The Routledge Handbook of Public Transport 9780367747244, 9780367418724, 9780367816698

Table of contents :
Cover
Half Title
Title
Copyright
Contents
Preface
Contributors
Part A Policy context
1 Regulatory frameworks in public transport including tendering
2 Public transport governance
3 Mobility as a Service and public transport
4 Public transport use: the ‘soft’ side of the story
5 Intercity modal competition
6 Public transport integration
7 Intermodal connections
8 Public transport and the environment
9 Community participation in public transport decision-making
10 Land value gains and value capture: the potential for financing public transport infrastructure
11 Public transport evaluation
12 Public transport productivity and efficiency assessment
Part B Modal settings
13 Bus – from workhorse to thoroughbred
14 Rail – urban, suburban and regional
15 Air passenger services
16 Seeking economic and social success with sustainable urban ferry services
17 Flexible transport services
18 Informal paratransit in the Global South
19 Taxis
Part C Public transport environment
20 The health impacts of public transport
21 Demand estimation for public transport network planning
22 The first/last mile connection to public transport
23 Public transport and the built environment
24 Intelligent mobility and big data for planning, trust, and privacy
Part D Specific public transport delivery issues
25 The provision of service information for public transport
26 Public transport and social inclusion
27 Public transport and travel with dogs
28 Public transport use in later life
29 Parking provision, parking demand and public transport accessibility
30 Intermodal strategies combining cycling and public transport to improve service and acceptability
31 Accessibility and design for all
32 Network planning and design
Part E Smart card data for public transport planning
33 Smart card data and its use in public transport research: an overview
34 The use of smart card data to analyse platform crowding
35 The use of smart card data to analyse railway station waiting times
36 Smart card data and planning for public transport: a case study from South East Queensland, Australia
37 Variability of passenger travel patterns observed using smart card data in Japan
38 Analysis of skip-and-stop planning using smart card data
39 Analysis on the impact of bus frequency reductions on commuters using smartcard data
Part F The future
40 Automated vehicles and vehicles of the future
41 Public transport technology for the future
42 Smart mobility governance
43 Public transport and the future of mobility
Index

Citation preview

THE ROUTLEDGE HANDBOOK OF PUBLIC TRANSPORT

The Routledge Handbook of Public Transport is a reference work of chapters providing in-depth examination of the current issues and future developments facing public transport. Chapters in this book are dedicated to specific key topics, identifying the challenges therein and pointing to emerging areas of research and concern. The content is written by an international group of expert contributors and is enhanced through contributions from practitioners to deliver a broader perspective. The Handbook deals with public transport policy context, modal settings, public transport environment, public transport delivery issues, smart card data for planning and the future of public transport. This comprehensive reference work will be a vital source for academics, researchers and transport practitioners in public transport management, transport policy and transport planning. Corinne Mulley was the inaugural Chair of Public Transport at the Institute of Transport and Logistics Studies at the University of Sydney. She is now Professor Emerita at the University of Sydney and Honorary Professor at Aberdeen University. Corinne is a transport economist and is active in transport research at the interface of transport policy and economics, in particular on issues relating to public transport. She has provided both practical and strategic advice on transport evaluation, including economic impact analysis, benchmarking, rural transport issues and public transport management. Corinne’s research is motivated by a need to provide evidence for policy initiatives, and she has been involved in such research at local, state/ regional, national/federal and European levels. John D. Nelson is Chair in Public Transport at the Institute of Transport and Logistics Studies (ITLS), University of Sydney, which he joined in 2019 from the University of Aberdeen, where he was Sixth Century Chair in Transport Studies and Director of the Centre for Transport Research. Before moving to Aberdeen in 2007, he was Professor of Public Transport Systems at Newcastle University, UK. John is particularly interested in the application and evaluation of new technologies to improve transport systems (with a particular focus on public transport and shared transport solutions) as well as the policy frameworks and regulatory regimes necessary to achieve sustainable mobility. He is Series Editor for Routledge’s Transport and Mobility and Transport and Society book series and recently edited a special issue on the future of public transport for Research in Transportation Business and Management. Stephen Ison is Professor of Air Transport Policy within Leicester Castle Business School at De Montfort University, Leicester, UK. He has published widely in the area of transport policy and economics and has edited, authored, or co-authored 18 books in the area and published over 130 peer-reviewed journal papers. He is a member of the Scientific Committee of the World Conference on Transport Research, Editor of the journal Research in Transportation Business and Management, Associate Editor of the journal Transportation, Planning and Technology and Book Series Editor of Transport and Sustainability.

THE ROUTLEDGE HANDBOOK OF PUBLIC TRANSPORT

Edited by Corinne Mulley, John D. Nelson and Stephen Ison

First published 2021 by Routledge 2 Park Square, Milton Park, Abingdon, Oxon OX14 4RN and by Routledge 52 Vanderbilt Avenue, New York, NY 10017 Routledge is an imprint of the Taylor & Francis Group, an informa business © 2021 selection and editorial matter, Corinne Mulley, John D. Nelson and Stephen Ison; individual chapters, the contributors The right of Corinne Mulley, John D. Nelson and Stephen Ison to be identified as the authors of the editorial material, and of the authors for their individual chapters, has been asserted in accordance with sections 77 and 78 of the Copyright, Designs and Patents Act 1988. All rights reserved. No part of this book may be reprinted or reproduced or utilised in any form or by any electronic, mechanical, or other means, now known or hereafter invented, including photocopying and recording, or in any information storage or retrieval system, without permission in writing from the publishers. Trademark notice: Product or corporate names may be trademarks or registered trademarks, and are used only for identification and explanation without intent to infringe. British Library Cataloguing-in-Publication Data A catalogue record for this book is available from the British Library Library of Congress Cataloging-in-Publication Data Names: Mulley, Corinne, editor. | Nelson, John D., editor. | Ison, Stephen, editor. Title: The Routledge handbook of public transport/edited by Corinne Mulley, John D. Nelson, and Stephen Ison. Description: New York: Routledge, 2021. | Includes bibliographical references and index. Identifiers: LCCN 2020045890 (print) | LCCN 2020045891 (ebook) | ISBN 9780367418724 (hardback) | ISBN 9780367816698 (ebook) Subjects: LCSH: Transportation and state. | Transportation – Environmental aspects. | Transportation – Planning. | Transportation – Technological innovations. | Civil engineering. Classification: LCC HE193. R688 2021 (print) | LCC HE193 (ebook) | DDC 388—dc23 LC record available at https://lccn.loc.gov/2020045890 LC ebook record available at https://lccn.loc.gov/2020045891 ISBN: 9780367418724 (hbk) ISBN: 9780367747244 (pbk) ISBN: 9780367816698 (ebk) Typeset in Bembo by Apex CoVantage, LLC

CONTENTS

Prefaceix Contributorsxii PART A

Policy context

1

  1 Regulatory frameworks in public transport including tendering David A. Hensher

3

  2 Public transport governance Fabio Hirschhorn and Wijnand Veeneman

21

  3 Mobility as a Service and public transport Göran Smith

33

  4 Public transport use: the ‘soft’ side of the story Veronique Van Acker, Sazkia Sandoval and Mario Cools

46

  5 Intercity modal competition Roger Vickerman

61

  6 Public transport integration Lucy Budd and Stephen Ison

72

  7 Intermodal connections Bianca Eiben and Jianqiang Cui

82

  8 Public transport and the environment Ka Ho Tsoi and Becky P.Y. Loo

93

v

Contents

  9 Community participation in public transport decision-making Lake Sagaris 10 Land value gains and value capture: the potential for financing public transport infrastructure Corinne Mulley, Barbara T.H. Yen and Min Zhang

107

123

11 Public transport evaluation Glen Weisbrod and Naomi Stein

138

12 Public transport productivity and efficiency assessment Jonathan Cowie

154

PART B

Modal settings

169

13 Bus – from workhorse to thoroughbred Frits Olyslagers, Corinne Mulley and John D. Nelson

171

14 Rail – urban, suburban and regional Simon Blainey and John Preston

186

15 Air passenger services Lucy Budd and Stephen Ison

199

16 Seeking economic and social success with sustainable urban ferry services James D. Frost and Mary R. Brooks

210

17 Flexible transport services John D. Nelson and Steve Wright

224

18 Informal paratransit in the Global South Roger Behrens, Saksith Chalermpong and Daniel Oviedo

236

19 Taxis Matthew W. Daus

252

PART C

Public transport environment

271

20 The health impacts of public transport Melanie Crane and Christopher Standen

273

vi

Contents

21 Demand estimation for public transport network planning Wenzhe Sun and Jan-Dirk Schmöcker

289

22 The first/last mile connection to public transport Christoffel Venter

306

23 Public transport and the built environment Murtaza Haider and Ahmed El-Geneidy

322

24 Intelligent mobility and big data for planning, trust, and privacy Caitlin D. Cottrill PART D

342

Specific public transport delivery issues

353

25 The provision of service information for public transport Nigel Halpern

355

26 Public transport and social inclusion John Stanley and Janet Stanley

367

27 Public transport and travel with dogs Jennifer L. Kent, Corinne Mulley, Laura Goh and Nick Stevens

381

28 Public transport use in later life Charles Musselwhite and Maria Attard

393

29 Parking provision, parking demand and public transport accessibility405 Barbara T.H. Yen and Corinne Mulley 30 Intermodal strategies combining cycling and public transport to improve service and acceptability Lake Sagaris

419

31 Accessibility and design for all Nick Tyler

439

32 Network planning and design Gustav Nielsen

453

vii

Contents PART E

Smart card data for public transport planning

469

33 Smart card data and its use in public transport research: an overview Chinh Q. Ho

471

34 The use of smart card data to analyse platform crowding Loan T.K. Ho, Chinh Q. Ho

479

35 The use of smart card data to analyse railway station waiting times Geoffrey Clifton

491

36 Smart card data and planning for public transport: a case study from South East Queensland, Australia Barbara T.H. Yen

499

37 Variability of passenger travel patterns observed using smart card data in Japan Toshiyuki Yamamoto, Shasha Liu and Toshiyuki Nakamura

507

38 Analysis of skip-and-stop planning using smart card data Seungjae Lee and Madiha Bencekri 39 Analysis on the impact of bus frequency reductions on commuters using smartcard data Yuya Utsumi, Jan-Dirk Schmöcker and Toshiyuki Nakamura PART F

515

523

The future

533

40 Automated vehicles and vehicles of the future Louis Alcorn and Kara Kockelman

535

41 Public transport technology for the future Roberto Palacin

551

42 Smart mobility governance Kate Pangbourne

568

43 Public transport and the future of mobility Glenn Lyons

583

Index594 viii

PREFACE

Public transport is collective transport accessible by the general public, often provided on fixed routes (whether on road or tracks). It includes transport by trains, trams, bus rapid transit (BRT), light rapid transit (LRT), coaches, local buses, aircraft, taxis and the newer forms of ridesharing such as Uber and Lyft. Transport has always been in the public eye, and many influential transport academics have acknowledged this. Denys Munby, for example, claimed “There is no escape from transport. . . . Transport more than any other industry has always been in the public eye and always will be. . . . Almost every transport decision is a public issue” (1968, p. 7). This is as true for public transport today as it was for transport in general over 50 years ago, thus making this Routledge Handbook of Public Transport both timely and highly relevant. Public transport plays a crucial role in promoting sustainability and encompasses policy that is multifaceted, embracing economic, environmental, land use and social aspects, all of which are detailed in this Routledge Handbook of Public Transport. Public transport also has an important spatial dimension, with issues relating to the urban, peri-urban and rural form each sharing common problems but also exhibiting distinctly different challenges. Within urban areas in particular, public transport is central to the mass movement of people from A to B and is land efficient in doing so – for example, a full bus of 65 passengers carries the equivalent of 50 cars (with 1.25 average occupancy), with trains and metros even more space efficient than buses for mass transit, both above and below ground. Public transport is an important area of study; it is problem rich, embracing a range of issues stimulating a great deal of debate in areas which impact everyday lives. The delivery and management of public transport involves regulatory frameworks, environmental impacts, choice of technology, market structure, productivity considerations and debates on private versus public ownership from a political and economic efficiency perspective. Very few sectors demonstrate such a breadth of issues, problems and solutions. The chapters in this Handbook seek to highlight these often complex issues, providing a state-of-the-art research-based approach to public transport. Transport developments have throughout time underpinned important technological and social change. Theo Barker (1996) highlights how changes from the industrial revolution onwards have been associated with transport changes that have had a great impact on society. As a study area, public transport is now at an important crossroads. Technology developments mean that the whole area of public transport is developing rapidly; new innovations or ix

Preface

technologies – for example, the field of autonomy, or attitudes towards sharing – will set in train dramatic changes in the way in which public transport is organised and used. Smart technology developments, for example, have disrupted many aspects of collective transport to provide a more personalised transport experience. In editing this Handbook of carefully invited chapters, the editors recognise that the area of study is broad. The contributions are truly international, with 63 authors from 19 countries. The Handbook has been developed to be complementary to existing textbooks on public transport, providing a more in-depth context and research-informed focus on topics. Some chapters have been based on original research, while others afford an informed state-of-the-art review of the literature pertaining to specific aspects of the public transport environment and debate. Each chapter aims to set out a research agenda demonstrating how the future of public transport can be influenced by evidence-based research. It is also anticipated that these contributions will be useful to those tasked with formulating future policy relating to passenger transport.

Book sections and key topic areas Through the process of planning this Handbook, the editors identified six key topic areas of focus, namely: Part A: Policy Context covers areas which define the space in which public transport operates. This area demonstrates how policy in some areas is very context specific, but there are aspects of specific advice which are more generic (governance, environment, evaluation, intermodal connections, integration, land value uplift, efficiency measurement). Other aspects of policy show more of a developmental approach, with chapters exploring a number of contexts (regulatory frameworks, Mobility as a Service, the ‘soft’ side of public transport demand, community participation). Part B: Modal Settings covers in more detail the specific modes that make up the public transport ecosystem. These chapters set modal development in its historical setting, identify distinguishable modal issues and look to the future development of the modes within a wider setting. Part C: Public Transport Environment focuses on those areas that many would not normally expect to be part of a Handbook of Public Transport but which warrant inclusion so as to illustrate the interconnectedness of public transport issues and other aspects of complexity ranging from the first/last mile connections to public transport and the health impacts of public transport. Part D: Specific Public Transport Delivery Issues aims to cover aspects of public transport delivery which are perhaps more specialised but determine, in one way or another, the success of public transport delivery. Public transport delivery is problem rich, and this area serves to show the wide diversity of issues that are recognisable as issues of substance. This ranges from the information needed by travellers to use the system successfully to a consideration of how making public transport more pet friendly might help deliver a more sustainable city, as well as how to design good public transport networks, specific implementation to aid older travellers and issues around public transport information. Part E: Smart Card Data for Public Transport Planning begins with an overview of the role of smart card data in underpinning various aspects of planning. The overview chapter is followed by a diverse set of case studies from different parts of the world illustrating the types of problems that can be addressed with this data. Part F: The Future contains chapters more specifically framing how the future of public transport might look. These chapters are the culmination of the diverse set of topics addressed by earlier chapters. x

Preface

This Handbook serves to show that public transport continues to be a complex, rapidly changing area, never so much as at the present time, given the COVID-19 pandemic which emerged during the preparation of this Handbook and which has greatly impacted public transport in three important ways: first, in that many people have been furloughed or are working from home and as such have had no need to use public transport; second, the need to self-distance so as to avoid infection, which has rested uneasily with the use of public transport as a mode of travel and third, in relation to aviation, there has been a need to prevent travel so as to seek to halt the spread of COVID-19. Where relevant, the impact of COVID-19 is addressed in each chapter and the topic areas covered in this Handbook. Editors of Handbooks are always criticised both for their selection of topic areas to include and for their omissions. Given the breadth of public transport issues and the debate they engender, we do not expect to escape this charge. However, we have collected in one place a farreaching set of chapters which can rightfully claim to exemplify the importance of public transport in today’s world and in the world into the future. We sincerely hope that you enjoy reading this Handbook just as much as we have enjoyed editing it. As a final note, we have some acknowledgements to make: This Handbook would not exist without our contributors, whom we must thank for so willingly contributing their expertise to this timely venture. Without exception, they have shown patience and responded quickly to the many questions asked from demanding editors. We are indeed grateful for their support and input. We thank Guy Loft from Routledge, whose expertise identified the gap in the market that this Handbook addresses and whom we must thank for providing us with the opportunity to fulfil the role as editors in its production. It has been a short year from first asking authors to contribute to the delivery of the manuscript. We need also to thank Matthew Ranscombe, also from Routledge, who has patiently supported us and answered our many queries. Apart from our authors, there is a long list of individuals who have contributed behind the scenes – too many to list here, but they know we know who they are – and we offer our thanks to them for lightening the load for us. At a very late date, one of the authors, Nick Stevens, stepped in and used his Photoshop skills to make sure all the figures were to the required resolution  – a task that had left us perplexed. And, finally, we have reflected on the undertaking we feel is now completed. Working together as editors on opposite sides of the globe has been an enjoyable experience, unaffected by the pandemic and facilitated by technology. Corinne Mulley, Sydney, Australia John D. Nelson, Sydney, Australia Stephen Ison, Leicester, UK

References Barker, T. (1996). The world transport revolution. History Today, 46(11), 20–26. Munby, D. (1968). Transport. Penguin.

xi

CONTRIBUTORS

Louis G. Alcorn works as a consultant at WSP, specialising in transit project development and finance. He holds Master’s degrees in both Transportation Engineering and Community and Regional Planning from the University of Texas at Austin and a Bachelor’s degree in Urban Studies from the University of California, Berkeley. The US Department of Transportation selected Louis as a Dwight D. Eisenhower graduate transportation fellow in 2019. Before graduate school, Louis held roles in transit service planning and capital finance at Capital Metropolitan Transportation Authority and the San Francisco Municipal Transportation Agency, respectively. His research interests focus on improving accessibility through data-driven prioritisation of space-efficient travel modes. Maria Attard is Associate Professor, Head of Geography and Director of the Institute for Climate Change and Sustainable Development at the University of Malta. She is Co-Editor of Research in Transportation Business & Management and Associate Editor of Case Studies on Transport Policy and sits on the editorial board of the Journal of Transport Geography. Between 2002 and 2008, she was a consultant to Malta’s government and helped develop the first white paper on transport policy (2004) and implement the 2006 Valletta Strategy, including Park and Ride, pedestrianisation and road pricing (2007). She also supported the planning for the 2011 public transport reform. She has participated in over 30 projects and is a country expert on the TEN-T ScanMED Corridor Studies and several other EU projects. She sits on the Steering Committee of the WCTR and is a cluster co-chair for NECTAR. She holds a BA Hons and MA in Geography from the University of Malta and a PhD from University College London (UCL). Roger Behrens is a Professor in the Department of Civil Engineering at the University of Cape Town (UCT). He is Director of the Centre for Transport Studies and of the African Centre of Excellence for Studies in Public and Non-Motorised Transport. He graduated with a Master’s in City and Regional Planning from UCT in 1991 and with a PhD in 2002. His current research activities relate to the integration and improvement of paratransit services, the dynamics and pace of changing travel behaviour, the use of transport systems by pedestrians and the urban form prerequisites for viable public transport networks.

xii

Contributors

Madiha Bencekri is a PhD candidate in the Transportation Engineering Department and Research Fellow at the Transportation Planning Lab at the University of Seoul. She obtained her Master’s degree in Urban and Regional Development from the University of Seoul in 2017 and her telecom engineering degree at the National Institute of Post and Telecommunication (Rabat, Morocco) in 2010. She has worked as a government officer at Casablanca Transportation Authority and Casablanca Regional Government. Simon Blainey is Associate Professor in Transportation at the University of Southampton’s Transportation Research Group. He has particular expertise in railway demand and operational modelling and in strategic transport modelling and planning. He has published around 75 articles, conference papers and book chapters. He is programme lead for the University of Southampton’s MSc in Transportation Planning and Engineering and leads the University’s module ‘Railway Engineering and Operations’. He is a member of the governance board for the Data and Analytics Facility for National Infrastructure and a former Secretary of the Royal Geographical Society’s Transport Geography Research Group. Mary R. Brooks is Professor Emerita at Dalhousie University’s Rowe School of Business and a founding editor of Research in Transportation Business & Management. From 2016–2018, she served as Chair of the Marine Board of the U.S. National Academy of Sciences and chaired the 2015–2017 Council of Canadian Academies’ assessment of the value of commercial marine shipping to Canada. Her research focuses on competition policy in liner shipping, port strategic management and short sea shipping. She has authored and published more than 25 books and technical reports, more than 25 book chapters, and more than 80 articles in peer-reviewed scholarly journals. In 2018, she was recognised for her lifetime contribution to the field with the Onassis Prize in Shipping. Lucy Budd is Professor of Air Transport Management and Programme Director of the MSc in Air Transport Management at the Leicester Castle Business School at De Montfort University, Leicester, UK. Owing to her disciplinary background in human geography, Lucy is particularly interested in the relationships between people and places and the role of different transport modes in facilitating changing patterns of human mobility. Saksith Chalermpong is Associate Professor in Civil Engineering at Chulalongkorn University, Thailand, and currently serves as Associate Director of Chulalongkorn University Transportation Institute. His research interests include urban transport planning, public and informal transport and equality issues in transport policy. He received his Bachelor’s degree in Civil Engineering from Chulalongkorn University, his Master’s degree from MIT and his Doctoral degree from UC Irvine, both in the field of Transport. Geoffrey Clifton is a Senior Lecturer in Transport and Logistics Management in the Institute of Transport and Logistics Studies at the University of Sydney. Geoffrey’s research focuses on places and groups for which the traditional models of public transport provision are not adequate. In particular, Geoffrey focuses on the challenge to grow public transport patronage in suburban areas and the provision of transport for people with special needs due to age or disability, including community transport. Geoffrey’s PhD thesis focused on the impact that frequency and connectivity have on the demand for bus services along ‘strategic bus corridors’, whilst other research projects have focused on the use of stated choice methods and optimal pricing theories for public transport planning and management.

