch as school and Monday-Friday shopping trips, withmuch less evening and weekend public transport use. Two-car households tend to make very little use of publictransport. There may be exceptions for households ofabove-average size, or where some journeys to work aremade by public transport.

It should be noted that in practice car ownershipdepends particularly on household size and composition.Non-car-owning households, whose members are oftenelderly, tend to be smaller, and contain only about 20% ofthe population. Rail trip rates are also affected by carownership, but to a lesser degree, and rail distancetravelled rises with car ownership. The majority of railusers, especially those using commuter services is thesouth east and elsewhere, come from car-owninghouseholds.

In 1999, 22.8 million private cars were licensed inBritain, corresponding to about 0.39 per head, or 1.02 perhousehold. The most common category was the one carhousehold, some 44% of the total. Another 27% ofhouseholds had two or more cars, thus leaving 28%without a car. The proportion of households with one ormore cars has grown less rapidly than car ownership intotal, as average household size has fallen. Its rate ofgrowth has also declined.

Marked variations occur by area. In 1997/99, about 38%of households in London and the metropolitan areas did

Total person-km (motorised modes only) rose by 19%.By mode, the fastest growth was in air travel, at 65%(typically, this mode has grown by about 5-6% per annumin most industrialised countries). This was followed byprivate car at 18%. However, in aggregate, public transportmodes also showed growth of 9% in bus and coach modes,5% in tram and metro modes, and 8% in rail.

Such growth in motorised travel does not necessarilyrepresent growth in trips. As an example, estimates ofgrowth based on data from the British NTS are shown inTable 2.12 In this case, a more comprehensive record ismade possible through the inclusion of non-motorisedtrips. Between 1990 and 2001 the average annualdistance travelled per person increased by about 4%.However, numbers of trips made fell slightly possiblydue to an ageing population (Table 2.2). Time spent intravel per person per day (averaged over the wholepopulation) has remained remarkably stable at about 60minutes since the mid 1970s although it may havegrown from a slightly lower figure in the previous twodecades. There are of course major variations byindividual and person type.

The growth in person-kilometres corresponded to anincrease in average speed of 10.1%. Hence, a similarnumber of trips (and thus, by implication, out-of-homeactivities) were made, but of increasing average length andspeed. If the pattern of land uses were fixed, then a wider

14

range of opportunities (for example, in terms of shoppingfacilities, or employment) could be accessed within thesame time budget. There may also be negative outcomes -the greater distance travelled could indicate that facilitieshave become less accessible than before (for example, dueto greater centralisation of health services).

This concept is supported by evidence from othercountries, notably work by Brög (1993) in Germany, astudy in Norway , and a recent survey in Lyon (Nicolasand Pochet, 2001), all consistent with an average timebudget very close to 60 minutes. For example, the work byBrög in German cities indicates that for intra-urban travelthe average time spent per person per day is very stable (atabout 60 minutes), as is the number of activities outsidethe home. In comparing surveys carried out in Essen andHanover in 1976 and 1990 it was found that average traveltime per person in Essen had changed from 60 minutes to59 minutes, and in Hanover from 61 to 62. Activities perperson per day likewise changed very little (stable at 2.7,and from 2.9 to 2.8 respectively). One could see thus theurban transport market as a whole as a ‘saturated’ market,with little scope for dramatic expansion. More substantialchanges were seen in the mix of modes used, and totaltravel distance (a shift from non-motorised modes to cardriver and public transport). However, in Hanover inparticular, public transport did not decline as a share oftotal trips, but rose from 16 to 22% over this period(Brög,1993).

Such time budget constraints are less likely to apply toweekend, leisure and long-distance travel.

Despite growing traffic congestion, average door-to-door speed has risen, for several reasons:

� A shift from non-motorised to motorised modes forshort trips, e.g. from walk to bus; from walk to car.

� A shift from slower public transport modes (primarilybus) to car, as car availability has risen: in most cases,except where extensive priority is given, car will besubstantially faster than bus for door-to-door journeys.