xiii

Contributors

Mario Cools is an Associate Professor at the Faculty of Applied Sciences of the University of Liège, where he is responsible for the research domain ‘transport and mobility’ of the research unit LEMA. He holds a Master’s degree in Applied Economics, major quantitative business economics and minor operations research (University of Antwerp, 2004) and a Master’s degree in Applied Statistics (Hasselt University, 2005). After obtaining his statistics degree, he worked as a PhD candidate at the Transportation Research Institute of Hasselt University, where he obtained his PhD in Transportation Sciences in 2009. Mario is author and co-author of scientific publications in research domains such as travel behaviour, transport policy and activitybased travel demand models. Caitlin D. Cottrill is Senior Lecturer in the School of Engineering at the University of Aberdeen and serves as the Director of the Centre for Transport Research. Her primary research interests span the interrelated topics of transport, individual behaviour, technology and data, linked by an underlying commitment to encouraging sustainable and efficient mobility. Her work has a strong focus on facilitating data sharing between transport service providers and travellers in a privacy-preserving manner in order to encourage better decision-making. Application areas for these interests include emerging topics such as intelligent mobility, Mobility as a Service (MaaS) and smart cities. She has additionally worked to ensure that this research takes place in a multidisciplinary context, with collaborators from the areas of computing science, engineering, statistics and information sciences. Jonathan Cowie is Lecturer in Transport Economics at Edinburgh Napier University, UK, and previously held lecturing posts at University College Scarborough and Glasgow Caledonian University. He is the author of The Economics of Transport, published by Routledge in 2009, and joint editor of the Routledge Handbook of Transport Economics, published in 2017. Jonathan has authored many conference and journal papers in the area of supply-side economics of transport services (both public and freight) and is a Fellow of the Higher Education Academy and longtime member of the Scottish Economic Society. Melanie Crane is a Research Fellow for the Australian Prevention Partnership Centre and is based in the Sydney School of Public Health at the University of Sydney. She is also part of an international Wellcome Trust–funded project investigating complex urban systems for sustainability and health to inform urban transformation. Her research and teaching extend across non-communicable disease prevention; active travel; urban sustainability and complex health prevention intervention, and evaluation. Dr Crane has recently led two independent reports on the health impact assessment of new public transport infrastructure for the State Government of New South Wales, Australia. Jianqiang Cui is Lecturer in Planning at Griffith University, Australia. Dr. Cui’s research interests lie in the fields of urban and environmental planning, transport planning and travel behaviour. Her research has appeared in a number of high-impact scholarly journals such as Transport Reviews, Urban Design International, Habitat International, International Planning Studies and Transport Policy. She serves as Associate Editor of Tunnelling and Underground Space Technology (Elsevier). She is also Associate Editor of Travel Behaviour and Society (Elsevier). Matthew W. Daus currently serves as Transportation Technology Chair at the City University of New York’s (CUNY) University Transportation Research Center of the City College of New York, where he conducts research and continues to be extensively published as an expert xiv

Contributors

on ground transportation regulation and technology. As CUNY Distinguished Lecturer for the past nine years, he taught courses on transportation history, policy, sustainability, for-hire regulation and technology. Mr. Daus has served for the past ten years and continues to serve as President of the International Association of Transportation Regulators (IATR), a non-profit educational and advocacy peer group of government transportation regulators from around the world promoting best regulatory and innovative practices. Commissioner Daus was the longestserving Chair of the New York City Taxi and Limousine Commission (TLC), serving for 8½ years. Prior to his tenure as Commissioner, Mr. Daus served in executive and other positions in NYC government for almost 20 years at several agencies including as General Counsel to the TLC and the NYC Community Development Agency, as Special Counsel to the TLC and NYC Trade Waste Commission, as a NYC Human Rights Prosecutor, and as Commissioner of the NYC Civil Service Commission. Mr. Daus is a partner and currently chairs the Transportation Practice Group at Windels Marx Lane & Mittendorf, LLP. Bianca Eiben works at the Department of Transport and Main Roads, Queensland Government, Australia. Bianca graduated from Griffith University with a Bachelor of Urban and Environmental Planning, Class I Honours for her thesis Planning for Public Transport: A Study of the Gold Coast Light Rail Park ‘N’ Ride. Through her time at university, Bianca developed a passion for transport and sustainability. Ahmed El-Geneidy is Full Professor at the School of Urban Planning, McGill University. He is also Chair of the World Society of Transport and Land Use Research. His research interests include land use and transport planning, public transport operations and planning, travel behaviour analysis including both motorised (car and public) and non-motorised (walking and cycling) modes of transportation, travel behaviour of disadvantaged populations (seniors and people with disabilities) and measurements of accessibility and mobility in urban contexts. James (Jim) D. Frost is a marketing and business development specialist with 35+ years’ experience in port marketing, container shipping, short sea shipping and the ferry industry. A consultant for the past 25 years, he has managed a container feeder/transhipment service operating between the traditional trading links of Halifax and Boston and was also marketing manager for Marine Atlantic, a large crown corporation which operates ferry services in Atlantic Canada. Jim has a BA from McGill, an MA from Queen’s and an MBA from Saint Mary’s University. He has authored three books and many articles on the business and economic history of the Maritime Provinces of Canada, as well as several articles on short sea shipping and gateway development. Recent projects have included one of 31 studies commissioned for the Canadian Transportation Act Review as well as a study examining opportunities for hub-and-spoke transshipment services in the five regions of Canada. Laura Goh is a Research Officer at the Sydney School of Architecture, Design and Planning at the University of Sydney. Laura has worked as an urban planner in government, and her research interests focus on the impact of changing urban governance and management structures on legislation and policy in the areas of health, transport, housing, heritage and public land across different jurisdictional contexts. Past projects have focused on the impact of neoliberalism on public land use, emerging regulatory issues related to Commonwealth land management and the implementation of planning strategies to improve health in higher density living. Laura is also a passionate educator and has lectured in planning in Australia for the last decade. xv

Contributors

Murtaza Haider is a Professor of Real Estate Management at Ryerson University. He also serves as the research director of the Urban Analytics Institute. Professor Haider holds an adjunct Professorship of Engineering at McGill University. He is the author of Getting Started with Data Science: Making Sense of Data with Analytics, which was published by the IBM Press/ Pearson. Professor Haider’s research interests include business analytics, data science, housing market dynamics, transport/infrastructure/urban planning and human development in Canada and South Asia. Professor Haider is a syndicated columnist with Post Media. His weekly column on real estate markets appears nationally in The Financial Post and local newspapers, including Ottawa Citizen, Vancouver Sun, Calgary Herald, Edmonton Sun, and Montreal Gazette. He also writes occasionally for The Globe and Mail and The Toronto Star. Murtaza Haider holds a Master’s in Transport Engineering and a PhD in Civil Engineering from the University of Toronto. Nigel Halpern is Professor of Air Transport and Tourism Management with the Department of Marketing at Kristiania University College in Oslo, Norway. He has previously worked at Molde University College, Norway; the Centre for Civil Aviation at London Metropolitan University, UK; the UK Department for Transport, Local Government and the Regions; the UK Civil Aviation Authority and PGL Travel in the UK, France and Spain. Nigel teaches and conducts research and consultancy in transport and tourism. He is co-author of the Routledge book Airport Marketing (2013) and the Routledge Companion to Air Transport Management (2018). David A. Hensher is Professor and Founding Director of the Institute of Transport and Logistics Studies at the University of Sydney. David is a Fellow of the Australian Academy of Social Sciences, recipient of the 2009 International Association of Travel Behaviour Research (IATBR) Lifetime Achievement Award, Recipient of the 2006 Engineers Australia Transport Medal, recipient of the Smart 2013 Premier Award for Excellence in Supply Chain Management, recipient of the 2014 Institute of Transportation Engineers (Australia and New Zealand) Transport Profession Award and recipient of the 2016 Award for Outstanding Research as part of the inaugural University of Sydney Vice-Chancellor’s Awards for Excellence. David is also the recipient of the 2019 John Shaw Medal, which honours an industry champion who has made a lasting contribution to Australia’s roads. He has published over 675 papers in leading international transport and economics journals as well as 18 books. He has over 60,000 citations of his contributions in Google Scholar and an H-index of 108. Fabio Hirschhorn is a Post-Doctoral Researcher at Delft University of Technology. Fabio draws on mixed methods to analyse the governance and policy making of urban mobility. His main interest is in understanding how the dynamic interplay between political actors, institutional environments and steering mechanisms can influence (and be influenced by) the performance outcomes of passenger transportation systems in metropolitan areas. Prior to working in academia, Fabio spent a number of years designing, implementing and supervising urban transport projects in different countries in Latin America with the World Bank. Before that, he worked as a corporate lawyer in Brazil. Chinh Ho is Senior Lecturer in Applied Spatial Data Analytics at the University of Sydney. He has a research track record of 30+ journal articles in the areas of logistics and transport, statistical modelling and big data. His research interests cover four main areas: integrated land use and transport planning, big data analytics, emerging transport technologies and modelling group decisions. xvi

Contributors

Loan Ho is a Research Associate at the Institute of Transport and Logistics Studies (ITLS) at the University of Sydney. Her expertise is spatial data analytics in public transport network planning and design to obtain insights for decision-making and policy formulation using big datasets, including smart card data, Public Transport Information and Priority System (PTIPS) and open data sources. Stephen Ison is Professor of Air Transport Policy within Leicester Castle Business School at De Montfort University, Leicester, UK. He has published widely in the area of transport policy and economics and has edited, authored or co-authored 18 books in the area and published over 130 peer reviewed journal papers. He is a member of the Scientific Committee of the World Conference on Transport Research, Editor of the journal Research in Transportation Business & Management (Elsevier), Associate Editor of the journal Transportation, Planning and Technology (Routledge) and Book Series Editor of Transport and Sustainability (Emerald). Jennifer L. Kent is a Robinson Fellow in the Urban and Regional Planning Program at the Sydney School of Architecture, Design and Planning. Jennifer’s research interests are at the intersections between urban planning, transport and human health. She specialises in combining quantitative and qualitative data with understandings from policy science to trace the practical, cultural and political barriers to healthy cities. Key issues examined to date include the links between health and higher density living, the interpretation of health evidence into urban planning policy, the health impact of extended commute times and cultural and structural barriers to sustainable transport use. Her findings are policy relevant and have been incorporated into state and federal urban planning agendas. She publishes regularly in highly ranked scholarly journals across the fields of urban planning, public health and transport, and her work is widely cited within these disciplines. Kara Kockelman has been a Professor of Transportation Engineering at the University of Texas at Austin for the past 23 years. She is a registered professional engineer and holds a PhD, MS and BS in Civil Engineering, a Master’s in City Planning, and a minor in economics from the University of California, Berkeley. She has been the recipient of an NSF CAREER Award, Google Research Award, MIT Technology Review Top 100 Innovators Award and several ASCE and WTS awards. She recently served as President of the North American Regional Science Association and on the Eno Center for Transportation’s Advisory Board, as well as being on three TRB committees. She has authored over 180 journal articles (and two books), and her primary research interests include planning for shared and autonomous vehicle systems, the statistical modelling of urban systems, energy and climate issues, the economic impacts of transport policy and crash occurrence and consequences. Seungjae Lee is Professor of Transport Planning at the University of Seoul. He obtained his PhD in the Department of Civil and Environmental Engineering, University College London, in 1994, and worked as a Research Fellow in the Department of Statistical Science, University College London, and then Korea Transport Institute before joining the University of Seoul. He founded the International Journal of Transportation and served as an editor-in-chief (ESCI). He has also served on several editorial boards of SCI/SSCI journals, such as Journal of Advanced Transportation, Transportmetrica, International Journal of Sustainable Transportation, and Proceedings of Municipal Engineer, Institution of Civil Engineering, among others. Shasha Liu received a BS and PhD in Traffic and Transportation Engineering from Beijing Jiaotong University, China, in 2013 and 2019, respectively. She is currently a Post-Doctoral xvii

Contributors

Researcher in the Institute of Materials and Systems for Sustainability, Nagoya University, Japan. Her research interests include transportation and land use, transportation big data analytics and public transport demand forecasting. Becky P.Y. Loo is Professor of Geography, Director of the Institute of Transport Studies and Founding Co-Director of the Joint Laboratory on Future Cities at the University of Hong Kong at the University of Hong Kong. Her core research interests are transportation and e-technologies (defined as microelectronics, informatics and telecommunications). In particular, she has done research on applying spatial analysis, surveys and statistical methods in analysing pertinent issues related to sustainability, regional transport infrastructure and development, transit-oriented development, walkable communities and road safety. Glenn Lyons is the Mott MacDonald Professor of Future Mobility at UWE Bristol – seconded for half his time to Mott MacDonald. Throughout his research career, he has focused upon the role of new technologies in supporting and influencing travel behaviour both directly and through shaping lifestyles and social practices. A former secondee to the UK Department for Transport and more recently to the New Zealand Ministry of Transport, Glenn has led major studies on traveller information systems, teleworking, virtual mobility, travel time use, user innovation, road pricing, public and business attitudes to transport and future mobility. He is now actively examining the future prospects for technological innovations, including driverless cars and MaaS; he is also addressing the challenge of decarbonising transport. Recent and ongoing engagements include helping transport authorities address future uncertainty in their planning, policymaking and investment and examining the need for transport planning practice to evolve. Corinne Mulley was the inaugural Chair of Public Transport at the Institute of Transport and Logistics Studies at the University of Sydney. She is now Professor Emerita at the University of Sydney and Honorary Professor at Aberdeen University. Corinne is a transport economist and is active in transport research at the interface of transport policy and economics, in particular on issues relating to public transport. She has provided both practical and strategic advice on transport evaluation, including economic impact analysis, benchmarking, rural transport issues and public transport management. Corinne’s research is motivated by a need to provide evidence for policy initiatives, and she has been involved in such research at the local, state/regional, national/federal and European levels. Charles Musselwhite is Associate Professor in Gerontology at the Centre for Innovative Ageing at Swansea University. He has a particular interest in improving public policy and practice around the built environment and transportation, taking into account an ageing population, and his research looks at road user safety in later life, giving up driving and creating age-friendly neighbourhoods and communities. He completed a PhD at the University of Southampton in 2004, exploring driver behaviour and risk across the lifecourse. Charles is a member of the British Society of Gerontology (BSG), co-leading the special interest group on mobilities and transport in later life. Charles is Editor-in-Chief for Elsevier’s Journal of Transport & Health and on the editorial board for the Ageing and Society and Research in Transportation Business & Management journals. Toshiyuki Nakamura is a designated Associate Professor at the Institute of Innovation for Future Society at Nagoya University. After obtaining a Master’s degree in 2007, he worked at a Transportation Planning consultant in Tokyo (Institute of Behavioral Sciences, Japan). He xviii

Contributors

received his PhD from Kyoto University, Japan, in 2012, where he also worked as an Assistant Professor from 2011 to 2017. He then moved to Nagoya University, where his research interests include travel behaviour with respect to various transportation modes and data such as smart card and location data. John D. Nelson is Chair in Public Transport at the Institute of Transport and Logistics Studies (ITLS), University of Sydney, which he joined in 2019 from the University of Aberdeen, where he was Sixth Century Chair in Transport Studies and Director of the Centre for Transport Research. Before moving to Aberdeen in 2007, he was Professor of Public Transport Systems at Newcastle University, UK. John is particularly interested in the application and evaluation of new technologies to improve transport systems (with a particular focus on public transport and shared transport solutions), as well as the policy frameworks and regulatory regimes necessary to achieve sustainable mobility. He is Series Editor for Routledge’s Transport and Mobility and Transport and Society book series and recently edited a special issue on the future of public transport for Research in Transportation Business & Management. Gustav Nielsen is an independent expert and former Senior Research Planner at the Institute of Transport Economics, Norwegian Centre for Transport Research, in Oslo, Norway. He was educated as a Master of Civil Engineering at the Norwegian University of Technology, Trondheim, and Master of Town Planning at Heriot-Watt University, Edinburgh. For many years, he has been a recognised researcher, planner and lecturer in the field of environment-friendly transport, public transport and urban planning in Scandinavia. He has worked through national, Nordic and international projects as researcher and research leader, consultant and advisor to central and local government and the private sector. He is currently working on a book on network planning as a means to car-free mobility. Frits Olyslagers is an urban transport system development specialist with extensive global experience of more than 40  years’ specialisation in bus operations and 25  years in international bus systems development, involving urban transport policy, institutional frameworks and organisational reform, planning and implementing bus systems and BRT, including business and operations planning and capacity building. He holds a Master’s degree in Development Practice and is a Certified Practicing Project Manager with a background in planning and managing bus operations in Australia, Pakistan and Vietnam. Internationally, he is extensively involved in strategic planning for urban transport and commercial bus system development. Through institutional reform and a commercially oriented business approach, Mr. Olyslagers works to develop financially sustainable strategies for public transport within the framework of network management, contract partnerships and sound risk management. In senior management roles, he has undertaken change management to restore bus transport operations to profitability through reorganisation, efficient route scheduling and cost management. Daniel Oviedo is Assistant Professor in Urban Transport and Development Planning at the Bartlett Development Planning Unit of University College London. Daniel holds a Bachelor’s degree in Civil Engineering and a Master’s in Transport Engineering and Planning from Universidad de los Andes in Colombia and a PhD in Development Planning from UCL. His research focuses on the social consequences of urban transport in global south cities, particularly accessibility, inequality and social exclusion; the role of informal transport in urban development and inclusion and the effects of innovations and disruptions in transport in accessibility and sustainability. xix

Contributors

Roberto Palacin is a Reader in Transport Futures at Newcastle University and has a background in mechanical engineering, design and railway systems engineering. Over the past two decades, he has been involved in research on subjects such as strategic development of transport systems, energy efficiency of urban and mainline rail systems and conceptualisation and assessment of new urban mobility approaches. His research interests revolve around two main aspects, application of a systems approach to transport energy conservation and human-systems interaction. It includes mobility and mass capacity in the context of MaaS (urban and long distance), decarbonisation of transport systems, network connectivity, influence of policy, and the development of ergonomic and design-led urban mobility environments. Kate Pangbourne is a University Academic Fellow in the Institute for Transport Studies, University of Leeds, UK. She has an MA (Hons) in Philosophy with English literature (Edinburgh), an MSc in Sustainable Rural Development (Aberdeen) and a PhD in Scottish transport governance (Aberdeen). Pre-academia, she worked for Scottish Natural Heritage, the advisory body to the Scottish government on conservation of the natural heritage and sustainable development. With an interdisciplinary background encompassing environmental sustainability, transport geography, technology, social science and philosophy, she addresses changing behaviours, practices and governance. Her current research investigates understanding smart mobility discourse and governance as well as messaging to influence travel behaviour using persuasive technologies. She currently holds a five-year Living with Environmental Change Fellowship award from the Engineering and Physical Sciences Research Council to research and progress persuasive communications for sustainable travel behaviour change (www.adapt.leeds.ac.uk). John Preston is Professor of Rail Transport and Head of the Transportation Research Group at the University of Southampton. His research in transport covers demand and cost modelling, regulatory studies, land-use and environment interactions and economic appraisal and evaluation. His initial work concentrated on rail, but subsequent work has covered all the major modes of transport. He has held around 160 research grants and contracts, worth almost £22 million, and has published around 370 articles, book chapters and conference and working papers. He has provided advice on railway matters to the UK Passenger Demand Forecasting Council, the House of Commons Transport Select Committee, the National Audit Office, the Welsh Government, the Korean Railroad Research Institute and the International Transport Forum. He is a co-chair of the World Conference on Transport Research Society’s (WCTR) Rail special interest group, a committee member of the International Association of Rail Operations Research (IAROR) and a member of the Future Traffic Regulation Optimisation (FuTRO) Project Control Board and the Vehicle/Train Control and Communications Systems Interface Committee. Lake Sagaris is Associate Adjunct Professor in Transport Engineering, crossposted to the Institute for Sustainable Development, and researcher affiliated with the Center for Sustainable Urban Development (CEDEUS), Centre of Excellence in Bus Rapid Transit, Pontificia Universidad Católica de Chile. She is an internationally recognised expert on civil society, gender and cycle-inclusive urban planning as they relate to urban-regional governance for transitions toward more sustainable, socially just cities. An award-winning writer, she began her working life in Chile in 1980, with the London Times, Toronto Globe and Mail and other media. She holds a Master’s of Science and a PhD in Urban Planning and Community Development (University of Toronto). She uses participatory action research and community-government partnerships to mobilise “ecologies of modes and actors” to transition toward more sustainable transport. These xx

Contributors

experiences have led to awards, most recently a distinction as Remarkable Woman in Transport, and presentations in Latin America, Europe, Canada, Taiwan, the United States and India. Sazkia Sandoval is an Ecuadorian who graduated with honors in 2013 from Pontificia Universidad Católica del Ecuador in the field of Business Engineering with a specialisation in Marketing. In 2015, she studied an MBA with a specialisation in international management at KU Leuven, Belgium, in which she graduated magna cum laude. Her Master’s thesis focused on how culture can influence people’s transportation decisions. Back in Ecuador, she worked in a retail company as a brand manager. Recently, she moved to El Salvador and runs her own recycling project, focusing on sustainable roof manufacturing from recycled materials. Jan-Dirk Schmöcker is an Associate Professor within the School of Global Engineering and the Department of Urban Management at Kyoto University. He studied at Technical University Berlin, Newcastle University and Imperial College London, from where he obtained his PhD. He further worked at Tokyo Institute of Technology. His main research topic is modelling of passenger behaviour in networks, including aspects such as fare structures, crowding, bunching and real-time information. Jan-Dirk’s current research further looks at long-term demand adaptation to transport infrastructure investments, technological advances and trends associated with “shared mobility”. He co-edited a book titled Public Transport Planning with Smart Card Data. Göran Smith is a Senior Researcher at RISE Research Institutes of Sweden. He studies how digital innovations initiate, facilitate and shape sustainability transitions. In particular, his research is focused on understanding what governments can do to pave the way for emerging mobility services while also ensuring that they contribute to making transport systems less car centric. Göran, moreover, works as a Regional Developer at Region Västra Götaland and holds an Honorary Senior Principal Research Associate position at the Institute of Transport and Logistics Studies at the University of Sydney Business School. Christopher Standen is a Research Fellow in Applied Urban Development at the School of Population Health at the University of New South Wales. His research interests span transport, housing and public health – including travel behaviour, road safety and health equity. He has a particular focus on enhancing assessment methods for urban development projects and policies to better capture positive and negative health and wellbeing impacts, and to ensure they are more equitably distributed. Janet Stanley is Associate Professor, Principal Research Fellow at the Melbourne Sustainable Society Institute, Faculty of Design, University of Melbourne, and Visiting Professor at Hiroshima University. Her work focuses on the interface between social, environmental and economic issues across policy and system design and at community levels, particularly specialising in research on urban planning, transport, climate change, bushfires, children and youth wellbeing and social exclusion. She has approximately 150 publications, with five co-authored or edited books, including How Great Cities Happen: Integrating People, Land Use and Transport, Edward Elgar, UK. Her latest book is on the prevention of wildfires. Janet is a board member of the Mornington Peninsula Foundation, supporting disadvantaged children. John Stanley joined the Institute of Transport and Logistics Studies (ITLS) at the University of Sydney in July 2008 as Adjunct Professor. Prior to taking on this role, he had nine years as Executive Director of Bus Association Victoria, after eight years as Deputy Chairman of the Australian xxi

Contributors

National Road Transport Commission. He was a member of the Ministerial Advisory Committee advising Victoria’s last two planning ministers on Melbourne’s long-term Metropolitan Planning Strategy and is currently Chair of People and Parks Foundation. John has published widely on transport and land use policy and planning and is co-author of the books An Introduction to Transport Policy and How Great Cities Happen and co-editor of A Research Agenda in Transport Policy. Naomi Stein is a principal at EBP US. She works with communities to incorporate economics into the transportation and infrastructure planning and investment process. She is a leader in the applied practice of transit economics and prioritisation in the United States. Her experience includes co-authoring guidance for the American Public Transportation Association on transit economic evaluation, as well as documenting best practices in the same for the Transportation Research Board. She has assessed the economic role of transit and analyzed the benefits and impacts of future transit investment scenarios for regions across the United States, including in Virginia, Nevada, Colorado and Georgia. Naomi is the Principal Investigator on a current national research project on improving prioritisation methods for public transportation investments. Prior to joining EBP, she was a researcher in the Regional Transportation Planning & High-Speed Rail Research Group at MIT. Nick Stevens is a statistician most recently specialising in data and survey analysis in the field of transport. Following a number of commissions for Transport for New South Wales and for Australian bus companies, he has most recently been involved in analysis for projects at Sydney University. He has taught master’s courses at the University of Sydney and Macquarie University. He is now located in the United Kingdom. Wenzhe Sun received a BE degree from Tongji University, China, in 2013 and ME and PhD degrees from Kyoto University, Japan, in 2016 and 2020, respectively. He is currently a PostDoctoral Researcher in the Laboratory of Intelligent Transport Systems at Kyoto University. His research interests include data-driven public transport planning and operation, machine learning applications on predictive transit operation and automated bus transit. Ka Ho Tsoi is a PhD candidate in the Department of Geography of the University of Hong Kong. His research interests are sustainable mobility and travel behaviour, accessibility and landuse planning and space-time analysis. Nick Tyler is the Chadwick Professor of Civil Engineering and the Director of the UCL Centre for Transport Studies. He explores people’s interactions with the environment through city-based projects around the world and life-scale experiments in his laboratory in London. He is particularly involved with the design and operation of public transport systems which are accessible to all. He is interested in combining detailed neuro-/physiological activity and the conscious responses of people to their environment in order to create more acceptable and accessible design of urban space. His laboratory extends this approach to study larger and more complex environments and interactions, such as city streets, public squares, stations, trains and buses, and their interactions with people in terms of their behavioural, physical, psychological, physiological and neurological responses. Yuya Utsumi was a Bachelor’s and Master’s student in the Intelligent Transport Laboratory, Department of Urban Management, Kyoto University, from 2017 to 2020. He has now graduated and is working for the Japanese National Expressway Co-operation (NEXCO). His xxii

Contributors

co-edited chapter forms part of his research and is part of a series of work in collaboration with Shizutetsu. He has further researched tourism behaviour of Thai visiting Japan and their attitudes to recommending their experiences to other travellers. Veronique Van Acker is a Researcher at LISER, Luxembourg, in the Urban Development and Mobility Department. She is also a Guest Professor in Spatial Analysis at Ghent University, Department of Geography. Veronique’s research concentrates on the interaction between the built environment and travel behaviour. Her research topics include, among others, the importance of soft factors such as lifestyles and attitudes; behavioural change towards sustainable mobility; travel satisfaction and wellbeing; peak car and differences between generations and acceptance, use and spatial/social impacts of new mobility technologies. Wijnand Veeneman is Associate Professor of Governance of Infrastructures at Delft University of Technology and Scientific Director of Next Generation Infrastructure, a knowledge platform of six major Dutch infrastructure managers. He leads a group that looks at the relation between governance and infrastructure system performance, from cyber to energy, from efficiency to sustainability. His own research focuses on understanding that relation in public transport and broader mobility services. His research is targeting designing the way that decisions are made to improve the value of infrastructure systems for society. Christoffel Venter is a Professor in Civil Engineering and a researcher at the Centre for Transport Development at the University of Pretoria in South Africa. His research and teaching interests centre on the planning, design, and operations of multi-modal transport systems in countries of the global South, with a recent focus on issues of access to public transport, first/ last mile issues and their impacts on the behaviour of current and potential public transport passengers. Most of this work has been conducted with support from the Bus Rapid Transit Centre of Excellence, sponsored by the Volvo Research and Educational Foundations (VREF). Roger Vickerman is Emeritus Professor of European Economics at the University of Kent and Visiting Professor at the Transport Strategy Centre, Imperial College London. He has published widely on many aspects of transport and is the author of six books and over 200 papers. He was Editor-in-Chief of the journal Transport Policy 2010–2016 and is currently Editor-inChief of Elsevier’s Encyclopaedia of Transportation. He has served as an advisor to parliamentary committees in the United Kingdom and as a consultant to the European Commission, various government departments and regional and local government authorities in the United Kingdom and worldwide. He was awarded the Jules Dupuit Prize of the World Conference on Transport Research Society in 2016. Glen Weisbrod is Chairman of EBP US, formerly Economic Development Research Group. His specialises in the use of benefit-cost, economic impact and program evaluation techniques for all modes of transportation, particularly public transportation. He has authored reports and led workshops on these topics for agencies spanning the United States, Canada, Scotland, England, Netherlands, France, Australia and Japan. He was previously Chair of the US Transportation Research Board’s Committee on Transportation and Economic Development and also served on its Committees on Transportation Economics, Land Development and Social and Economic Factors. For the American Public Transportation Association, he co-authored its guide to communicating the economic benefits and value of local transit and a national report on the economic impact of public transportation investment. xxiii

Contributors

Steve Wright is a Research Fellow in the Centre for Transport Research (CTR) at the University of Aberdeen. Following a PhD from Newcastle University and subsequent research work in traffic control and network flow modelling, he has more recently been involved in a variety of projects related to the planning, operation and evaluation of flexible and shared transport alternatives to private car use and solutions for the first/last mile problem (e.g. demand responsive transport services, taxis including shared taxi services, carpooling, feeder bus services and automated vehicle solutions). His special interest lies in the application of new and proven techniques to facilitate, and expand the market for, such services. Toshiyuki Yamamoto is a Professor in the Institute of Materials and Systems for Sustainability, Nagoya University, Japan. He received a BS, MS and PhD from Kyoto University, Japan in 1992, 1994 and 2000, respectively. His research interests include vehicle ownership and use, vehicle sharing systems, travel behaviour analysis, time use and activity-based analysis, intelligent transport system and traffic safety. Barbara T.H. Yen is an Assistant Professor in the Department of Transportation and Logistics Management, National Chiao Tung University, Taiwan. Barbara is a transport economist, specialising in behaviour analysis, public transport, value capture, travel demand management and performance evaluation. Barbara has been awarded a significant amount of competitive funding from government and industry to carry out research studies in both Taiwan and Australia. These have included studies on public transport travel behaviour management, value capture and road safety. Most recently (2021), work on smart card analysis in Taiwan won second best prize in a competition hosted by Taipai City Government. Min Zhang is a Post-Doctoral Fellow at the School of Earth and Environment, University of Queensland, having been awarded a PhD in Transport Planning in May 2019. Her primary research interests are in using quantitative methods, such as geographic information systems (GISs), big data mining and modelling and econometrics, to study issues in urban and transportation planning. These issues cover value capture, transport behaviour analysis and urban mobility.

xxiv

PART A

Policy context

1 REGULATORY FRAMEWORKS IN PUBLIC TRANSPORT INCLUDING TENDERING David A. Hensher

Introduction This chapter focuses on economic regulation, which has an overarching role to ensure that prices and quantity, including quality, of services are aligned with objectives such as the public interest in the provision of public transport. Such regulation is exercised on both natural monopolies and market structures with imperfect or excessive competition. Matters of safety and environmental regulation, often referred to as social regulation (see Viscusi et  al., 2005), while also important, are lightly covered given the focus on economic regulation. The range of theories that guide regulatory reform are synthesised before looking closely at specific approaches to procuring services in bus contract design. The chapter complements Hensher (2018), which explores the role of contracting in the delivery of efficient and effective services as a way of revealing the potential strengths and weaknesses of alternative ways to garner greater performance from the delivery of bus services that are primarily under the control of the public sector but which are increasingly delivered by the private sector on behalf of the public sector.

Economic theories of regulation There is an extensive literature on the regulatory framework within which goods and services are provided in every country. Economists have developed sophisticated interpretations of how regulation can be used to guide, or control, the way in which services are provided. The role that various agents play in the public and private sector is controversial, and there is no absolute agreement on these roles, despite there being a large volume of theory and practice documented in support of one or more ways in which services might or should be provided. Within the public transport sector, the ambiguity and disagreement remain despite efforts over the last 30 years at least to both promote and reform the way in which public transport is provided. The full spectrum has ranged from public monopoly (nationalisation) to economic deregulation, with competition for the market through tendering being used as a compromised way of controlling the market (in lieu of negotiated contracting) while aspiring to a cost-efficient (and desirably network-effective) outcome that befits a competitive setting. Before taking a closer look at the structural change in the provision on public transport, drawing on examples in bus supply in 3

David A. Hensher

various geographical jurisdictions, some of the theoretical contributions are synthesised as a guide to what might be best described as aspirational or ideological views of the world of service delivery, which is often drawn on to promote a particular position with respect to the role of economic regulation. Kay and Vickers (1990) make a useful distinction between ‘structural’ and ‘conduct’ economic regulation. Structural economic regulation concerns the regulation of the market structure and includes restrictions on entry or exit (the interpretation associated with competitive tendering); in contrast, conduct economic regulation is used to regulate the behaviour of suppliers and consumers in the market and includes price controls and minimum quality standards, monitored as appropriate through actionable benchmarking. A distinction is often made between public and private interest theories (Den Hertog, 2012). Public interest theories of regulation assume that sufficient information and appropriate enforcement powers exist to ensure that the public interest is enhanced by an essentially benevolent regulator. In contrast, private interest theories promote a position that regulators are not well informed on demand, cost and service quality and tend to be less benevolent, resulting in private self-interest at the cost of the public interest. Intervention by a public sector authority is typically aligned with market failure and the need to support a social welfare outcome through efficient, albeit appropriate, government intervention. Where there is often a lack of appropriate public sector knowledge of an industry sector, there is a case for a contribution from the industry sector itself provided that this complements the role of government and does not singularly promote a private-sector commercial interest that is not aligned with achieving government objectives. The challenge herein is with how it can be ensured that this alignment provides a mechanism to ensure that the less informed public sector becomes more informed through private-sector participation to deliver a social welfare outcome when public funds are at risk (Den Hertog, 2012). Delegation of powers within each sector, but most notably in the government sector, is often a cause of regulatory failure, where ‘expertise’ is in the hands of those with limited and often inaccurate information on the industry sector they are responsible for through implementation of the regulatory framework. The formation in the United Kingdom of specialised regulatory agencies (e.g., Office of Rail) is one way of ensuring the relevant specialised skills, and the focus in Singapore through the Land Transport Authority of expert knowledge of specific sectors is laudable, building significant trust and respect between the principal and the agent. Public interest theories are most often applied to explain regulation in terms of achieving economic (cost) efficiency (Joskow & Noll, 1981, p. 36); however, such theories are also often interpreted more broadly to correct inefficient or inequitable market practices (Posner, 1974; Den Hertog, 2012), designed to achieve a broad socially efficient (including equity or distributional implications) use of scarce resources as opposed to an economically efficient allocation of resources. Although market failure has historically been used to justify government intervention, there has been significant criticism of this position, linked to the failure to recognise transaction costs and the role they play. Transaction and information costs which underlie market failure are assumed to be absent in the case of government regulation (Williamson, 2002). Market failure is a result of a divergence between the price or value of an additional unit of a particular good or service and its marginal resource cost. The theory of second best has demonstrated that the partial aim of efficient allocation does not make the economy as a whole more efficient if unavoidable inefficiencies persist elsewhere in the economy, as is typical in transport with underpriced alternatives (such as the private car) to public transport where there are observed external effects, taxation, imperfect competition and inadequate information. An appealing regulatory 4

Regulatory frameworks in public transport

theory must explain how and why regulation is comparatively the best transaction cost–minimising institution in the efficient allocation of resources for particular goods, services or societal values (Zerbe, 2001). Competition may replace some elements of regulation but not all. In general, the market mechanism itself is often able to produce institutions to compensate for any inefficiencies, with private enterprise developing appropriate ways to avoid adverse selection and quality concerns through best practice performance. The assumption of market failure when a dominant firm supplies the market has been criticised by many authors, notably Demsetz (1968), with significant returns a result of superior efficiency as well as the possibility of competition for the market (Baumol et al., 1982) as opposed to competition in the market. A key concern of the public interest theory is that the normative theory of economic welfare is being used as a positive explanatory theory of economic regulation (Joskow & Noll, 1981; Den Hertog, 2012). Empirical testing of public interest theories relative to private interest theories has concentrated on the effectiveness and not the efficiency of regulations with limited consideration on such matters as are prices lower, is price discrimination absent, is there a reduction in costs, did congestion decline and is the political influence of interest groups identifiable? An alternative perspective was presented by Stigler (1971) that became known as the Chicago theory of government. Stigler argued that economic regulation is designed to benefit specific industry groups, for example, the suppression of transport by the trucking sector to protect the railways, whereas Posner (1971) argued that it benefited consumer groups, for example subsidised prices for all public transport users, even though the cost of such services varies by location (e.g., regional vs. urban, inner and outer urban). Regulatory practices appear to confirm this prediction, where regulated industries are either monopolistic, such as rail transport in most countries, or highly competitive, such as freight and ride-sharing services such as Uber, Ola and taxis (see also Chapter 19). Importantly, in the context of public transport, the theory of economic regulation also predicts that the benefits will take the form of transfers directly through subsidies (typically provider side) or indirectly through price or quantity regulation or restriction to market entry. In Stigler’s view, competitive industries have much to gain from economic regulation and are in a better position than consumers to bring favourable regulation about. In practice, such regulation of competitive sectors is rarely seen. One explanation is found in Becker (1983), whose theory suggested that the loss of economic welfare is greater where the elasticity of service supply is greater and that in competitive sectors, the elasticity of such supply is large (Den Hertog, 2012); hence, the transfers of personal or business income and the welfare losses associated with regulation are so large that the countervailing pressure invoked eliminates any investment in political influence. In more general terms, the criticism of the Chicago model’s view of the power of organised groups has led to a recognition that there are many more influences at work, as promoted through the Virginian School of Public Choice (Tullock et al., 2002), which has a strong distributional focus. In the public transport sector, the winds of change moved in many countries from the late 1980s onwards, linked in particular to the desire to promote cost efficiency and innovation, to decrease the extent of political intervention, to reduce government borrowing (essentially to move debt off the government books, as in the United Kingdom in the Thatcher era), to increase political popularity, to undermine trade-union power (the Thatcher interest), to lower the level of cross-subsidies and to eliminate the inefficiencies caused by uninformed or rentextracting regulators. These developments align to varying degrees with almost all theories of economic regulation and as such might be best described as a hybrid interpretation of the guidance from theory. The following sections trace some of the key reforms in the bus sector, noting that both competition in the market (i.e., economic deregulation) and competition for the market 5

David A. Hensher

(i.e., competitive tendering) or negotiated contracts have a been used as ways to achieve a number of the promoted benefits of reforms under economic regulation (Hensher & Stanley, 2008), where the regulator has varying degrees of knowledge and sophistication with respect to protecting the economic interests of the sector they have societal responsibility for.

Reform in the bus sector What has been learnt over the last 30 years? Although there has been a limited amount of economic deregulation, the dominating approach to reform in the bus industry (excluding the deregulated long-distance coach sector) has been through competition for the market, in large measure due to the claimed natural monopoly nature of services, be they urban or non-urban (see also Chapter 13). While there is clear competition with other modes (notably the car for school travel and the car more generally), the economics of bus services (including patronage levels) has in the main mitigated against competing operators, be it in a route or area-wide geographical setting. The procurement model has been either competitive tendering or negotiation, with the majority of contracts taking on a gross cost form (typically a total cost per service kilometre), although in earlier years, there was a great deal of interest in net cost contracts (Van de Velde & Alexandersson, 2020). The change in focus might best be traced to the reduced attractiveness of a contract where the risks are ambiguous and where growth prospects for patronage were shown so often to be illusory. Regulators have tended, under the influence of lawyers, to over prescribe the details of a contract, resulting in major concerns ex ante about exactly what is required in areas such as service quality and service variations. Although the theory of incomplete contracts suggests that a simplified ex ante contract would ensure greater clarity (Hensher, 2010), it requires acceptance of the willingness on behalf of the regulator and operator to sit down ex post and clarify any points of ambiguity as part of a shared responsibility in delivering an efficient and effective service. This is at the heart of a trusting partnership (Stanley & Hensher, 2008; Hreljaa et al., 2018). For example, the recent five-year bus contracts in Singapore in two tranches (won by Tower Transit and Go Ahead, respectively, in rounds 1 and 2) were gross cost management contracts, with all assets being owned by the public sector (through the Land Transport Authority as the regulator), with explicit wording during the bid phase that the winner will sit down with the regulator and clarify any matters, including possible variations, prior to signing the contract but after the bid submission and award announcement of the successful bidder. In Singapore, the successful bidder was not the least cost bidder but one deemed able to commit to delivering significant improvements in service quality. Throughout this period of significant reform in many countries, notably in Western Europe, Australasia, South Africa and some Latin American and Asian countries that have a desire to move away from public monopoly provision of public transport, there has been a great deal of real-world testing of various regulatory reforms, discussed and summarised in many papers from the Thredbo conference series. There have been different procurement models, contract designs and obligations associated with the operator and the regulator. Often, many of the challenges have been associated with the strategic ‘societal’ role of public transport in contrast to how to improve the (cost) efficiency and (network) effectiveness of services when faced with markets that vary from very sparse to very dense and the associated debate on whether the focus of service delivery should be on spatial coverage and/or (corridor) frequency (Walker, 2011). Importantly, no matter what regulatory model is used to offer up services in the market, the market will remain a very powerful arbiter of its success since it is the setting within which patronage 6