� An increase in trip length, reducing the proportion ofwalk and wait time within public transport trips. Theaggregate totals shown in Table 2.12 include inter-citytravel, which has grown rapidly.

� Primarily within the car mode, a reassignment and/orredistribution of trips so that relatively more travel isnow made where congestion is (or was) less severe, suchas suburban and rural areas.

The overall stability in the average travel time budgetcould also serve as a means of checking plausibility offorecasts - growth in travel which implies increased totaltravel time would be questionable. The stability alsoimplies an inverse relationship between distance travelledand average speed. For example, if within a 60-minutetravel time budget door-to-door average speed were 10kph, then 10 km would be covered. An increase to 15 kph,would imply an reduction in travel time of 33%, to 40minutes. In practice, with a constant travel time budget,distance travelled would increase by 50%, to 15 km. Thisoverall relationship would not necessarily apply equally to

each mode. For example, a greatly accelerated service(such as rail rapid transit replacing bus) would not onlyenable and stimulate more travel within the existing publictransport user market, but also divert car trips.

There are of course wide variations by person type andindividual in travel time budgets.

2.5.4 Trends within the public transport market

Table 2.13 shows some selected examples of trends for EUcountries in terms of three major public transport modes(bus and coach, metro and tram, rail). An increase isevident in most cases, the noteworthy exceptions being busand coach travel in Great Britain and also in Germany - thedecline in Germany is largely or wholly associated with asharp drop in public transport use in eastern regions afterreunification. Even in the USA, growth is evident, albeitfrom a low absolute base.

Table 2.13 Growth in passenger km 1990-1998

Passenger km (billions)

1990 1998 Change

DenmarkBus and coach 9.3 11.1 19.4%Tram and metro 0.0 0.0Railways 5.1 5.6 9.8%

All modes 14.4 16.7 16.0%

GermanyBus and coach 73.1 68.2 -6.7%Tram and metro 15.1 14.4 -4.6%Railways 62.1 66.5 7.1%

All modes 150.3 149.1 -0.8%

NetherlandsBus and coach 13.0 15.0 15.4%Tram and metro 1.3 1.4 11.1%Railways 11.1 14.8 33.3%

All modes 25.4 31.2 23.0%

Great BritainBus and coach 46 45 -2%Tram and metro 6.5 7.3 12.3%Railways 33.4 35.4 6.0%

All modes 86 88 2%

All EU countriesBus and coach 370.0 402.5 8.8%Tram and metro 48.6 50.1 3.1%Railways 270.4 280.8 3.8%

All modes 689.0 733.4 6.4%

USABus and coach 196.0 239.0 21.9%Tram and metro 19.0 22.0 15.8%Railways 21.0 23.0 9.5%

All modes 236.0 284.0 20.3%

Source: Department for Transport 2002b.

15

2.6 Concluding observations

By segmenting the market for public transport a betterunderstanding may be obtained than simply using totalvolumes of travel. This is made possible partly throughexisting operator-based ticket data, but also throughsources such as the National Travel Survey in Britain. Infuture, smartcards may enable a more reliablesegmentation to be attained.

Overall trends indicate that the absolute volume ofpublic transport demand, in terms of passenger-km, is stillgrowing in western Europe and North America, albeitfalling as a share of all motorised travel.

It remains to be determined to what extent theseinternational differences may be explained in two main ways:

� The same elasticities and exogenous factors may apply,but marked differences in funding produce, for example,much lower fare levels. Given long-run elasticity valuesnow known to apply in Great Britain, these wouldproduce correspondingly large differences in overalldemand.

� More fundamental differences may exist. These arelikely to apply to factors other than those such as realfares and service levels, for which internationalcomparisons (see Chapters 6 and 7) suggest broadlysimilar elasticity values. The income elasticity for localpublic transport use may be positive, and differencesexist in the overall perception of public transport,affecting modal choice between it and the car.