Regulatory frameworks in public transport

arises. The mapping of operators (suppliers) with the preferences of regulators (on behalf of government) who increasingly are taking on the role of transport planners and service designers, in an essentially subsidised non-commercial setting, has become a high-agenda theme in public transport. Arguments have been made in favour of one regulatory framework over another. Before focusing on two important themes, namely incentives in contracts and actionable benchmarking to monitor efficiency and effectiveness, additional comment needs to be provided on gross vs net contracts, tendered vs negotiated contracts and who owns the assets (management contracts in particular). Gross cost contracts are typically defined by a cost per kilometre that is paid to an operator to provide a specific service. The operator typically carries no risk with respect to patronage and service levels and is contracted to provide services and paid for such services (including a margin) regardless of demand. In contrast, a net cost contract typically involves a cost per kilometre under the condition that the fare box revenue is retained by the operator, or at least shared between the operator and the regulator. There is risk sharing here, since the revenue retained is not known and is dependent on both the market and the efforts of the operator. Gross cost contracts dominate in most countries for many reasons (see Merkert & Preston, 2018), most noticeably the greater number of bidders in competitive tenders but also as a way of minimising risk from a regulatory viewpoint when revenue risk under a net cost contract can result in a decline in service quality when revenue targets are not met and, as is often the case, when the financial sanctions are inadequate, such as, for example, in Santiago (Batarce & Ávila, 2019). Aarhaug et al. (2018) provide a comprehensive assessment of Norwegian local bus contracts since 1995. They find that despite cost increases, repeated rounds of tenders attract many more bids and result in noticeable reductions in unit prices. Furthermore, they conclude that the number of bidders is influenced by contract design, in particular the scale of the offer in terms of annual service kilometres, having a gross in contrast to net contract and providing facilities such as depots and parking areas. Importantly, they raise concerns about the future dominance of large multinational operators and warn about the potential loss of competition if this were to occur. Larger-scale contracts increase this risk, making it infeasible for smaller, efficient operators to compete, especially if the contract requires provision of all assets (in contrast to a management contract). Competitive tendering became popular in the early 1990s as a way of breaking the stranglehold of the urban bus sector on inefficient public monopoly supply (see Hensher & Wallis, 2005). Windfall gains of around 30 percent were achieved in round one in many countries simply because the incumbents were never subjected to competition amongst bus service providers or performance targets and associated sanctions. Such windfall gains are never realised beyond the first round, and indeed subsequent rounds typically result in increases in costs (but not to the pre-tendered levels) and a risk of service quality decline (Wallis et al., 2010; Aarhaug et al., 2018), especially where overprescribed service levels are ambiguous through complex contracts. The appeal of negotiated contracts has arisen in part from evidence that incumbent operators are cost efficient and committed to long-term investment in the bus sector, which is always at risk with tendered contracts that have limited contract years, typically five years or less. In addition, there is a strong interest in building trust between the regulator and operators (which still maintain an arm’s-length commercial interest and avoid regulatory capture1) to encourage performance enhancement and investment beyond obligations under a tightly defined and time-limited contract (Stanley & Hensher, 2008). Any support for a negotiated contract must be accompanied by appropriate benchmarks to ensure that a value-for-money outcome is obtained which is effectively equivalent to a tendered outcome or better and where there is provision for tendering under non-compliance. The threat of tendering is a very strong antidote. In order 7

David A. Hensher

to make a case for negotiation, information on performance through benchmarking (to reveal counterfactual evidence) is essential (see Hensher, 2018 and a later section subsequently). Although there is an argument that competitive tendering is transparent whereas negotiation is not and hence attracts greater political support, this is highly questionable. There is growing concern that tendering authorities do not reveal the assessment process in sufficient detail to have confidence in its transparency but also that details of the losers’ bids compared to the winners are never or rarely released. Under negotiation, actionable benchmarking can be used to release the agreed costings, at least to the auditor general or some other party where commercial information is not at risk. Since the funding is typically associated with taxpayer funds, there must be an obligation to ensure that the value-for-money test is transparently applied. Where there has been demonstrated cost efficiency under negotiation or even tendering, there should be a case made to retain the incumbent operator, provided agreed-upon standards are met going forward. Negotiation has an advantage of reduced transactions costs, which are known to be typically much higher through tendering. Hensher et al. (2016) develop a method to correct for differences in risk associated with the incumbents’ and new entrants’ bid price. Within all of the contract settings discussed previously, there are two models of asset utilisation; one with the regulator owning all the assets (as in Singapore, Adelaide and Perth in Australia and many US jurisdictions) and the other where the operator acquires the assets (as in Sydney, Melbourne and Brisbane in Australia). In many settings, this difference can be attributed to the historical context, especially where an operator was initially a public monopoly in contrast to a business initially commenced by a private operator, where the latter almost always provided the assets (vehicles and depots). Asset ownership has interesting implications on risk and the number of bidders, as well as being used by some governments as a way of guaranteeing continuity of service if an operator defaults. By taking away the obligation of asset ownership, it has been suggested that this will result in more bidders and hence a better price achieved in the contracting of a service supplier. The debate on whether assets are owned by the principal or the agent continues to this day; however, some argue forcefully that ownership matters, with rights of ownership of an asset defined as the rights to use the asset, the right to appropriate returns from the asset and the right to change the form and/or substance of an asset (Wong & Hensher, 2018). This argument is aligned with theories around incentives, which are central to efficient contracts and property rights. One position relates to obligations on asset transfer under failed contracts. Specifically, assets are regarded in some settings as essential equipment (e.g., existing rolling stock), and hence there are obligations to pass these assets on to either a new operator who subsequently wins a tender or franchise or an operator brought in as part of a transition to ensure service continuity until a new operator is awarded the contract. An interesting separation in recent years has been between the depot and the buses. In NSW, Australia, for example, the depot is separated out from the rest of the tendered service, which enables a successful bidder who is not the incumbent to initially access the incumbent loser’s depot (claimed by the latter as their private property), but where the access to use fails, government has established an agreement with a local council to provide another site for a depot. In addition, the terms of engagement mean the local council forbids using the incumbent’s site for other developments. New forms of asset ownership in the contracted public transport sector are also emerging. While a regulator owning assets and leasing them to operators can promote standardisation and greater attention to life cycle costs (Nash & Bray, 2014), there is a recognition, however, that the private sector could fulfil such a role more effectively and also better promote innovation. This forms the basis for a new middle ground where the regulator owned some depots/vehicles, whilst the operator owned others, hence promoting innovation at the margin (Nash & Bray, 2014). 8

Regulatory frameworks in public transport

There has been a great deal of focus on the need to build greater trust between the regulator and the operator. A  trusting partnership is seen as particularly important because of the problems posed by incomplete contracts. A  changing market environment makes the complete specification of contractual obligations extremely difficult. Furthermore, much experience (e.g., in many contracts in the Netherlands; see Bakker & van de Velde, 2009) suggests that a contractual focus on such detail discourages operator innovation and encourages an operational focus on cost cutting to increase profits. Where the government and operator(s) work in a trusting partnership, especially at the system design level, the best outcomes are expected to result (see Hreljaa et al., 2018). This is at the heart of relational contracting. This expectation partly reflects the shortage of skilled people and the associated need to draw on all available skills to the maximum extent possible, wherever they are located. It also reflects the expectation that if the government and operator are jointly focused on achieving common goals (patronage and related outcomes), rather than on watching each other, the best patronage outcomes are likely to follow. This notion of a trusting partnership has evolved through the Thredbo conference series as being grounded in five Cs (Hensher & Stanley, 2010): 1 2 3 4 5

common core objectives tied to public policy purposes; consistency of behaviour and direction; confidence in a partner’s capacity to deliver; respect for each other’s competencies; and demonstrated commitment to good faith in making and keeping arrangements and in principled behaviour.

Incentive contracts and why are they rare? The bus contracts discussed previously typically have very few, if any, incentive structures built into them. Drawing on Hensher et al. (2016) to provide a synthesis of the key issues. Hart and Holmstrom (1987) suggest that optimal contracts (incentive contracts) are often extremely complicated, and, indeed, they risk a regulatory nightmare in managing them. In the presence of moral hazard, optimality means inclusion of all relevant information and detailed specification of multiple contingencies. That contracts are usually simple in practice is a result of incomplete information, leading to what Bajari and Tadelis (2001) describe as a ‘nonconvex’ procurement problem resulting in extreme contracts. They argue that there is a fundamental difference between a fixed-price contract and an incentive contract, where a fixed-price contract requires no cost measurement. This leads to a clear discontinuity in the cost of measuring and monitoring costs and implies that fixed-price contracts will dominate contracts that are ‘close’ to fixed price, and as it becomes costlier to measure costs, fixed-price contracts will dominate a larger set of incentive contracts. Similarly, they suggest a fundamental difference between a cost-plus contract and incentive contracts, as there is a risk of costly distortion where incentives are introduced. Therefore, solutions close to cost plus will be dominated by cost-plus contracts. Schwartz and Watson (2004) use a legal framework to explain simple contracts, arguing that contract law, such as the prohibition of contract renegotiation bans, discourages complex contractual forms by making renegotiation relatively inexpensive (Hensher et al., 2016). There is a trade-off between the costs of contract complexity with gains from efficient investment incentives where higher contracting costs result in simpler contracts. In addition, agents have preferences for high or low renegotiation costs depending on the complexity of a contract, where a complex contract requires high renegotiation costs to retain the incentive scheme. 9

David A. Hensher

So, should incentive contracts be used more often? Dye and Sridhar (2005) uses moral hazard severity to suggest that simple contracts can be optimal, despite the potentially vast array of performance measures that are ‘marginally informative’ (see Holmstrom, 1979). Paul and Gutierrez (2005) looking from a practitioner angle, however, disagree. Hensher et al. (2016) investigated this in the bus context. Using a stated choice experiment and a mixed logit model to account for operator preference heterogeneity, the findings offer informative guidance to bus operators (and regulators) in preferences for specific performance-based contracts given their attitudes towards contracts that require them to bear some risk in return for a higher margin when meeting the performance standard specified. Given prior views, reflected in operator preferences, on the likelihood of performance requirements being achieved by the operator, the regulatory authority can use this evidence to identify the likely reaction of bus operators to varying margins and associated bonuses and penalties. This provides valuable insights into the extent to which an operator awarded a contract under the specific financial offer (regulating the specific base margin, bonus and penalty), with the specific risk profile, is likely to prefer and accept the contract. A simple model can be used to construct a number of alternative performance-based contracts (PBCs) which will reveal the contract that is most likely to be preferred by the operator (as well as suggesting contracts where operators would be neither worse off nor better off). The following formula, from Hensher et al. (2016), based on a sample of 64 operators in Australia, can be used by regulators as an additional (behavioural) tool to garner knowledge on the likely support from operators (in a trusted partnership) for specific risk profiles associated with contracts under consideration. Ucontract = 0.5574 × Base (profit) margin as percentage markup on costs + 0.8339 × Bonus under current level of effort % × Chance of occurring − 0.7846 × Penalty under extra effort % × Chance of occurring Knowing the operator level of utility associated with each of the potential PBCs, the regulatory authority can design a contract to give the operator a reasonable incentive and amount of risk for increasing effort and achieving performance requirements. For example, if the current contract is formulated in such a way that all the operator costs are reimbursed and the operators are provided with a stipulated fee for providing the services (i.e., cost-plus contracts), the regulator can estimate the base margin (stipulated fee/operation costs) and feed this into the equation to compute the operator’s current level of satisfaction (or utility). In the next round of negotiation or competitive tendering, if the regulator desires operators to increase their efforts in order to achieve a predefined objective, for example, growing patronage and/or increasing service reliability, the authority can design a contract in which the chances of receiving a bonus/penalty at the current/increased levels of effort are controlled via performance indicators specified via contract clauses. The contract parameters that the authority needs to determine are the level of bonus (consistent with the objective) and penalty (when failing to achieve the predefined objective) required to ensure the operator’s level of utility per the current contract. Vigren and Pyddoke (2020) suggest that although the conceptual literature suggests that passenger incentives could increase public transport ridership, the few studies evaluating real contract outcomes are less positive, as none has yet been able to attribute increased ridership to passenger incentive contracts. As an example of a practical effort to build incentives into contracts, the impacts that contract design has had on the performance of Santiago’s (Chile) public transport system, Transantiago, over the last 13  years are considered, drawing on the findings in Batarce and 10

Regulatory frameworks in public transport

Ávila (2019). Transantiago was implemented in 2007 through a public bidding process that awarded five groups of trunk bus lines and nine groups of feeder bus lines to 11 companies, totalling approximately 4,600 buses. Several contract modifications were made in the early years of operation, including the incorporation of performance indicators for provided capacity, frequency, and regularity. To meet this objective, an analysis of all contracts was conducted, starting with the 2012 renegotiation through the last quarter of 2016, focusing on the structure of the financial incentives in the concession contracts, as these largely determine operators’ performance and the effectiveness of the contract changes (Gómez-Lobo & Briones, 2014). The financial incentives were analysed from the perspective of how operators receive revenue because these determine the quality of the operators’ service delivery (Sappington, 2005). A  system performance audit examining operator revenue, compliance indicators, evolution of payment variables and fines showed that the financial incentives had not been effective. This failure is reflected in a low-quality transport service in which operators achieved steady revenue growth despite their failure to meet expected performance levels. During the period analysed, most of the contractual modifications increased the payment per passenger, rewarding operators financially more than the fines paid for not complying with minimum quality requirements. In addition, the contract incentives were leveraged by operators to benefit themselves and reduce user benefits. For example, operators attained a high performance score for frequency by not stopping at all bus stops, similar to the many settings in which on-time running is the only performance metric subject to fines (as in Australia). Such a contract clearly protects the operators’ revenue, giving them no incentive to improve their services. However, the incentives for adding new vehicles and increasing the size of the bus fleet did have a positive impact, as the buses emitting the most pollution were removed from service, and the fleet grew because of an increase in the amount paid per driven kilometre. Since 2012, the contracts with operators have become more like gross cost contracts with quality incentives, with payment linked to quality compliance. The payable driven kilometres were specified in each operator’s operation plan according to a series of contract requirements. The quality-related requirements were wait times, occupancy rates and overcrowding rates. The problem of incentives to provide quality in the Santiago bus system, however, is that the revenues obtained by providing more quality are lower than the costs incurred. Batarce and Ávila (2019) recommend that incentives, essentially payments per passenger or kilometre, be determined based on the bus operator’s costs and the sensitivity of demand to service quality, a theme promoted by Hensher and Prioni (2002) more than 18 years ago. Hensher and Prioni (2002) developed a way of measuring service quality that results in an intuitively appealing formula that is transparent, is incentive compatible, is easy to administer and monitor and can be integrated into the specification of a competitive tender. Known as the service quality index (SQI), it provides a mechanism to benchmark service quality on a number of criteria and offers various ways to improve on service quality where it falls short of best practice, after controlling for differences in operating environment. An identification of SQI prior to tendering would allow the responsible authority to gain information on customers’ satisfaction with the current levels of service quality and to include this information in the form of service quality targets in the contract specification. Table 1.1, adapted from Hensher and Prioni (2002), gives an example of how one might integrate SQI targets into the tender process. The weights attached to each of the studied service quality attributes2 were obtained from a stated choice preference experiment estimated as a simple multinomial logit model. Let us assume that from a survey of a sample of existing users, the user-defined quality of current service of three operators is identified. Operator 1 achieved an SQI of 1.4 by providing a service that is on average two minutes late, clean enough for 60% 11

David A. Hensher Table 1.1 Including SQI targets in the contact design Current Service description

SQI

Attributes

Target after

Operator

Reliability

Bus fare

Clean enough

Travel time

Etc.

Realised

2.5 yrs.

5 yrs.

1

2 minutes late

2.1

60%

25 minutes

. . .

1.4

1.6

1.8

2

1 minutes late

2.4

78%

26 minutes

. . .

1.3

3

1 minutes late

2.0

80%

21 minutes

. . .

2.0

of the sampled users and costs on average $2.1. Operators 2 and 3 have SQIs respectively of 1.3 and 2.0. Assuming that these operators are comparable, Operator 3 is best practice. Regulators can use the SQI in the contract design to specify how much service improvement they require relative to the current levels as illustrated in the last two columns of Table 1.1. Although one might impose the requirement that each and every bus operator must be at best practice, this may discourage bidders, and so it is preferable to set a target level that is recognised as achievable by potential bidders. The level should be incentive compatible. Given the gap between an operator’s SQI and that of best practice (e.g., 0.6 for Operator 3), a formulation SQI + z, where z is the predesignated improvement over a period of time (e.g., 0.2 in both subperiods), is suggested. The SQI + z formula provides a target in line with a predesignated increase in the service quality level. In the case of the service previously provided by incumbent Operator 1, authorities impose an SQI target of 1.6 after 2.5 years and a final SQI target of 1.8 at the end of the contract (5 years). Other experiences around the world are consistent with the results described previously. In Sweden, for example, Jansson and Pyddoke (2010) analysed the effects of contract incentives on the quality of the transport system in terms of punctuality and cancelled bus and train trips. Their study concluded that the Stockholm transport authority’s incentives were weak, as the monetary amounts of rewards and fines were very small and did not influence operator behaviour. In England, Gómez-Lobo and Briones (2014) demonstrated the importance of including quality incentives in contracts to prevent concession holders from benefiting at the expense of high-quality service delivery. The study revealed significant improvement in the system since 2000, which was when the London transport authority included quality incentives in its contracts for distance travelled and for performance indicators such as schedule compliance and punctuality. The authority also included the incentive of a two-year contract extension if an operator exceeded performance standards. The regulatory experiences of Stockholm and London highlight the importance of including compliance indicators and financial incentives in contracts in order to improve the quality of the system. These indicators must measure compliance with standards for regularity, frequency, agreed-upon demand levels, driven kilometres, fleet status and more, desirably within an SQI framework that accounts for the role that each service dimension plays in customer satisfaction. Implementing this type of contract requires a control and monitoring mechanism that enables the measurement of actual versus expected performance and then linking the performance achieved to penalties and rewards that influence the behaviour of the service providers. This provides an introduction to actionable benchmarking.

12

Regulatory frameworks in public transport

Actionable benchmarking as part of an effective regulatory toolkit Regardless of what procurement model is used, the need to monitor performance remains unquestionable not only to ensure that the agreed-upon performance is maintained (and can be improved on) throughout the duration of a contract but as a way of providing information that the regulator can use to improve practice and especially to develop an actionable benchmarking program to ensure that best practice outcomes in terms of cost efficiency, network effectiveness and customer satisfaction are achieved. Data from the Sydney metropolitan and outer metropolitan areas for 25 contract regions is used to illustrate how this works in practice. The two concepts of particular regulatory interest are efficiency and effectiveness (Table 1.2). Efficiency (doing things right) may be considered in cost per vehicle/service kilometre terms as intermediate outputs in the supply chain. Effectiveness (doing the right things), on the other hand, is a productivity measure which may be related to passenger kilometres/trips delivered per vehicle kilometre. The latter recognises that whilst minimising costs is important, the end objective of public transport is to move people (not just vehicles) for a given budget. To measure the cost efficiency of bus operations, there is a need to identify the features of service provision that incur a disproportionate cost impact on an operating region and are in effect the reality of operating in that specific jurisdiction which the operator has no effective control over. It is therefore necessary to recognise and account for these differences in a process called normalisation (not to be confused with standardisation) to make valid comparisons. Herein lies the difference between the actual gross cost and normalised net cost of bus operations (not to be confused with net cost contracts). In the context of metropolitan and outer metropolitan bus operations in Australia (suspected to be the case in most countries), the main influences that are outside the control of an operator are: (i) the speed on the road (often the result of traffic congestion but also road alignments, including traffic lights and level of bus priority); (ii) the amount of in-service kilometres you can get out of each bus each year (called bus utilisation and influenced by depot location), which impacts the amount of capital and hence capital cost and (iii) the spread of service hours, which can be defined to describe the proportion of service hours on evenings and weekends when higher labour costs associated with penalty rates are typically incurred. The method used to normalise the results is explained in detail in Hensher (2018). The average gross and net cost per in-service kilometre is compared in Figure 1.1 for bus operators for Sydney (Metro and Outer Metropolitan), Brisbane, Perth and Adelaide. Government operators, whose labour costs are generally 25–40% higher due to more rigid work practices, political interference and the sheer staff-to-bus ratio, have been excluded from the sample. In competitively tendered metropolitan Sydney, gross costs are high due to congestion, but on a net cost basis, the bus operators are more effective than in outer metropolitan Sydney, where bus contracts up to 2014 are negotiated. Net cost in Brisbane (where all contracts are Table 1.2 The concepts of efficiency and effectiveness as they relate to bus operations Inefficient

Efficient

Ineffective Pursuing the wrong goals (e.g., satisficing Blind pursuit of cost minimisation at the behaviour) for very high cost expense of customer experience and network effectiveness Effective Achieving the right goals (e.g., Doing the right things right – maximising patronage maximising patronage) but at high cost and enhancing mobility cost effectively

13

David A. Hensher

Figure 1.1 Gross and net cost per service kilometre compared across Australia (ITLS estimate for 2017 based on 2010–11 data)

negotiated) is significantly higher than gross cost, reflecting typically higher average speeds and higher bus utilisation, probably attributable to the bus priority afforded by its busway system (Hensher et al., 2020). Bus services in Perth (all tendered) are the most expensive to procure based on this analysis. The average cost for metro/outer bus operators in Australia for 2017 was $4.58 per service kilometre. The three operating differences can either have a favourable or unfavourable impact on the operating costs of a contract region. Looking now in more detail for all 25 contracts in the Sydney and Outer Sydney metropolitan areas which includes four government operator contract regions, Figure 1.2 shows the degree of variance for each contract region between the operating measures and the industry average. A positive variance means the operating difference has a favourable impact on costs (i.e., reduces costs) relative to the industry average. In simple terms, the effect of normalisation will be to increase cost efficiency which has favourable variances in operating differences relative to the industry average. However, in practice, the overall net effect of normalising for the three operating differences is a complex function of the variances in each operating difference relative to the industry average; the influence each operating difference has on costs, measured through correlation and the gap between the gross cost efficiency and the industry average. Nevertheless, knowing that average peak speed has a major influence on the differences in gross cost efficiency, the direction of change can be anticipated in the normalised or ‘net’ cost efficiency key perfomance indicator (KPI). Specifically, for Contract Regions 5–9 and 13, the net cost efficiency should be reduced to normalise for their unfavourable average peak speeds, and, conversely, the net cost efficiency for Contract Regions 16, 17 and 18 should increase to normalise for their favourable average peak speeds; however, this will be somewhat mitigated by unfavourable vehicle utilisation in the case of Contract Regions 16 and 17. A comparison of gross and net cost efficiency for each contract region is provided in Figure 1.3 for the 2012 financial year, the last year that the author had access to performance data.3 The ranking of industry cost efficiency can be presented in a number of different ways to identify poorer-performing operators: simple ranking of first to last, quartile analysis with identification of contract regions in the bottom quartile or relative to the industry median. 14

Regulatory frameworks in public transport

Figure 1.2 Variances in operating differences compared to industry average

Figure 1.3 Comparison of gross and net cost efficiency KPIs for financial year 2012

Figure 1.4 shows the net cost efficiency relative to the industry median of $5.39 per service kilometre and highlights which operators are potentially inefficient (above the median). The network effectiveness or productivity of the bus and coach industry can be assessed by comparing passengers carried and vehicle kilometres travelled. Network effectiveness (boardings per service kilometre [Figure 1.5]) is a key results area which both government and operator can influence through service review processes and their respective influence on the drivers of patronage growth. In this respect, the operator is best placed to manage service quality and reliability to make bus services more attractive for both existing and new users. The highest network effectiveness is achieved in the four State Transit metropolitan contract areas, reflecting their high-density central business district commuter services and greater patronage potential in their catchment areas. Most other contract areas fall in the range of 0.5 to 1.0 boardings per service kilometre. 15

David A. Hensher

Figure 1.4 Net cost efficiency KPI vs. industry median for financial year 2012

Figure 1.5 Network effectiveness

Only one outer-metropolitan contract area is in the top ten: R20. The five lowest performers are all outer-metropolitan contract regions: R23, R18, R16, R19 and R17. For network effectiveness, around 21 regions have shown a decrease in service kilometres from 2011 to 2012, this following an increase in the previous period, but for many of the regions, the number of boardings has either decreased or not experienced increases of the same magnitude, meaning most regions have shown a decline in this measure.

Conclusions In concluding this chapter, the future is contemplated in terms of where competition and ownership under a regulatory framework in the bus sector might head. Wong and Hensher (2018) 16

Regulatory frameworks in public transport

envisaged a number of issues which may emerge to become important policy areas over the next 30 years, related to multimodal contracts, access contracts, next-generation economic deregulation, intermediate mode regulation, autonomous vehicle regulation and Mobility as a Service (MaaS) (see also Chapter 3). A move across government to enlarge contract regions and include complementary modes to enhance system integration and the customer experience is starting to be seen. This opens up opportunities for the private sector to form joint ventures to compete for these larger contract offerings. Access contracts to rail and bus rapid transit hubs provide a way for operators looking at expanding their service offerings to cover the first/last mile to/from stations, whether this be in the form of fixed-route buses, flexible bus services (microtransit), carsharing or cycle hire (see also Chapters 17 and 22). The implications of this on existing public transport demand and contracts remain unclear. In Australia, though, Sydney’s Region 6 contract offering is a pioneering first of its kind to combine fixed-route buses and on-demand services (Perera et al., 2019). Under the next-generation economic deregulation, public transport contracts are shifting from their output-based form (delivering kilometres on defined modes) to outcome-based models which seek to deliver accessibility using any mode, maximising for network efficiency. There are opportunities to combine elements of competitive tendering and autonomous market initiative to create the next-generation service delivery model. In addition, intermediate mode regulation is becoming important with the growth in ridesourcing and microtransit provided by transportation network companies (and to a lesser extent cycle hire and carsharing) who have had to battle outdated regulation to become mainstream. Opportunities exist for a more streamlined approach based on a common platform and incentive payments to better integrate intermediate modes with other modes (e.g., public transport). The ownership model for autonomous vehicles, including buses, will determine its implications for productivity, traffic congestion, road capacity and the urban form. Regulations and incentives can help pool vehicles and move the community towards shared mobility. Pricing signals can help discourage autonomous zero occupancy deadheading – the influx of which will clog cities. Mobility as a Service contracts are also gaining great interest. This entails a personalised, one-stop travel management platform digitally unifying trip creation, purchase and delivery across all modes which can help move people away from vehicle ownership towards mobility consumed as a service. Mode-agnostic mobility contracts offered by brokers/aggregators (Wong et al., 2019) of the system to suppliers of transport assets/capacity can help deliver such a service. There will also be the opportunity to implement road pricing defined by time of day, geography and modal efficiency within this system to help optimise for network efficiency (Wong et al., 2020), including, for instance, preventing an influx of point-to-point transport. The COVID-19 pandemic has changed the public transport landscape, with significant reductions in patronage due to both mandated social distancing and government messaging that people should stay away from public transport as a shared mode until advised otherwise (see Beck & Hensher, 2020). We saw drops as low as 20 percent of pre-COVID-19 levels over the March to May 2020 period in many countries, with some slow increase as restrictions have been relaxed in a number of countries. So what does this mean for public transport regulation, especially the contracting regime? Where there is gross cost contracting, governments have carried the burden of lost revenue from the fare box, which has been a saviour, allowing many public transport services to continue at their post-COVID-19 levels, although how long this can continue if patronage does not return to acceptable levels is unknown. In some jurisdictions, for example, in the UK rail setting, governments have switched out of net cost franchise arrangements for a while to support rail services under a revised gross cost contract regime (see also Chapter 14). This added burden on the state, in some sense, aligns with the public interest 17

David A. Hensher

theories of regulation discussed in this chapter, which assume that sufficient information and appropriate enforcement powers exist to ensure that the public interest is enhanced by an essentially benevolent regulator, at least during a period of significant disruption and uncertainty.

Acknowledgements This paper contributes to the research program of the Volvo Research and Educational Foundations Bus Rapid Transit (BRT) Centre of Excellence. We acknowledge the Foundation for partial funding support.

Notes 1 If regulatory agencies come to be dominated by the industries or interests they are charged with regulating, the result is that an agency, charged with acting in the public interest, instead acts in ways that benefit the industry it is supposed to be regulating. 2 Bus travel time, fare, ticket type, frequency, arrival time at bus stop, walking time to bus stop, seat availability, information at bus stop, access to vehicle, bus stop facilities, temperature on bus, driver attitude and general cleanliness on board. 3 Since 2012, there have been a number of changes in operators, associated with direct selling of businesses but also to changes through competitive tendering, the latter under the Sydney and not Outer Sydney contracts.

References Aarhaug, J., Fearnley, N., Gregersen, F., & Norseng, R. (2018). 20 years of competitive tendering in the Norwegian bus industry – an analysis of bidders and winning bids. Research in Transportation Economics, 69, 97–105. Bajari, P.,  & Tadelis, S. (2001). Incentives versus transaction costs: A  theory of procurement contracts. RAND Journal of Economics, 32, 387–407. Bakker, B., & van de Velde, D. (2009). Superincentive public transport contracting in the Greater Amsterdam area. Paper presentation. Thredbo 11 Conference. Batarce, M., & Ávila, F. (2019). Misguided quality incentives: The case of the Santiago bus system (unpublished paper). Baumol, W. J., Panzar, J. C.,  & Willig, D. (1982). Contestable markets and the theory of industry structure. Harcourt Brace Jovanovich. Beck, M. J., & Hensher, D. A. (2020). Insights into the impact of COVID-19 on household travel, work, activities and shopping in Australia – the early days under restrictions. Transport Policy. doi:10.1016/j. tranpol.2020.07.001 Becker, S. (1983). A theory of competition among pressure groups for political influence. Quarterly Journal of Economics, XCVIII, 371–400. Demsetz, H. (1968). Information and efficiency: Another viewpoint. Journal of Law and Economics, 12, 1–22. Den Hertog, J. (2012). Economic theories of regulation. In R. J. van den Berg & A. M. Pacces (Eds.), Regulation and economics (pp. 25–96). Edward Elgar Publishing. Dye, R. A., & Sridhar, S. S. (2005). Moral hazard severity and contract design. RAND Journal of Economics, 36, 78–93. Gómez-Lobo, A., & Briones, J. (2014). Incentives in bus concession contracts: A review of several experiences in Latin America. Transport Reviews, 34, 246–265. Hart, O., & Holmstrom, B. (1987). Theory of contracts. In T. Bewley (Ed.), Advances in economic theory. Econometric Society Monographs, Fifth World Conference. Cambridge University Press. Hensher, D. A. (2010). Incompleteness and clarity in bus contracts: Identifying the nature of the ex ante and ex post perceptual divide. Research in Transportation Economics, 29, 106–117. Hensher, D. A. (2018). Public service contracts  – the economics of reform with special reference to the bus sector. In J. Cowie & S. Ison (Eds.), The Routledge handbook of transport economics (Chapter 7, pp. 93–107). Routledge.

18

Regulatory frameworks in public transport Hensher, D. A., Ho, C., & Knowles, L. (2016). Efficient contracting and incentive agreements between regulators and bus operators: The influence of risk preferences of contracting agents on contract choice. Transportation Research Part A, Policy and Practice, 87, 22–40. Hensher, D. A., & Prioni, P. (2002). A service quality index for area-wide contract performance assessment. Journal of Transport Economics and Policy, 36, 93–113. Hensher, D. A.,  & Stanley, J. K. (2008). Transacting under a performance-based contract: The role of negotiation and competitive tendering. Transportation Research Part A, Policy and Practice, Special Issue on Public Transport Reform (Thredbo 10), 42(10), 1295–1301. Hensher, D. A., & Stanley, J. K. (2010). Contracting regimes for bus services: What have we learnt after 20 years? Research in Transportation Economics, 29, 140–144. Hensher, D. A., & Wallis, I. (2005). Competitive tendering as a contracting mechanism for subsidising transportation: The bus experience. Journal of Transport Economics and Policy, 39(3), 295–321. Hensher, D. A., Wong, Y. Z., & Ho, L. (2020). Review of bus rapid transit and branded bus service network performance in Australia. Research in Transportation Economics. doi:10.1016/j.retrec.2020.100842 Holmstrom, B. (1979). Moral hazard and observability. Bell Journal of Economics, 10, 74–91. Hreljaa, R., Rye, T., & Mullen, C. (2018). Partnerships between operators and public transport authorities: Working practices in relational contracting and collaborative partnerships. Transportation Research Part A, Policy and Practice, 116, 327–338. Jansson, K., & Pyddoke, R. (2010). Quality incentives and quality outcomes in procured public transport – case study Stockholm. Research in Transportation Economics, 29, 11–18. Joskow, P. L., & Noll, C. (1981). Regulation in theory and practice: An overview. In G. Fromm (Ed.), Studies in public regulation (pp. 1–66). The MIT Press. Kay, J. A., & Vickers, J. S. (1990). Regulatory reform: An appraisal. In G. Majone (Ed.), Deregulation or re-regulation (pp. 223–251). Pinter Publishers. Merkert, R., & Preston, J. (2018). Workshop 2 report: Competitive tendering and other forms of contracting out: Institutional design and performance measurement. Research in Transportation Economics, 69, 86–96. Nash, C., & Bray, D. (2014). Workshop 5 report: The roles and responsibilities of government and operators. Research in Transportation Economics, 48, 286–289. Paul, A., & Gutierrez, G. (2005). Simple probability models for project contracting. European Journal of Operational Research, 165, 329–338. Perera, S., Ho, C., & Hensher, D. A. (2019, August). Resurgence of demand responsive transit services – insights from BRIDJ trials in inner West of Sydney, Australia. Paper presentation. 16th International Conference on Competition and Ownership of Land Passenger Transport (Thredbo 16). Posner, R. A. (1971). Taxation by regulation. Bell Journal of Economics and Management Science, 2, 22–50. Posner, R. A. (1974). Theories of economic regulation. Bell Journal of Economics and Management Science, 5, 335–358. Sappington, D. E. (2005). Regulating service quality: A survey. Journal of Regulatory Economics, 27, 123–154. Schwartz, A., & Watson, J. (2004). The law and economics of costly contracting. Journal of Law Economics and Organization, 19, 2–31. Stanley, J. K., & Hensher, D. A. (2008). Delivering trusting partnerships for route bus services: A Melbourne case study. Transportation Research Part A, Policy and Practice, 42(10), 1295–1301. Stigler, G. J. (1971). The theory of economic regulation. Bell Journal of Economics and Management Science, 2, 3–21. Tullock, G., Seldon, A., & Brady, G. L. (2002). Government failure. Cato Institute. van de Velde, D.,  & Alexandersson, G. (2020). Workshop 2 report: Practical considerations in implementing different institutional regimes. Research in Transportation Economics. doi:10.1016/j.retrec.2020.100920 Vigren, A., & Pyddoke, R. (2020). The impact on bus ridership of passenger incentive contracts in public transport. Transportation Research Part A, Policy and Practice, 135, 144–159. Viscusi, W. K., Vernon, J. M., & Harrington, J. E. (2005). Economics of regulation and antitrust. MIT Press. Walker, J. (2011). Human transit. Island Press. Wallis, I., Bray, D., & Webster, H. (2010). To competitively tender or negotiate: Weighing up the choices in a mature market. Research in Transportation Economics, 29(1), 89–98. Williamson, O. E. (2002). The theory of the firm as governance structure: From choice to contract. The Journal of Economic Perspectives, 16(3), 171–195. Wong, Y., & Hensher, D. A. (2018). The Thredbo story: A journey of competition and ownership in land passenger transport. Research in Transportation Economics, 69, 9–22.

19

David A. Hensher Wong, Y., Hensher, D. A., & Mulley, C. (2019, August). Delivering mobility as a service (MaaS) through a broker/aggregator business model. Paper presentation. 16th International Conference on Competition and Ownership of Land Passenger Transport (Thredbo 16). Wong, Y., Hensher, D. A., & Mulley, C. (2020). Emerging transport technologies and the modal efficiency framework: A case for mobility as a service (MaaS). Transportation Research Part A, Policy and Practice, 131, 5–19. Zerbe, R. O. (2001). Economic efficiency in law and economics. Edward Elgar Publishing.

20

2 PUBLIC TRANSPORT GOVERNANCE Fabio Hirschhorn and Wijnand Veeneman

Introduction Public transport is an essential service to both users and non-users. It provides an option for people driving on congested roads, for people who cannot – or prefer not to – drive a car, for people who need a means to access educational and leisure opportunities. It is thus also important for the overall population for ensuring that workers and providers of services of varied nature can reach their employment sites, for people who would prefer to see more street space available for other uses than parking, and for those – now and in the future – who can benefit from an alternative that is more sustainable than private cars. In sum, public transport is a vehicle for the delivery of a multitude of public values. Making public transport work in real life is challenging. Developing public transport systems that constitute an effective transportation alternative for people involves dealing with significant financial, technological, and infrastructure constraints (UN-Habitat, 2013). Nonetheless, it is the governance and policymaking of public transport that constitute the most complex challenge (Marsden & Reardon, 2017; Stough & Rietveld, 1997). It requires linking the diverse perspectives and interests that actors hold in relation to the way in which public transport should work and the outcomes it should deliver – that is, governance needs to ensure collective decisionmaking and coordinated action for public transport to prioritise and deliver public values. The magnitude of the governance challenge is illustrated by the crisis involving the global outbreak of COVID-19, which put public transport at a crossroads. Trains, metro, and buses are identified as ‘unsafe’ because of potentially hazardous physical proximity during travel. At the same time, the fundamental role of public transport to users and non-users became more prominent than ever, as it can ensure that essential workers reach their job locations, keeping indispensable services to cities and society running. This ambiguity triggers a number of complex questions ranging from the need to revise subsidy and remuneration of operators, practices in relation to vehicle occupancy and cleaning procedures, the redefinition of routes and frequency of services, to strategies for the planning of future bidding and concessions. Permeating these questions is the intricate decision-making and coordination around a multiplicity of values, such as individual safety, public health, economic development, and sustainability (social, environmental, and economic), all of which are expected from public transport and that governance is supposed to enable. 21

Fabio Hirschhorn and Wijnand Veeneman

How to tackle and understand the governance challenge? Traditionally, literature addresses (public transport) governance through the study of governance shifts. These are changes in mechanisms of governance, location of governance, governing capacities, and styles of governance (Kersbergen & Waarden, 2004). These shifts come about to address societal problems that are seen as most pressing, namely growing deficits, pollution, and pandemics, and/or due to emerging influential theories, that is, neoliberal ideals and networked governance, for example. Indeed, public transport governance studies historically look at how the introduction or reform of certain formal governance frameworks can help or hinder the achievement of diverse public values, including sustainability, accessibility, and safety (Hirschhorn et al., 2019). Nevertheless, this approach is limited in its ability to reveal complexities involved in the design and implementation of public transport governance, thus being insufficient to equip decision-makers in their task of coordinating collective decision-making for the delivery of public values. At least two important considerations explain why this is the case. First, most existing analyses tend to approach governance shifts from a fragmented perspective, setting apart a specific element of governance to estimate their isolated impact on performance, disregarding interdependencies between values. However, public values may not be aligned with each other, and preferences may change over time, generating potentially tough trade-offs. Promoting one value influences the ability of the system to deliver others, and managing trade-offs between values in political decision-making requires recognising these interdependencies. Second, governance shifts should not be seen as changes in formal frameworks exclusively and taking place in a vacuum. Other dimensions of governance processes – such as the role of informal institutions, political steering, actors’ agency, power relations, and framing in political decision-making – and the way they interact with formal frameworks are often neglected. However, they are also crucial explanatory factors of how governance shifts occur and the results they produce, including the public values that are promoted and safeguarded. This chapter tackles the public transport governance challenge by answering four questions. First, it clarifies what is meant by public transport governance. Subsequently, the chapter addresses the question of why public transport governance is complex and reviews academic works explaining how public transport governance shifts deliver public values. Based on this stock-taking exercise, the chapter concludes by looking ahead and answering the where question – that is, it proposes directions for policy and academia.

What is public transport governance? To understand public transport governance better, it is first important to clarify what it means. This is the objective of this section. The term public transport (hereafter PT) is used in this chapter to designate all collective modes of land passenger transport services available to the general public on a non-discriminatory and continuous basis. In most cases, PT works according to predetermined routes, timetables, or frequencies, and access to it depends on the payment of fares. The word ‘public’ in the term is thus associated with the non-discretionary possibility of access to these services and not with the ownership nature of the transport operator – that is, PT services can be offered by either public or private firms. Finally, the chapter considers PT services offered within the urban context, linking the cities to the surrounding hinterland and providing links between cities. From a systemic perspective, PT can be conceptualised as a complex sociotechnical system in that it comprises both complex physical–technical systems and networks of interdependent 22

Public transport governance

actors (Bauer & Herder, 2009). PT is made up of interacting and interdependent elements such as infrastructure, technology, finance, and actors (individual or collective) and is supposed to fulfil a broad societal function – passenger transportation. For PT to work well, the systems that allow the planning and provision of services are necessary, as well as the systems that make the decisions on said planning and provision of services (White, 2017) (see also Chapter 21). Each actor involved in PT – including politicians and public officials, operators’ managers, drivers, users, and non-users – has a different perception of reality; actors have different opinions and preferences on the way PT systems and its subsystems should work and the functions they should fulfil. Because of these differing interests and perceptions, cooperation and coordination between actors cannot be taken for granted. This is the space for governance. Broadly speaking, governance, and governance theories, are concerned with creating and examining the ways in which societies create and uphold norms and instruments to deal with matters that require coordination in the pursuit of collective interests (Bevir, 2013). It involves the formal and informal structures and processes shaping the interactions between public and private actors in collective decision-making and through which they coordinate practices in view of predefined goals (Hufty, 2011). Analytically, governance can manifest itself in three main facets, politics, polity, and policy (Treib et  al., 2007). First, governance has a political dimension, which concerns player constellations, power, and conflicts in the political process. Second, the polity dimension of governance concerns the structures and rules that influence the players (but can also be shaped by them), that is, the institutional environment. Finally, the policy dimension refers to the instruments and content of policies designed and implemented in the sector. Therefore, the governance of PT concerns the coordination of the decision-making processes to identify and promote the collective goals to be achieved in connection to PT as a policy area. Accordingly, PT governance establishes the allocation, amongst diverse (public and private) actors, of the roles and responsibilities – along with the needed resources and discretionary ability (agency) – for the design and implementation of PT policies and services.

Why is public transport governance complex? The previous section clarified that, broadly speaking, (PT) governance is about the coordination of collective decision-making to achieve predetermined public values. It involves the identification of goals, the development of the policies necessary to reach those goals, and the allocation of powers and responsibilities across actors as well as the attachment of resources to those policies. But why is PT governance complex? This section addresses this question.