3 Summary of findings

This chapter summarises the principal findings of all thestudies considered and the additional research undertakenin the course of compiling this guide. Its purpose is toprovide an overview of the topic, helping readers to gainan overall understanding of the major issues and to learnhow to use some of the forecasting techniques presented.

The results are presented in as non-technical a manneras practicable, with a minimum of tables and figures andno direct references to sources in the literature andelsewhere. No details are given of the evidence considered,or the analysis of it which leads to the conclusions: that isto be found in the more technical chapters, at locationsindicated by the marginal references.

The subjects addressed in this chapter are the effects offares and service quality on public transport demand, theinfluence of income and car ownership, interactionsbetween land-use and transport, the demand generated bynew modes and services, and other transport policies.

3.1 Effects of fares

3.1.1 introductionFares are probably the most intensively studied transportdemand factor, for two reasons: fares and changes in faresare relatively easy to identify and quantify, and they arethe most readily, and probably the most frequently,adjusted factor.

3.1.2 Fare elasticitiesFares are fundamental to the operation of public transportsince they form a major source of income to operators. Ingeneral, if fares are increased, patronage will decrease.Whether revenue increases or decreases as a result of a fareincrease depends on the functional relationship betweenfares and patronage as represented by the demand curve.Usually this is expressed through the concept of‘elasticity’. In its simplest form the value of the fareselasticity is the ratio of the proportional change inpatronage to the proportional change in fares3. It has anegative value when, as is usually the case, fares andpatronage are inversely related: an increase in fares leadsto a decrease in patronage and vice versa. If the value ofthe elasticity is in the range zero to -1, then a fares increasewill lead to increased revenue. If the value exceeds -1, thena fare increase will lead to a decrease in revenue4.

Fare elasticities are dynamic, varying over time for aconsiderable period following fare changes. Therefore it isincreasingly common for analysts to distinguish betweenshort-run, long-run and sometimes medium-run elasticityvalues. There are various definitions of short-, medium-and long-run, but most authors take short-run to be 1 or 2years, and long-run to be around 12 to 15 (althoughsometimes as many as 20) years, while medium run isusually around 5 to 7 years.

As well as considering the direct effects of a change infares, it is often important to consider the effects of farechanges on other modes. The usual method to take intoaccount the effect that other modes have on the demand fora particular mode of public transport is to use cross-elasticities. These estimate the demand elasticity for acompeting mode with respect to the change in the givenmode, e.g. the demand elasticity of bus travel with respectto rail fares.

In this study elasticity values from many sources have beenexamined to provide an up-to-date overview of fareselasticities and the effects of various factors on the values.The principal results of this analysis are shown in Figure 3.1.

Fare elasticity varies significantly depending not only onthe mode, and the time period over which it is beingexamined, but also on the specific circumstances in whicha mode is operating. Broadly speaking: bus fare elasticityaverages around -0.4 in the short run, -0.56 in the mediumrun and -1.0 in the long run; metro fare elasticities averagearound -0.3 in the short run and -0.6 in the long run, andlocal suburban rail around -0.6 in the short run.

These results appear to indicate a significant changefrom those reported by Webster and Bly (1980) whichwere based on international aggregate measures of fareselasticity for all journey purposes and passenger typesacross all trip lengths and fares. This analysis led to theconclusion that overall fares elasticities are low, so thatincreases in fare levels will almost always lead to increasesin revenue. The analysis resulted in then accepted‘standard’ public transport fares elasticity value of -0.3.Given the dominance of before-and-after studies in the1980 report, it is likely this value is what would now becalled a short-run elasticity. In the current work the shortrun elasticity has been found to be about -0.4.

5.5.4

5.5.5

16

Two of the main reasons for this difference are as follows.Firstly, given that fare elasticity is different for differentjourney purposes, there may have been a shift in the types ofjourneys for which people are using public transport (forexample, more leisure travel). Secondly, for the samejourney purpose the elasticity may actually have changed.This could be due a variety of factors, some of which willinteract with each other: one of these is market turnoverbecause different generations and social groups of peoplewill have different perceptions of using public transport.Other factors include: rising incomes and car ownership andthe varying quality of public transport service over the last20 years. Suburban rail short run fare elasticity hasincreased slightly from about -0.5 to -0.6 in the UK.