Public values The collective goals that governance is expected to promote and secure reflect the public values held by the multiplicity of actors that have a stake in the functioning of PT. Public values thus represent the principles defining government’s responsibilities and rights, along with the obligations of citizens, regarding different policy areas; they constitute the purposes of stakeholders and their networks in relation to that sector (Jørgensen & Bozeman, 2007; Koppenjan et al., 2008). The precise formulation of a public value varies according to stakeholders’ perspective and interests, existing problems, and policy arenas (de Bruijn  & Dicke, 2006). Since every actor involved in PT governance processes has a limited perspective and set of priorities in relation to what PT is supposed to realise, the definitions of public values change in terms of time and 23

Fabio Hirschhorn and Wijnand Veeneman

context. Moreover, as further detailed in the following, achieving goals such as environmental sustainability may affect the ability of PT to deliver other goals, such as greater efficiency (see also Chapter 8). This means that public values may not be aligned with each other, and the conflict between public values entails tough trade-offs for decision-makers (Stewart, 2006; Veeneman et al., 2009). These dynamic interdependencies are the main reasons behind the challenge for PT governance.

The challenge for public transport governance Sociotechnical systems such as PT can be thought of as “vehicles through which important public values are delivered” (Steenhuisen et al., 2009, p. 491). As such, PT is not immune to the complexities alluded to in the previous section. Values attached to PT are multiple and varied in nature, including greater environmental sustainability, greater accessibility to work or leisure, efficiency in the production of the services to reduce the need for public subsidies, and time savings. Not only is the set of values that need to be taken into account in PT decision-making multifaceted, but by definition, these values drive two conflicting types of objectives: some values are achieved by increasing the number of passengers in PT, and other values are achieved by increasing the spatial availability of PT (Walker, 2008). Promoting each of these objectives means negatively affecting the other. In periods of economic hardship, for example, authorities may prioritise the value of efficiency above accessibility, reducing PT spatial coverage and/or service frequency to decrease the need for operational subsidies. This decision impacts the constituencies that will have less access to PT for the benefit of the public purse. On the other hand, if in a different moment, there is a push for greater accessibility to PT in suburban areas, routes will have to become longer and the number of bus stops will be increased, affecting operational costs and possibly influencing efficiency. Real examples of these interdependencies between values attached to PT are numerous. The opening of the market for rail transport in the United Kingdom brought changes in the efficiency of the operators but also unintended consequences in relation to safety and pricing (Department of Transport, 1993) (see also Chapter 14). Furthermore, the deregulation in bus services (outside London) has also been linked to lower quality of services and declining ridership levels (see also Chapters 1 and 13). This has led government to introduce a series of legislative measures shifting the emphasis over time, from a focus on competition as a major policy aim to one in which partnerships between operators and local transport authorities are encouraged as a means to support for service coordination (White, 2018). The introduction of tendering in Santiago triggered a host of changes in the behaviour of operators that asked for constant rethinking of the best way in which the contracts should be set up; a strong initial push for competitive tendering is now followed by a reconsideration due to perceived downsides (Galilea & Batarce, 2016). Likewise, different transport modes offer different values on different spatial scales, and, as such, they are valued differently by municipal, metropolitan, and national governments that might stimulate different investment decisions, as shown in case studies from Australia, the Netherlands, and South Africa (Veeneman & Mulley, 2018). Finally, and further complicating this scenario, there is also evidence that even goals linked to ridership objectives may not always be aligned with each other. In Madrid, in the 1990s and 2000s, for example, PT modal split and cost-recovery levels moved in opposite directions after the introduction of regulatory reforms (Vassallo et al., 2009). This is where governance plays an important role, linking and coordinating the multiple perspectives on public values. The corporate governance of the railway company makes sure 24

Public transport governance

that a decision of a planner to extend the services is financially feasible. It links the perspective of the planner to the perspective of the finance department. Market governance lets the price of the service convey the costs to the person deciding to use the system. It links the perspective of the customer to that of an entire supply chain. Safety governance requires train drivers to keep safety of the traveller a top priority. It links the perspective of the driver to that of passengers and their families. As such, governance can be seen as the integrating factor, widening the value and system perspective of the decision-maker, linking them to other decision-makers in fragmented multiactor systems. Research can identify, help understand, propose, and explain possible safeguarding mechanisms and management strategies to deal with public values throughout the entire governance process, starting from the identification of goals and moving through the development of the policies, the attachment of resources to those policies, and the analysis of the results achieved. In other words, analysing and understanding PT governance is critical for decision-makers to manage and balance values’ trade-offs. The next section examines the way in which most literature has worked to equip decision-makers by revealing the link between governance and the achievement of public values in PT.

How does public transport governance deliver public values? Having clarified the meaning and the challenging task of PT governance in the previous sections, the chapter moves to addressing the subsequent question of how PT governance can eventually deliver public values.

Governance shifts Kersbergen and Waarden (2004) explain that there is a tendency across disciplines to study governance by focusing on shifts; that is, literature across different fields, the authors claim, describes and analyses changes taking place in the forms and mechanisms of governance, the location of governance, governing capacities, and styles of governance. The multilevel governance framework, for example, describes a shift in the location of governance. It acknowledges the multiactor dispersed policymaking performed within and across politico-administrative institutions located at different territorial levels (Stephenson, 2013). The new institutional economics identify changes in governance mechanisms whereby actors coordinate actions via hierarchy, market, and hybrid structures, depending on the transaction costs each form entails (Williamson, 2010). These shifts in governance can originate from changes in the type of problems with which societies are confronted and from influential ideas and theories. Literature on networked forms of governance, for example, suggests that the nature of unruly policy problems, such as growing congestion and pollution and global warming, require elected officials and public managers to develop new governance skills and tools. They are expected to develop trusting and collaborative relationships with a host of governmental and non-governmental actors to complement or substitute governance through hierarchies and markets (Sørensen & Torfing, 2009). Concerning shifts originated by influential ideas and theories, Bevir and Rhodes (2016) indicate that the shift for a greater emphasis on networks instead of markets and hierarchy can be traced back to the rise of neoliberal ideals in the 1970s and 1980s, giving primacy to values like efficiency and effectiveness, and the associated reforms under the New Public Management (NPM) label. 25

Fabio Hirschhorn and Wijnand Veeneman

Examining governance shifts in public transport In the field of PT in particular, governance studies have also historically worked by examining shifts. In the 19th century, Chadwick (1859), interested in the economic efficiency and fare levels in the English railway sector, compared the potential effects of shifts between competition for the field – namely competition to have access to a market or an area to deliver PT services – and competition within the field – related to the competition between different transport providers operating in the same market. Since the 1980s, governance shifts, as well as their analysis, have proliferated in PT. This has been the case both in response to particular problems which need to be tackled – undesirable continued growth of public subsidies and declining passenger numbers – and in the wake of influential ideals and theories – the neoliberal ideals operationalised in NPM reforms, as well as the theory of contestable markets (Baumol, 1982) and competitive tendering (franchise bidding) as a regulatory mechanism for monopoly operations (Demsetz, 1968). As a result, efficiency became the key public value of PT driving governance shifts. The shifts in PT governance in the last decades primarily involved experimentation with mixes of deregulation (reducing the number of rules to which transport operators are subject in the market in which they operate, resulting in greater freedom to define service characteristics), liberalisation (allowing other operators, in addition to the incumbent, to access the market), and/or privatisation (transference of the ownership of a company or agency from the public sector – such as the national or local government – to the private sector) (van de Velde & Hirschhorn, 2021). In the European context, for example, these shifts first became prominent in the United Kingdom, where the long-distance coach sector was deregulated in 1980 and local and regional bus sector outside London too, in 1986. The British government also introduced competitive tendering in the railway sector (known as railway “franchising”) in 1994. Other reform experiences with competitive tendering increased within the bus sector  – in London (1984), Copenhagen (1991), Sweden (1989), France (reform of contracting in 1981 and stricter tendering rules in 1994), and the Netherlands (2001). The literature in the field of PT governance has since been built on a body of work to understand if and how governance shifts involving these mixes can promote or hinder diverse performance goals in PT (Hirschhorn et  al., 2019). In these analyses, authors describe and compare the different ways in which tasks and responsibilities are allocated amongst actors in the PT sector. Their aim is to investigate how such allocation (or changes thereof) may translate into variations in levels of performance indicators such as operational costs, ridership, modal split, or user satisfaction. From an analytic perspective, this academic work can be mapped and grouped according to the type of task and responsibility being shifted across the three levels of PT management and control  – strategic, tactical, and operational (van de Velde, 1999). The ‘strategic level’ refers to deciding on public transport ‘aims’ such as policy goals in terms of accessibility and modal share. The ‘tactical level’ refers to service design (routes, frequencies, fares, vehicle design, etc.), that is, determination of ‘means’. The ‘operational level’ refers to operational management, for example, crew and vehicle rostering or facility and vehicle maintenance. A first crucial dichotomy in these studies refers to the distinction between PT systems based on competitive tendering of monopoly rights by a transport authority and those systems based on free market competition (van de Velde, 1999). In authority-initiated regimes, governments have the legal monopoly of initiative in the sense that autonomous market entry is legally impossible and all production or market entry is the result of choice of the authority to produce or request the production of services. In turn, in market-initiated regimes, the supply 26

Public transport governance

of transport services is based upon autonomous market entry with more or fewer regulatory checks at the entrance. Most studies concerning this strategic-level characteristic of PT are concerned with identifying which of these systems is most efficient or how ridership levels are affected (see Cowie, 2012; Sakai &Takahashi, 2013). Concerning analyses of governance shifts in elements of the tactical level of PT, issues such as the use of awarding competitive mechanisms or the contractual allocation of risks and responsibilities frequently appear as the object of studies. Authors investigate the cost savings that are linked to the use of competitive tender and negotiated contracts and differences connected to the use of gross-cost and net-cost arrangements (see Filippini et al., 2015; Zhang et al., 2015). Another tactical governance shift frequently examined concerns changes in fare policies, especially the introduction of integrated fare systems, and how this can help improve ridership levels (see Abrate et al., 2009; Sharaby & Shiftan, 2012). Finally, concerning governance shifts in the operational level, studies mostly focus on variations in the ownership nature of transport operators. Authors seek insight into whether publicly owned or private operators perform better in terms of values such as technical efficiency or customer satisfaction levels (see Roy & Yvrande-Billon, 2007; Swarts & Warner, 2014).

Where to? Ways forward in the practice and study of public transport governance This final section critically reflects on the previous sections. The next subsection suggests guiding principles that can allow decision-makers to better understand, design, and conduct PT governance. This is followed by an opportunity for academics to support decision-makers in their task.

The need to recognise complex interdependencies between values There is a gap between the current mainstream academic work relating to PT governance and the actual functioning of governance in reality. Existing analysis tends to approach governance shifts from a fragmented perspective, setting apart a specific element of governance to estimate their isolated impact on a single type of performance. Discussing who initiates service delivery, private or public actors, for example, is just a single feature in the much wider governance and decision-making context; furthermore, as described previously, it is one that is mostly analysed from a perspective of one value: efficiency. However, when looking at governance, a holistic approach appears more relevant. A governance system needs to be set up and function to deal with a wide range of values that, as highlighted in a previous section, may change over time and may not be aligned with each other. Due to the dynamic conflict between public values, decisions targeting one value will most likely influence the ability of the system to deliver others. Therefore, designing governance is an exercise in dealing with complexity. Rather than avoiding it, decision-makers should embrace it. Governance design and functioning are incomplete if not recognising these interdependencies to account for and manage such intricacies. The broader literature on political science and public administration offers insights that help in elaborating upon some guiding principles to support in this task. PT governance can follow a whole-of-values, adaptive, and context-aware approach. A whole-of-values approach refers to the importance of recognising and taking into account ex ante the entirety of public values potentially affected by a governance shift being planned, as well as the interdependencies between these values (de Bruijn et al., 2004). Decision-makers 27

Fabio Hirschhorn and Wijnand Veeneman

should endeavour to address head on the interdependencies and conflicts between diverse goals, making them explicit from the policy design and planning stage. The aim is to ensure a systematic and integrated vision of PT and related policy areas rather than the usual fragmented ‘system of parts’ approach. One mechanism to enable this effort for completeness can, for example, be the promotion of further communication across government departments working separately to develop policies on seemingly separate issues that, in fact, are interconnected. Likewise, having interdisciplinary teams that include individuals with backgrounds other than transport can also be valuable (Hirschhorn et al., 2020). However, whilst advancing the need for a whole-of-values vision, it is necessary to accept that cognitive and resource limitations are inevitable and affect governance and policymaking. It is not feasible to expect that policymakers can have full ex ante insight into the totality of values affected by a governance design decision, and perhaps ‘muddling through’ is not only the single feasible approach but also the most effective one (Lindblom, 1959). This is the reason an adaptive approach to governance, the second principle, is important. As delineated here, an adaptive approach complements the whole-of-values vision by proposing that decision-making should strive to find a middle ground between an all-encompassing approach – that disregards cognitive and resource limitations – and a fragmented and incremental view – that disregards the complex interdependencies between values. Adaptability thus entails being able to adjust not only to changing circumstances (the dynamic preferences and conflicts between values) but also to cognitive and resource limitations, allowing decision-makers to approach governance design and policymaking in a structured and focused manner, that is, developing a systematic ex ante analysis focused on selecting most relevant issues at stake, filtering and examining only strategic ones within the particular context. In other words, unable to review all the existing values and interdependencies at play, and seeking to do better than merely thinking one or two steps ahead, decision-makers can use their cognitive and time resources in first deciding amongst fundamental factors to be analysed – a higher-level scanning – and then examining in detail only the options within the chosen approach (Etzioni, 1986). An adaptive approach can combine elements of complex long-term planning with incrementalism, always having a set of strategically defined societal goals in the backdrop. This can be operationalised in a number of measures, such as: (i) strategic problem structuring definition; (ii) recourse to the opinion and input from a broad set of stakeholders – from different government departments as well as from outside government; (iii) experimentation in pilots and living labs of different scales; (iv) the combined use of diverse meta-governance strategies, mixing more or less hands-on and direct or indirect governance strategies; (v) the use of flexible policy instruments that can be tweaked or replaced based on empirical knowledge gained over time; and (vi) mitigation instruments that can coexist with formal frameworks and compensate for their partial unintended effects without requiring completely overhauling them (Etzioni, 1986; Kemp et al., 2007; Sørensen & Torfing, 2017). The third and final principle, rich context awareness, is a necessary condition of the previous two. Context awareness recognises that policy planning and decision-making processes always take place in a specific context, and context matters, because it turns (or fails to turn) causal potential into causal outcome (Pawson & Tilley, 1997). PT governance involves designing and implementing rules and processes that do not land on a clean slate; path dependencies, interests of incumbent actors, existing logics of action, and shared understandings play a role in determining the solutions proposed for and the eventual functioning of PT systems (Hirschhorn, 2020). Likewise, broader and much more complex socioeconomic trends or shocks in which the governing policy regime operates – including demographic events, pandemics, economic cycles – make a difference as well (as dramatically shown by the case of COVID-19). This means that 28

Public transport governance

single policies or specific formal institutional frameworks transferred between countries or cities, or reattempted in a different time, will not necessarily work the same way. Decision-makers must be sensitive to context when designing and changing governance rather than blindly following existing templates. Only this way can they explicitly and ex ante take stock of the most strategic values at stake, as well as their potential conflicts, in an adaptable manner, looking for the best design to promote the desired values and manage trade-offs.

The need for a more comprehensive view on governance Academia can have a pivotal role in supporting decision-makers in recognising the complex interdependencies between values to implement a whole-of-values, adaptive, and contextaware approach to the governance of PT. To this end, scholars have to address some gaps in the way this topic has been dealt with so far. This research opportunity is further detailed in this subsection. Indeed, there is at least one important gap in current mainstream academic work in PT. This lacuna refers to the somewhat limited way in which the concept of governance has been understood and operationalised in most analyses in the field. When looking at governance shifts, studies focus primarily on how organisational elements of PT systems may influence performance outcomes. Whilst valuable, these efforts employ an overly narrow view of governance, restricting analyses to the effects brought by governance shifts in formal institutions – the set of formal rules that are created, communicated, and enforced by official channels, for example, constitutions, laws, and contracts (Farrell & Héritier, 2003; Helmke & Levitsky, 2004). In other words, analysis emphasises one portion of the polity dimension of governance only, neglecting issues such as the role of informal institutions, political steering, actors’ agency, power relations, and framing in political decision-making. Nevertheless, whilst analytically distinct, all dimensions of governance are empirically intertwined, and they do not work or produce effects separately. The interactions between these dimensions are decisive in the way collective decision-making processes develop, being crucial for the outcomes that are achieved. The interplay between elements across politics, polity, and policy dimensions vary according to the social and economic background in which they occur and thus depend on how actors (individual or collective) ‘play the game’. Institutions constrain and enable actors by facilitating or hampering certain actions and outcomes but can also be shaped according to how these actors interpret and enact them (Mahoney & Thelen, 2010). Evidence shows that informal institutions may, alongside formal ones, enable better PT planning and integration with land use policies, and key leaders can have a pivotal role in championing and enacting PT solutions and governance shifts (Hirschhorn et al., 2020; Hrelja et al., 2017). One crucial consequence of this multidimensional and contingent character of PT governance is that there are often important discrepancies between the imagined institutional design and their actual implementation and functioning. Recognising these discrepancies and, more importantly, understanding why and how they emerge is a challenge. However, the prevailing approach to the understanding and design of PT governance, described in this chapter, is unable to fully support decision-makers in grappling with these more complex questions of governance. The study of shifts in the forms, locations, and capacities of governance should be complemented with analyses that consider issues such as the role of informal institutions, political steering, actors’ agency, power relations, and framing in political decision-making. Whereas in other disciplines, especially in the social sciences, these topics have long been in the agenda (see Allison, 1971; Emirbayer  & Mische, 1998; Sabatier, 1991), only recently have they become 29

Fabio Hirschhorn and Wijnand Veeneman

more salient in PT (see Isaksson  & Heikkinen, 2018; Reardon  & Marsden, 2020). Therefore, there are clear opportunities for extending (not replacing) current research approaches by engaging with and benefiting from the insights produced by other disciplines in the social sciences. Understanding the dynamic interplay between specific actor constellations (with particular value preferences), policy instruments, steering strategies, and institutional environment (formal and informal rules) is key to improve governance design and understand what the possible effects of governance shifts can be.

References Abrate, G., Piacenza, M., & Vannoni, D. (2009). The impact of integrated tariff systems on public transport demand: Evidence from Italy. Regional Science and Urban Economics, 39(2), 120–127. Allison, G. T. (1971). Essence of decision: Explaining the Cuban Missile Crisis. Little, Brown and Company. Bauer, J., & Herder, P. (2009). Designing socio-technical systems. In A. Meijers (Ed.), Philosophy of technology and engineering sciences (pp. 601–630). North Holland. Baumol, W. J. (1982). Contestable markets: An uprising in the theory of industry structure. The American Economic Review, 72(1), 1–15. Bevir, M. (2013). A theory of governance. University of California Press. Bevir, M., & Rhodes, R. A. W. (2016). The 3Rs’ in rethinking governance: Ruling, rationalities, and resistance. In M. Bevir & R. A. W. Rhodes (Eds.), Rethinking governance: Ruling, rationalities and resistance. Routledge. Chadwick, E. (1859). Results of different principles of legislation and administration in Europe; of competition for the field, as compared with competition within the field, of service. Journal of the Statistical Society of London, 22(3), 381–420. Cowie, J. (2012). Contestability in bus markets – evidence from the British de-regulated market. Applied Economics, 44(36), 4777–4785. de Bruijn, H., & Dicke, W. (2006). Strategies for safeguarding public values in liberalized utility sectors. Public Administration, 84(3), 717–735. de Bruijn, H., van der Voort, H., Dicke, W., de Jong, M., & Veeneman, W. (2004). Creating system innovation how large scale transitions emerge. CRC Press. Demsetz, H. (1968). Why regulate utilities? Journal of Law and Economics, 11(1), 55–65. Department of Transport. (1993). Ensuring safety on Britain’s railways. Retrieved May 3, 2020, from www. railwaysarchive.co.uk/documents/HSE_Ensuring1993.pdf Emirbayer, M., & Mische, A. (1998). What is agency? American Journal of Sociology, 103(4), 962–1023. Etzioni, A. (1986). Mixed scanning revisited. Public Administration Review, 46(1), 8–14. Farrell, H., & Héritier, A. (2003). Formal and informal institutions under codecision: Continuous constitutionbuilding in Europe. Governance, 16(4), 577–600. Filippini, M., Koller, M., & Masiero, G. (2015). Competitive tendering versus performance-based negotiation in Swiss public transport. Transportation Research Part A: Policy and Practice, 82, 158–168. Galilea, P., & Batarce, M. (2016). Designing bus concession contracts. In J. C. Muñoz & L. Paget-Seekins (Eds.), Restructuring public transport through bus rapid transit: An international and interdisciplinary perspective (pp. 127–143). Bristol University Press. Helmke, G., & Levitsky, S. (2004). Informal institutions and comparative politics. Perspectives on Politics, 2(4), 725–740. Hirschhorn, F. (2020). En route to better performance: Tackling the complexities of public transport governance [Doctoral thesis, Delft University of Technology]. doi:10.4233/uuid:431ffda5-275c-4670-bf32-43d863261ec2 Hirschhorn, F., van de Velde, D., Veeneman, W., & ten Heuvelhof, E. (2020). The governance of attractive public transport: Informal institutions, institutional entrepreneurs, and problem-solving know-how in Oslo and Amsterdam. Research in Transportation Economics, 83, 100829. Hirschhorn, F., Veeneman, W., & van de Velde, D. (2019). Organisation and performance of public transport: A systematic cross-case comparison of metropolitan areas in Europe, Australia, and Canada. Transportation Research Part A: Policy and Practice, 124, 419–432. Hrelja, R., Monios, J., Rye, T., Isaksson, K., & Scholten, C. (2017). The interplay of formal and informal institutions between local and regional authorities when creating well-functioning public transport systems. International Journal of Sustainable Transportation, 11(8), 611–622.

30

Public transport governance Hufty, M. (2011). Investigating policy processes: The governance analytical framework (GAF). In U. Wiesmann & H. Hurni (Eds.), Research for sustainable development: Foundations, experiences, and perspectives (pp. 403–424). Geographica Bernensia. Isaksson, K., & Heikkinen, S. (2018). Sustainability transitions at the frontline: Lock-in and potential for change in the local planning arena. Sustainability, 10(3), 840. Jørgensen, T. B., & Bozeman, B. (2007). Public values an inventory. Administration & Society, 39(3), 354–381. Kemp, R., Loorbach, D., & Rotmans, J. (2007). Transition management as a model for managing processes of co-evolution towards sustainable development. International Journal of Sustainable Development & World Ecology, 14(1), 78–91. Koppenjan, J., Charles, M. B.,  & Ryan, N. F. (2008). Editorial: Managing competing public values in public infrastructure projects. Public Money & Management, 28(3), 131–134. Lindblom, C. E. (1959). The science of “muddling through”. Public Administration Review, 19(2), 79–88. Mahoney, J., & Thelen, K. (2010). A theory of gradual institutional change. In J. Mahoney & K. Thelen (Eds.), Explaining institutional change: Ambiguity, agency, and power (pp. 1–37). Cambridge University Press. Marsden, G., & Reardon, L. (2017). Questions of governance: Rethinking the study of transportation policy. Transportation Research Part A: Policy and Practice, 101, 238–251. Pawson, R., & Tilley, N. (1997). Realistic evaluation. Sage Publications. Reardon, L., & Marsden, G. (2020). Exploring the role of the state in the depoliticisation of UK transport policy. Policy & Politics. doi:10.1332/030557319X15707904263616 Roy, W., & Yvrande-Billon, A. (2007). Ownership, contractual practices and technical efficiency: The case of urban public transport in France. Journal of Transport Economics and Policy, 41(2), 257–282. Sabatier, P. A. (1991). Toward better theories of the policy process. PS: Political Science and Politics, 24(2), 147–156. Sakai, H., & Takahashi, Y. (2013). Ten years after bus deregulation in Japan: An analysis of institutional changes and cost efficiency. Research in Transportation Economics, 39(1), 215–225. Sharaby, N., & Shiftan, Y. (2012). The impact of fare integration on travel behavior and transit ridership. Transport Policy, 21, 63–70. Sørensen, E., & Torfing, J. (2009). Making governance networks effective and democratic through metagovernance. Public Administration, 87(2), 234–258. Sørensen, E., & Torfing, J. (2017). Metagoverning collaborative innovation in governance networks. American Review of Public Administration, 47(7), 826–839. Steenhuisen, B., Dicke, W., & de Bruijn, H. (2009). “Soft” public values in jeopardy: Reflecting on the institutionally fragmented situation in utility sectors. International Journal of Public Administration, 32(6), 491–507. Stephenson, P. (2013). Twenty years of multi-level governance: “Where does it come from? What is it? Where is it going?” Journal of European Public Policy, 20(6), 37–41. Stewart, J. (2006). Value conflict and policy change. Review of Policy Research, 23(1), 183–195. Stough, R., & Rietveld, P. (1997). Institutional issues in transport systems. Journal of Transport Geography, 5(3), 207–214. Swarts, D., & Warner, M. E. (2014). Hybrid firms and transit delivery: The case of Berlin. Annals of Public and Cooperative Economics, 85(1), 127–146. Treib, O., Bähr, H., & Falkner, G. (2007). Modes of governance: Towards a conceptual clarification. Journal of European Public Policy, 14(1), 1–20. UN-Habitat. (2013). Planning and design for sustainable urban mobility: Global report on human settlements. UN-Habitat. van de Velde, D. (1999). Organisational forms and entrepreneurship in public transport. Part 1: Classifying organisational forms. Transport Policy, 6(3), 147–157. van de Velde, D., & Hirschhorn, F. (2021). Regulatory reforms and competition in public transport. In Encyclopaedia of transportation. Elsevier. van Kersbergen, K.,  & van Waarden, F. (2004). “Governance” as a bridge between disciplines: Crossdisciplinary inspiration regarding shifts in governance and problems of governability, accountability and legitimacy. European Journal of Political Research, 43(2), 143–171. Vassallo, J. M., Pérez De Villar, P., Muñoz-Raskin, R., & Serebrisky, T. (2009). Public transport funding policy in Madrid: Is there room for improvement? Transport Reviews, 29(2), 261–278. Veeneman, W., Dicke, W., & de Bruijne, M. (2009). From clouds to hailstorms: A policy and administrative science perspective on safeguarding public values in networked infrastructures. International Journal of Public Policy, 4(5), 414–434.

31

Fabio Hirschhorn and Wijnand Veeneman Veeneman, W., & Mulley, C. (2018). Multi-level governance in public transport: Governmental layering and its influence on public transport service solutions. Research in Transportation Economics, 69, 430–437. Walker, J. (2008). Purpose-driven public transport: Creating a clear conversation about public transport goals. Journal of Transport Geography, 16, 436–442. White, P. (2017). Public transport: Its planning, management and operation (6th ed). Routledge. White, P. (2018). Prospects in Britain in the light of the bus services act 2017. Research in Transportation Economics, 69, 337–343. Williamson, O. E. (2010, June). Transaction cost economics: The natural progression. American Economic Review, 100, 673–690. Zhang, C., Juan, Z., & Xiao, G. (2015). Do contractual practices affect technical efficiency? Evidence from public transport operators in China. Transportation Research Part E: Logistics and Transportation Review, 80, 39–55.

32

3 MOBILITY AS A SERVICE AND PUBLIC TRANSPORT Göran Smith

Introduction Mobility as a Service (MaaS) is a digital concept that centres on integrating traditional public transport with other types of mobility services such as carsharing and ridesourcing. In recent years, MaaS has become a much-discussed topic within the transport industry. Proponents contend that MaaS can make it easier for users to combine multiple mobility services into customised offerings that match their individual mobility needs. Accordingly, they argue that MaaS can decrease the perceived need of owning private cars (for some) and thereby reduce the lockin effect of car ownership and support a modal shift away from private cars to active mobility, public transport, and other mobility services that build on shared vehicles and/or rides. Such a modal shift could potentially address some of transport systems’ pressing sustainability problems, like congestion, car parking, and carbon dioxide emissions. In the wake of the COVID-19 pandemic, MaaS has, moreover, been brought forward as a concept that can improve transport system resilience. MaaS is thought to strengthen the ties between mobility service providers and help users shift between different modes of transport, thus making it easier for both groups to quickly adapt to changed circumstances. However, the empirical evidence backing up these propositions is sparse. The understanding of when and under what conditions MaaS can attract what customer segments is still limited, as is the knowledge on how MaaS adoption influences travel behaviour. Despite the intuitive nature of MaaS and the widespread interest in the concept, the systematically evaluated demonstrations of comprehensive MaaS services can thus far be counted on the fingers of one hand. The realisation of MaaS has proven to be a demanding endeavour. MaaS does not fit well with existing legal frameworks, for instance, regarding customer relations in multimodal travel chains. Furthermore, MaaS challenges mobility services providers’ current models for interorganisational collaboration and business. Hence, MaaS developments have been accompanied by scepticism and mistrust. In particular, public transport authorities have been criticised for not permitting external actors to resell their products. Since public transport is considered the backbone of MaaS, this virtually blocks those other than the public transport authorities themselves from developing viable MaaS offerings. Nonetheless, these hurdles seem to have been mitigated, at least temporarily, in some places. Several MaaS services that integrate public transport with other mobility service offerings have 33

Göran Smith

been launched recently, such as Jelbi in Berlin (jelbi.de), Travis in Stockholm (gettravisapp.se), and Zipster in Singapore (mobility-x.com). A number of ongoing MaaS trials are, moreover, being closely monitored by researchers, for example, Tripi in Sydney, Amaze in Amsterdam, and EC2B in Gothenburg. Hopefully, these projects (and others) can start to accumulate an empirical body of knowledge regarding whether, and, if so, how, MaaS disrupts public transport, public transport authorities, and transport systems in general. Notwithstanding the shortage of data on the effects of MaaS, this chapter takes stock of extant MaaS experiences to discuss how public transport authorities can shape the development, diffusion, and use of MaaS. Using Smith (2020) as a starting point, this chapter first provides a definition of MaaS, followed by a brief review of expectations on how MaaS will influence public transport. A  framework that describes pathways to governing MaaS developments is introduced next and is subsequently used to analyse and discuss how public transport authorities in Finland, Sweden, and Norway have approached MaaS. Finally, the chapter concludes with a few statements on lessons learnt thus far.

What is Mobility as a Service? MaaS is an ambiguous term with multiple meanings. Since it was first introduced in Finland in 2014, the rapidly growing scientific literature on MaaS has provided a range of overlapping typologies (Kamargianni et al., 2016; Lyons et al., 2020; Sochor et al., 2018). Still, the mostcited definitions stem from grey material such as master thesis projects, industry magazines, and research reports: • • •

“A system in which a comprehensive range of mobility services are provided to customers by mobility operators” (Heikkilä, 2014, p. 8). “A mobility distribution model in which a customer’s major transportation needs are met over one interface and are offered by a service provider” (Hietanen, 2014, p. 2). “A  user-centric, intelligent mobility management and distribution system, in which an integrator brings together offerings of multiple mobility service providers, and provides end-users access to them through a digital interface, allowing them to seamlessly plan and pay for mobility” (Kamargianni et al., 2018, p. 3).

As illustrated in Smith and Hensher (2020), a shortcoming of these definitions is that they more often than not employ subjective and value-laden terms (e.g. major, comprehensive, usercentric, and seamless), which makes it arbitrary to apply them to objectively determine what is MaaS and what is not. To overcome this issue, this chapter suggests that MaaS can be understood simply as “a type of service that through a digital channel enables users to plan, book, and pay for multiple types of mobility services” (Smith, 2020, p. 3). Hence, at its core, MaaS introduces a digital one-stop shop (e.g. a smartphone app) that integrates scheduled and real-time information on several types of mobility services (e.g. public transport and taxi), as well as a unified method for gaining access to these. Additionally, MaaS might include possibilities for users to pay through periodical subscriptions and could synchronise the MaaS offering with transport policies (cf. level three and four in Sochor et al., 2018). Nevertheless, the fulfilment of these functionalities will not be a binary determinant for whether MaaS will be attractive to potential users. Rather, the attractiveness of MaaS can be described as a function of the perceived utility of the integrated mobility services (including interoperability and in comparison with alternative modes) and the service quality experienced by the users (Hensher et al., 2020; Karlsson et al., 2016; Smith et al., 2019a). As explained by Lyons et al. (2020), the ultimate MaaS service 34

Mobility as a Service

would go beyond multimodal information and payment functionalities to provide a level of service that is on a par with the private car in terms of the cognitive efforts required from users.

Expectations on how Mobility as a Service will influence public transport It has been widely acknowledged that transport systems are undergoing a technology-powered transformation that, inter alia, blurs the lines between public and private modes of mobility. The advent of new shared modes, such as ridesourcing and sharing systems for cars, bicycles, and scooters, has started to transform the mobility landscapes in many metropolitan cities, while autonomous technologies are anticipated to change the rules of the game even more dramatically in the not-so-distant future (Smith & Theseira, 2020). Veeneman (2019) notes that these new modes can have either synergetic or competitive relations with traditional public transport. For example, a survey of ridesourcing adoption in seven cities in the United States found that ridesourcing had attracted people away from public transport (Clewlow & Mishra, 2017), while a spatial analysis of docked bicycle sharing systems in two Chinese cities concluded that bicycle sharing can improve links between public transport stations and thus improve the catchment area and efficiency of urban public transport networks (Yang et  al., 2018). In order to pave the way for synergetic relations rather than competitive, Veeneman (2019) argues that governments have a key role in driving integration between modes. Although the importance of traditional policy fields such as infrastructure and land-use planning and transport regulation remains, MaaS provides a new promising tool for providing such integration. However, in a review of the expectation on MaaS among actors involved in MaaS developments in Sweden, Finland, and Australia, it was found that rationales regarding MaaS diverged (cf. Smith, 2020). Although a majority believed that MaaS will improve the integration between mobility services and thus help mobility services attract people away from private cars, others were either sceptical that MaaS will catch on or anticipated that MaaS will lead to undesirable outcomes such as induced travel demand, a modal shift away from public transport, increased transport exclusion (see also Chapter 26), or a higher complexity in planning and managing transport networks. In relation to public transport, it was contested whether MaaS will: bring new customer segments to public transport or primarily attract current public transport users, increase the cost-effectiveness of public transport by replacing ineffective routes or decrease its cost recovery by competing with the most efficient public transport routes, improve public transport authorities’ understanding of their users by collecting a fuller picture of travel habits or deplete this understanding by cutting public transport authorities’ access to user data, and enhance the brand value of public transport by associating it with modern mobility services or degenerate it by reducing public transport authorities to invisible suppliers (Smith, Sochor, & Karlsson, 2018). As discussed in the next section, it is, moreover, widely debated if the new operational tasks that MaaS entails should fall under public control or not. Depending on pathway, MaaS can either enlarge or reduce the current scope of public transport authorities. Regardless of this choice though, MaaS builds on new forms of relationships between many different actors, such as public and private mobility service providers, technology providers, and users. Consequently, MaaS is bound to change public transport authorities’ interorganisational relations. This will also likely put new requirements on their internal processes. For example, how are internal development projects prioritised vis-à-vis external ones, and how are strategies and action plans communicated? Furthermore, it will most probably demand new types of competences, such as 35

Göran Smith

managing networks of interdependent but more or less autonomous actors. Despite the limited data to date on the impacts of MaaS, and the diverging expectations, it thus seems reasonable to assume that MaaS will cause/require adjustments to the current structures and practices of public transport authorities in one way or another.

Pathways to governing Mobility as a Service developments The process of realising MaaS can be conceptualised as a case of disruptive and collaborative innovation (Smith, 2020). In other words, the realisation of MaaS can be understood as an innovation process that disrupts current structures and practices within the transport industry and which requires synchronised activities from many actors. Innovation processes are here understood to include multiple iterative stages that together describe how ideas are transformed into products, services, or processes that are perceived as new within a defined area and which are adopted, adapted, and used (cf. Rogers, 1995). Following this view, the MaaS innovation process can be divided into three interrelated core phases: development of MaaS during which operational MaaS services are materialised; diffusion of MaaS during which users get exposed to MaaS, form attitudes about MaaS, and decide to start using MaaS or not; and use of MaaS during which the use of MaaS is stabilised and becomes a mainstream part of transport systems (Smith, 2020). Since the realisation of MaaS requires activities from a diverse set of actors, the innovation process puts high demands on interorganisational and intersectorial co-ordination. For example, the MaaS development phase would benefit greatly from harmonisation of customer categories and technical data interfaces (Smith et al., 2020), while the MaaS diffusion and use phases require that the transport infrastructure and land-use planning at local, as well as regional and national, levels encourage the diffusion and use of mobility services (Smith  & Hensher, 2020; Smith et al., 2019a). Hence, the realisation of MaaS poses a great governance challenge (see also Chapter 2). Due to its unique capabilities (e.g. reforming regulation) and objectives (e.g. capturing public value), it can be argued that the public sector is best positioned to face this challenge. Beyond the need for synchronising and stimulating activities across sectors and policy fields, public sector governance of MaaS has, moreover, been advocated for based on the risks associated with MaaS, such as technological determinism, de-politicisation, and regulatory capture (Docherty et al., 2018; Pangbourne et al., 2018, 2020) (see also Chapter 1). In other words, public sector governance of MaaS seems to be needed to both facilitate MaaS developments and to steer the long-term trajectory towards contributing to policy objectives (Smith, 2020). Broadly speaking, governance, and theories of governance, are concerned with “the capacity to steer the economy and society, and involves identifying some effective means of deciding upon collective goals and then finding the means of reaching those goals” (Peters, 2014, p. 302) (see also Chapter 2). Since economies and societies are populated by a range of actors that have their own agencies and participate in various influential processes, governance is inherently a hybrid activity involving public-, private-, and civil-sector actors to different degrees and in different ways (ibid.). Nonetheless, the term governance is also used to describe a specific public administration theory (sometimes referred to as new public governance) which emphasises intersectorial collaboration between interdependent but autonomous actors as a key method for policymaking, negotiation, and implementation (Osborne, 2006; Rhodes, 1997). Here, the public sector manages sociotechnical developments by influencing public-private arenas through a mix of hands-on and hands-off interventions (Sørensen & Torfing, 2011, 2016). Handson intervention entails that the public sector participates in and oversees decision-making and is assumed to generate public sector accountability and control (Vento, 2019). Hands-off 36

Mobility as a Service

intervention implies that the public sector focuses on co-ordinating and incentivising external activities and is thought to give freedom for the private sector to innovative (ibid). Building on Kronsell and Mukhtar-Landgren (2018), Smith (2020) suggests that public sector actors that want to facilitate the realisation of MaaS can take four roles in the development, diffusion, and use of MaaS: MaaS Promoter, MaaS Partner, MaaS Enabler, and Laissez-Faire (a public sector actor can take several governance roles at once and change roles across MaaS innovation phases); see Table 3.1. The MaaS Promoter role entails that the public sector actor intervenes hands on in the realisation of MaaS. In other words, a MaaS Promoter manages MaaS developments by being directly involved in executing day-to-day decision-making and innovation (cf. the network participation and network management tools in Sørensen and Torfing [2009]). This could, for instance, imply mobilising resources to lead and co-ordinate the development of MaaS services and components during the MaaS development phase; funding MaaS marketing and/or acting as first customer of MaaS during the MaaS diffusion phase; and taking on the two new operative tasks during the MaaS use phase – MaaS integration, which encompasses collecting and curating data and tickets from mobility service providers, and MaaS operation, which comprises bundling MaaS services and delivering these to users (Smith et al., 2018). The MaaS Partner role entails that the public sector actor intervenes in the realisation of MaaS through a mix of hands-on and hands-off intervention in order to build collaboration networks and to support, participate in, and influence private sector-led innovation activities. This could, for instance, imply setting up and participating in knowledge sharing forums and MaaS experiments during the MaaS development phase, sharing user insights with those operating MaaS and legitimising their MaaS services during the MaaS diffusion phase, and taking on the MaaS integration task (but not the MaaS operation task) during the MaaS use phase. The MaaS Enabler role entails that the public sector actor intervenes hands off. The main goal is to create conducive conditions for MaaS developments carried out by private sector actors (cf. the network design and network framing tools in Sørensen and Torfing [2009]; see also Chapter 32). This could, for instance, imply pursuing institutional reforms and funding experimentation during the MaaS development phase, promoting the spread of mobility

Table 3.1 Pathways for governing MaaS developments, adapted from Smith (2020) Role

MaaS development

MaaS diffusion

MaaS use

MaaS Promoter

Takes the lead in transforming MaaS visions and ideas into operational services Participates in knowledge-sharing forums and in MaaS experiments

Acts as the lead customer for MaaS services and/ or advertises MaaS services Legitimises MaaS services, supports marketing, and shares user insights and data

Integrates mobility service data and tickets, and operates MaaS services

Opens for and funds MaaS experimentation and research

Promotes the diffusion of mobility services and/ or digital interfaces

MaaS Partner

MaaS Enabler

Laissez-Faire

Mediates data and products from mobility service providers to MaaS services

Feeds data and tickets for its own mobility services into MaaS services Monitors MaaS development processes while continuing business as usual

37

Göran Smith

services during the MaaS diffusion phase, and feeding data and tickets for the mobility services it manages to third-party resellers during the MaaS use phase. Finally, the Laissez-Faire role entails that the public sector actor withholds from intervening in the realisation of MaaS, and instead concentrates on monitoring the progress. Although this approach might appear passive, it could be a deliberate choice, that is, to ‘step away’ to give external actors leeway for innovation.