The 1980 report did not cover medium or long runelasticities at all. Therefore the likely value of medium runbus fare elasticity of around -0.56 cannot be comparedwith earlier estimates.

The realisation that long-term elasticities can exceed -1has serious implications for the public transport industry.While the immediate effect of a fare rise might be atemporary increase in revenue, the long-term effect islikely to be a decrease, although if future cash flows arediscounted operators may benefit from fare increases.Nevertheless, attempts to counter falling revenue with fareincreases alone will eventually fail. Reversal of negative

trends in public transport patronage requires serviceimprovements, and possibly fare reductions.

The relatively wide ranges of elasticity values about themeans shown in Figure 3.1 reflect variation in methods ofestimation, as well as variation between studies in anumber of other factors influencing demand and elasticity.Some of these factors, and their likely effects, arediscussed in the following sections.

Some travellers will have a choice between the threepublic transport modes included in Figure 3.1. This meansthat if the fare on one of them is increased, but the fares onthe others are kept constant, some travellers will switch tothe alternative public transport modes. However, if the fareson all public transport modes are changed simultaneously,there will be less incentive for travellers to switch. Henceone would expect elasticity values based on a fare changeby just one public transport mode to be larger than thosebased on data from a common fare change by all publictransport modes. The former are sometimes known as ‘Ownelasticities’ and the latter as ‘Conditional elasticities’ (seeSection 5.3 for more detailed explanation).

3.1.3 Types of faresFare systems can have various forms; for example, theycan be flat, zonal or graduated. Each of these has its

Figure 3.1 Summary of mean values and ranges of fare elasticities(The bars indicate the range of elasticity values found)

6.2

-1.60-1.40-1.20-1.00-0.80-0.60-0.40-0.200.00

Suburban rail UK OP SR

Suburban rail UK P SR

Metro UK OP SR

Metro UK P SR

Bus UK OP SR

Bus UK P SR

urban rail outside SE England SR

Suburban rail SE England SR

Bus outside London SR

Bus London SR

Metro UK LR

Bus UK LR

Bus UK MR

Suburban rail outside UK SR

Suburban rail UK SR

Suburban rail UK&outside SR

Metro outside UK SR

Metro UK SR

Metro UK&outside SR

Bus outside UK SR

Bus UK SR

Bus UK&outside SR

Public transport outside UK SR

Public transport UK SR

Public transport UK&outside SR

2004 study

1980 study

SR: Short runMR: Medium runLR: Long runP: PeakOP: Off-peak

Note: this figure is a simple graphical

representation of the elasticities

summarised in Table 6.55. The

accuracy of the plot is somewhat

limited; readers requiring more

precise elasticity values should refer

to Table 6.55

17

particular characteristics. Passengers may be offered singleor return cash fares, season tickets, travelcards orconcessionary fares.

While there are many types of fare it is often necessary,for the purposes of analysis, to define a single value, whichwill often be the average amount paid for a single trip.This can be calculated by summing all the passengerreceipts in the market segment of interest and dividing bythe equivalent number of passengers. Alternatively, thetotal revenue might include funding received tocompensate for concessionary fares that the operator offersas part of a policy agreement. Other studies may definefares differently.

3.1.4 Effect of types of fare changeThe magnitude of the fare changeFare elasticities may be affected by the magnitude of thefare change. In general greater fare increases producehigher values of elasticity than lower increases. Thedifferences are greatest for long-run elasticities.

The direction of the fare change

The response to a fare increase may not be equal andopposite to the response to a fare decrease. In other words,the elasticities may not be symmetrical. However, therehave been few opportunities for studies of the effects offare reductions, so that there is little convincing evidenceof this possible asymmetry.

The level of the fare

Fare elasticity is also affected by the current level of thefare relative to people’s income. This can be illustratedby the results for London buses When fares wereparticularly low, from October 1981 to March 1982, theelasticit