Examples from the Nordic countries MaaS has in recent years surfaced on many Nordic public transport authorities’ agendas. MaaS developments in the Nordic countries have, moreover, been frequently used as guiding examples in the global MaaS debate. In particular, the reform of transport legislation and the operation of Whim in Finland, and the UbiGo pilot in Sweden, have been scrutinised by both practitioners and researchers (e.g. Audouin & Finger, 2018; Hartikainen et al., 2019; Hensher et al., 2020; Hirschhorn et al., 2019; Kanger & Kivimaa, 2019; Mukhtar-Landgren & Smith, 2019; Sochor et al., 2016; Strömberg et al., 2018). Departing from the roles outlined in the previous section, the MaaS governance approaches of the public transport authorities in the metropolitan regions of Helsinki (Finland), Gothenburg (Sweden), and Oslo (Norway) are discussed next (for more detailed accounts of Nordic MaaS developments, see Hedegaard Sørensen et al. [2020] and Isaksson et al. [2019]). Helsinki: The capital city of Finland has in recent years gained a global reputation as a breeding and testing ground for mobility innovations. It has been argued that Helsinki is an ideal context for future mobility solutions, since its transport situation is manageable, Finland has a rich history in digital innovation and supports experimentation, the municipality has high targets for carbon dioxide emissions reduction in transport, and its residents are quick to adopt innovations (Davies, 2018). Likewise, Helsinki has time and again been hailed as a MaaS frontrunner in the transport industry press and in news media (e.g. Greenfield, 2014; Köllinger, 2018; Putkonen & Tikkanen, 2019). This status mainly comes down to two things. First, Helsinki is host to the first continuous operation of MaaS. The much-talked-about MaaS service Whim has been available to Helsinki residents since late 2016 and currently has around 10,000 users per month (cf. Hartikainen et al., 2019). Second, the regulation of personal mobility in Finland has been thoroughly reformed in recent years. A new general transport law has been introduced, which mostly entered into force during 2018. Its main objective is to streamline transport regulation and to open for and stimulate competition in order to pave the way for better mobility services (LVM, 2017). Amongst other things, mobility service providers are now required to ensure that essential, up-to-date data on its services is freely available via an open interface in a standard, easy-to-use, and computer-readable format. They must also give third parties access to the sales interfaces of their ticket and payment systems, via which it ought to be possible to reserve and/or purchase verifiable tickets on behalf of users. Third parties should be able to provide MaaS efficiently without restrictions (Finlex, 2017). In other words, both public and private mobility service providers in Finland are required to enable external MaaS innovation. The public transport authority in the greater Helsinki region (Helsingin seudun liikenne, HSL) has arguably not been in the driving seat of either the transport legislation reform or the Whim operation (Mukhtar-Landgren  & Smith, 2019) but has been affected by the former and has paved the way for the latter. In order to enable MaaS Global (the company operating Whim) and others to resell digital public transport tickets, HSL has developed an application programming interface (API) and a development portal (sales-api.hsl.fi) through which third-party 38

Mobility as a Service

actors can integrate and resell single tickets, daily and monthly passes, and zone extension. HSL has also developed general contract terms and signed a copy with MaaS Global, thus enabling the operation of Whim. Despite becoming the first public transport authority to fulfil the new transport legislation’s requirements on data and sales interface openness via these activities, a few of the other actively involved actors in Finland have alleged that HSL has impeded MaaS developments. Some have described HSL as slow moving and difficult to work with (Audouin & Finger, 2018), while the API has received criticism for having (purportedly) low flexibility and strict process requirements. All in all, HSL has thus far appeared to be sceptical of the market-driven model of MaaS in Finland. So also of the new general transport law, which forces mobility service providers to open data and ticket interfaces but does not oblige third-party resellers to share travel data back to mobility service providers. Still, HSL has been pushed to act as a MaaS Enabler of the development and operation of Whim (Audouin & Finger, 2019; Mukhtar-Landgren & Smith, 2019). As of spring 2020, two external actors are reselling public transport tickets through HSL’s API (MaaS Global and Korsisaari). Simultaneously, HSL is analysing different MaaS scenarios in order to identify the appropriate role for HSL in the future development, diffusion, and use of MaaS. The recognised alternatives range from keeping the status quo (as a MaaS Enabler) to taking a more hands-on role. Gothenburg: The public transport authority in Västra Götaland (Region Västra Götaland and Västtrafik, VGR/VT) started working actively on the MaaS topic in 2014. Its approach to governing MaaS has arguably gone through three major phases since then. VGR/VT’s interest in MaaS was awakened in association with a number of MaaS-related research and innovation projects that it participated in between 2011 and 2014. The most acclaimed outcome of these projects is the UbiGo pilot that was carried out in Gothenburg, which showcased that MaaS, at least in some cases, can attract users and influence their travel behaviour towards less car use (Sochor et al., 2015, 2016). Following the promising outcome of the UbiGo pilot, VGR/VT decided to procure a regionwide MaaS service by the end of 2014 (VGR, 2014). After a period of inactivity (characterised by internal indecisiveness and external criticism), VGR/VT commenced a dialogue with potential suppliers in spring 2016 in order to understand what procurement terms would be appropriate. The interest in bidding on the procurement was large, but it became evident during the dialogue that neither VGR/VT nor the potential bidders had sufficient knowledge and experience of MaaS to allow for a fruitful procurement process (Smith et al., 2017, 2019b). The procurement process was therefore cancelled. VGR/VT revised its MaaS strategy accordingly and chose to invest in a nationwide initiative that aimed to establish a publicly controlled national MaaS integration function. The logic behind this initiative was that a national MaaS integration function would lower the entry barriers for both those operating MaaS and mobility service providers and thus facilitate the development of many and diverse MaaS offerings that integrate multiple different mobility services (Smith et al., 2020). However, this initiative was discontinued as well due to lack of support from key actors (a governmental inquiry has since recommended the launch of a similar initiative and to, by law, require public transport authorities to participate [Government Offices of Sweden, 2020]). Consequently, VGR/VT had to amend its MaaS strategy once again. Its new (and current) tactic aims at enabling and stimulating private actors to integrate and resell digital public transport tickets as part of MaaS offerings. The hope is that this will stimulate private actors to establish a plurality of MaaS services, which in the short term generate knowledge of MaaS and of VGR/VT’s role(s) in MaaS and in the long term contribute to a more sustainable transport system. Since late 2018, VGR/VT has therefore focused on developing an API, integration 39

Göran Smith

processes, and contract terms for third-party resellers. These elements have been tested through a number of MaaS pilots, with the goal of moving from pilot contracts to regular contracts during 2020–2021 (the first regular contract was signed with the municipal parking company in Gothenburg in 2020). In parallel to this work, VGR/VT has been an initiator of the national roadmap for MaaS developments in Sweden, which, among other things, has co-ordinated and funded MaaS development activities and formulated impact targets for MaaS (cf. KOMPIS, 2017). Hence, VGRT/ VT has arguably thus far acted as a MaaS Promoter (e.g. via the national roadmap work), MaaS Partner (e.g. in the UbiGo pilot), and MaaS Enabler (e.g. by deciding to open for third-party resale) in the MaaS development phase. The first iteration of its MaaS strategy pointed towards a trajectory in which VGR/VT would act as a MaaS Promoter in the MaaS diffusion and MaaS use phases (partly via contract), while the second iteration was based on VGR/VT pulling back to a MaaS Enabler role during the MaaS use phase but still having some control over the MaaS integration task (via a jointly owned development company). The third and current strategy advocates a trajectory in which VGR/VT is involved as a MaaS Partner during the diffusion phase but only takes a MaaS Enabler role during the MaaS use phase by enabling external (more or less autonomous) actors to take on the MaaS integration and MaaS operation tasks. Put differently, VGR/VT currently plans to become increasingly less hands-on involved in MaaS as the realisation of MaaS progresses. The main reasoning behind this strategy is that it is assumed to provide the quickest path to realise, and thus learn about, MaaS under current institutional conditions. As of spring 2020, two MaaS pilots are ongoing in Västra Götaland, and three are about to commence. Oslo: The MaaS topic can in Oslo be traced back to 2015, when the public transport authority of the Oslo and Viken counties (Ruter) introduced a new strategy. Entitled M2016, this strategy’s main message was that Ruter’s focus would shift from public transport to mobility services (Ruter, 2015). Ruter is committed to contributing to long-term targets for the sustainability of Oslo and Viken (Hirschhorn et al., 2020). The shift in focus was thus motivated by the objective of giving residents in Oslo and Viken more mobility choices as well as by the need to make it easier for the residents to live without owning private cars. The latter was assumed as a prerequisite in order to be able to comply with the nationally formulated zero growth target for car traffic. In growing urban regions, such as Oslo, this goal implies that mobility services, along with cycling and walking, must absorb the projected growth in personal mobility demand. In practice, Ruter is tasked with developing a dense and flexible network of integrated, high-quality mobility services that make it possible for Oslo and Viken residents to leave their cars behind without compromising freedom and flexibility in their everyday lives. The M2016 strategy assumes that Ruter will be involved in developing these mobility services to some degree. At the same time, it expects private mobility services to take over some tasks that today are handled by Ruter (Ruter, 2015). In other words, the strategy predicts that Ruter’s role will change, but it is not clear exactly how yet. To identify its role(s) in a possible MaaS future, Ruter started to investigate MaaS during 2016. In an initial scenario analysis, Ruter evaluated different value chain setups (cf. Smith, Sochor,  & Karlsson, 2018). Drawing on the ongoing MaaS developments in Gothenburg, Vienna, Helsinki, Copenhagen, and Hannover, as well as on developments in the hotel and telecommunication industries, the scenario analysis concluded that a model in which the public sector takes the MaaS integration task and in which Ruter operates MaaS in competition with others was best suited to balance public control and the private sector’s innovation capacity. Ruter has since chosen to carry out a number of MaaS-related pilot projects. The overall plan is to pilot MaaS and other technology-related services in the period 2019–2021 in order to 40

Mobility as a Service

prepare for a large-scale pilot that is to be commenced around 2021. As of spring 2020, Ruter has one MaaS pilot in operation, which has drawn much inspiration from the 2013–2014 UbiGo pilot in Gothenburg. Ruter’s hope is that this pilot, in combination with the others, will generate the knowledge needed to make informed strategic decisions about the way forward in relation to MaaS as well as in relation to the future mobility service landscape in Oslo and Viken in general. In contrast to the developments in Sweden and Finland, Ruter was for some time the sole major actor in Norway that made comprehensive efforts to realise MaaS. However, a stateowned nationwide travel planner was introduced in 2018 (En-tur), and it has been indicated that the long-term plan is that this travel planner should enable users to find and pay for all public and private mobility services in Norway. However, how, when and to what extent this will materialise is still up in the air. Thus, it is not clear how it might influence Ruter’s plans. Nonetheless, Ruter has arguably taken a front-seat position in the development phase of MaaS in Norway thus far, primarily acting as a MaaS Promoter. According to the initial MaaS scenario work, it seems that Ruter plans to keep this role in the MaaS diffusion and MaaS use phases, too (recently, Ruter released a new overarching strategy that provides further evidence for this pathway [Ruter, 2020]). The choice of being hands-on involved in all MaaS development phases is largely based on the logic that Ruter is capable of developing and delivering competitive services and uniquely positioned to promote all types of sustainable travelling in Oslo and Viken. Ruter, moreover, believes that control over the MaaS operation task is needed to be able to nudge users’ travel behaviour (ibid.).

Concluding remarks As illustrated in the previous section, public transport authorities across the Nordic countries have taken somewhat different positions in relation to MaaS in terms of the mix of hands-on and hands-off intervention. It should be noted that these choices are influenced by both (intra-) organisational and external institutional conditions. For instance, the strong involvement of the national government in Finland has limited the perceived action space of Finnish public transport authorities (Mukhtar-Landgren & Smith, 2019). In contrast, public transport authorities in Sweden and Norway have had more front-seat (and less challenged) positions in the MaaS developments. Among other things, this has led to MaaS more frequently being framed as a means to assist public transport growth in Sweden and Norway, while the goals for MaaS more often have been motivated by national economic arguments in Finland (Smith et al., 2018). Recent MaaS developments across Europe demonstrate similar differences. In the United Kingdom, the public transport authority in West Midlands (Transport for West Midlands, TfWM) has supported MaaS Global in launching Whim. Although the user uptake has been slow and the service offering limited, this indicates that TfWM is taking a MaaS Enabler role. In Germany, the public transport authority in Berlin (Berliner Verkehrsbetriebe, BVG) has established a new business unit that focuses on MaaS. Empowered by a private technology provider, this unit has launched Jelbi – a MaaS service that allows users to find, book, and pay for BVG’s public transport offerings as well as a range of privately managed mobility services in Berlin. BVG is thus arguably taking a MaaS Promoter role. In the Netherlands, the National Ministry of Infrastructure and Water Management (Ministerie van Infrastructuur en Waterstaat, I&W) has set aside €20 million and created a framework agreement in order to prepare for procuring seven MaaS pilots to take place across the country during the next three years. The main goals are to speed up the realisation of MaaS in the Netherlands, gain experience in preparing for and operating MaaS, and collect empirical evidence of the actual impact of MaaS. Hence, the 41

Göran Smith

national strand of government is taking a front-seat role in MaaS development, which might influence the perceived action spaces and strategies of Dutch public transport authorities. In Belgium, the public transport company of the Flemish government (Vlaamse Vervoersmaatschappij De Lijn, De Lijn) launched an API for its data and tickets in 2016. Thus far, it has enabled seven external actors to launch MaaS-like services in Flanders. In parallel to these, De Lijn plans to offer its own MaaS service in the near future, in that case, acting both as a MaaS Enabler and as a MaaS Promoter. The MaaS developments have been slow moving and convoluted across the contexts covered in this chapter (Finland, Sweden, and Norway), and most public transport authorities have yet to move into the MaaS diffusion and MaaS use phases. Consequently, it is far too early to determine which MaaS governance pathways produce what outcomes. Yet some initial propositions can be made based on the experience to date. For example, in Smith (2020), the case is made that the ongoing processes in Sweden, Finland, and Australia indicate that the following interventions can support MaaS developments: • •

•

Establish an inspirational long-term vision for MaaS, which is based on transport policy objectives and links with other policy fields (e.g. land-use and transport infrastructure). Develop an innovation agenda for MaaS that co-ordinates mid- and short-term activities. Beyond measures directly related to the MaaS concept, this agenda should aim to strengthen mobility services and active mobility and to weaken the private car regime. Facilitate experimentation and joint knowledge generation by investing in experimentation activities and by establishing institutional conditions that favour learning and risk taking.

Importantly, these activities should be the subject of critical analyses as a means to revisit and refine governance approaches as MaaS evolves. Besides, the recommendations are untested and arguably quite general. Hence, further refinement and validation as well as contextspecific adaption and concretisation are needed to transform them into readily applicable advices. As illustrated in this chapter, public transport authorities’ MaaS governance pathways seem to diverge across jurisdictions as well as over time. Optimistically, this will create ample opportunities for comparative research studies that could contribute to this work. More generally, the ongoing and forthcoming MaaS operations should be thoroughly and transparently evaluated in order to improve the understanding of the societal effects of different types of MaaS in different types of situations, as well as of when what types of government interventions are appropriate.

Acknowledgements This chapter builds on my doctoral thesis ‘Making Mobility as a Service: Towards Governance Principles and Pathways’ (Smith, 2020) and its appended papers. The research reported in those publications was done in collaboration with many MaaS researchers and practitioners to whom I am deeply thankful. It should also be acknowledged that in parallel to my doctoral research project, I have been employed as a civil servant at Region Västra Götaland (denoted VGR/VT in this chapter), where I have been involved in drafting and implementing its MaaS strategies. Finally, the research was funded by Region Västra Götaland and K2 – The Swedish Knowledge Centre for Public Transport and supported by Chalmers University of Technology and the Institute of Transport and Logistics Studies at the University of Sydney Business School.

42

Mobility as a Service

References Audouin, M., & Finger, M. (2018). The development of mobility-as-a-service in the Helsinki metropolitan area: A multi-level governance analysis. Research in Transportation Business & Management, 27, 24–35. doi:10.1016/j.rtbm.2018.09.001 Audouin, M., & Finger, M. (2019). Empower or thwart? Insights from Vienna and Helsinki regarding the role of public authorities in the development of MaaS schemes. Transportation Research Procedia, 41, 6–16. doi:10.1016/j.trpro.2019.09.003 Clewlow, R. R., & Mishra, G. S. (2017). Disruptive transportation: The adoption, utilization, and impacts of ride-hailing in the United States. https://escholarship.org/content/qt82w2z91j/qt82w2z91j.pdf Davies, G. (2018). Special report: The future of mobility in Helsinki. Automotive World. www.automotive world.com/research/special-report-the-future-of-mobility-in-helsinki/ Docherty, I., Marsden, G., & Anable, J. (2018). The governance of smart mobility. Transportation Research Part A: Policy and Practice, 115, 114–125. doi:10.1016/j.tra.2017.09.012 Finlex. (2017). Act on transport services (2017:320). www.finlex.fi/en/laki/kaannokset/2017/en20170320_ 20180731.pdf Government Offices of Sweden. (2020). Ett nationellt biljettsystem för all kollektivtrafik. www.regeringen. se/499447/contentassets/339c31bcd5a74b12b92e689b898dfb37/ett-nationellt-biljettsystem-forall-kollektivtrafik-sou-202025 Greenfield, A. (2014, July  10). Helsinki’s ambitious plan to make car ownership pointless in 10  years. The Guardian. www.theguardian.com/cities/2014/jul/10/helsinki-shared-public-transport-plan-carownership-pointless Hartikainen, A., Pitkänen, J. P., Riihelä, A., Räsänen, J., Sacs, I., Sirkiä, A., & Uteng, A. (2019). Whimpact: Insights from the world’s first mobility-as-a-service (MaaS) system. Ramböll. https://ramboll.com/whimpact Hedegaard Sørensen, C., Isaksson, K., Oldbury, K., Paulsson, A., Smith, G., & Rignell, M. (2020). Kollektivtrafikmyndigheter och smart mobilitet: Nordiska erfarenheter och perspektiv på MaaS och autonoma bussar (K2 Outreach 2020:1). www.k2centrum.se/sites/default/files/fields/field_uppladdad_rapport/k2_out reach_2020_1.pdf Heikkilä, S. (2014). Mobility as a service – a proposal for action for the public administration, case Helsinki [Master’s thesis, Aalto University]. https://aaltodoc.aalto.fi/handle/123456789/13133 Hensher, D. A., Ho, C. Q., Mulley, C., Nelson, J. D., Smith, G., & Wong, Y. Z. (2020). Understanding mobility as a service (MaaS): Past, present and future. Elsevier. Hietanen, S. (2014). “Mobility as a service” – the new transport model? ITS  & Transport Management Supplement: Eurotransport, 12(2), 2–4. www.itsineurope.com/its10/media/press_clippings/ITS Supp_ et214.pdf Hirschhorn, F., Paulsson, A., Sørensen, C. H.,  & Veeneman, W. (2019). Public transport regimes and mobility as a service: Governance approaches in Amsterdam, Birmingham, and Helsinki. Transportation Research Part A: Policy and Practice, 130, 178–191. doi:10.1016/j.tra.2019.09.016 Hirschhorn, F., van de Velde, D., Veeneman, W., & ten Heuvelhof, E. (2020). The governance of attractive public transport: Informal institutions, institutional entrepreneurs, and problem-solving know-how in Oslo and Amsterdam. Research in Transportation Economics. doi:10.1016/j.retrec.2020.100829 Isaksson, K., Oldbury, K., Hedegaard Sørensen, C., Paulsson, A., Smith, G., & Rignell, M. (2019). Problembilder som formar framtidens mobilitet – exemplet “mobilitet som tjänst” i kollektivtrafikplanering. In J. Syssner (Ed.), Ett nytt kontrakt för samhällsbyggande (pp. 177–200). Linnefors Förlag. Kamargianni, M., Li, W., Matyas, M., & Schäfer, A. (2016). A critical review of new mobility services for urban transport. Transportation Research Procedia, 14, 3294–3303. www.sciencedirect.com/science/ article/pii/S2352146516302836 Kamargianni, M., Matyas, M., Li, W., Muscat, J., & Yfantis, L. (2018). The MaaS dictionary. www.maaslab. org/copy-of-maas-publications Kanger, L., & Kivimaa, P. (2019). The emergence and consolidation of mobility-as-a-service in Finland. www. tipconsortium.net/wp-content/uploads/2019/04/finland-TLHC-v5.pdf Karlsson, I. C. M., Sochor, J., & Strömberg, H. (2016). Developing the “service” in mobility as a service: Experiences from a field trial of an innovative travel brokerage. Transportation Research Procedia, 14, 3265–3273. doi:10.1016/j.trpro.2016.05.273 Köllinger, C. (2018, July 20). How Helsinki became a “mobility as a service” leader. Eltis – The Urban Mobility Observatory. www.eltis.org/discover/news/how-helsinki-became-mobility-service-leader

43

Göran Smith KOMPIS. (2017). Roadmap for combined mobility as a service in Sweden. https://kompis.me/wp-content/ uploads/2018/01/roadmapp_eng_10jan_2018-1.pdf Kronsell, A., & Mukhtar-Landgren, D. (2018). Experimental governance: The role of municipalities in urban living labs. European Planning Studies, 26(5), 988–1007. doi:10.1080/09654313.2018.1435631 LVM. (2017, October  27). Act on transport services. LVM (Press release). www.lvm.fi/-/act-on-transportservices-955864 Lyons, G., Hammond, P., & Mackay, K. (2020). The importance of user perspective in the evolution of MaaS. Transportation Research Part A: Policy and Practice, 131, 20–34. doi:10.1016/j.tra.2018.12.010 Mukhtar-Landgren, D., & Smith, G. (2019). Perceived action spaces for public actors in the development of mobility as a service. European Transport Research Review, 11(1), 32. doi:10.1186/s12544-019-0363-7 Osborne, S. P. (Ed.). (2006). The new public governance? Public Management Review, 8(3), 377–387. doi:10.1080/14719030600853022 Pangbourne, K., Mladenović, M. N., Stead, D., & Milakis, D. (2020). Questioning mobility as a service: Unanticipated implications for society and governance. Transportation Research Part A: Policy and Practice, 131, 35–49. doi:10.1016/j.tra.2019.09.033 Pangbourne, K., Stead, D., Mladenović, M.,  & Milakis, D. (2018). The case of mobility as a service: A  critical reflection on challenges for urban transport and mobility governance. In G. Marsden  & L. Reardon (Eds.), Governance of the smart mobility transition (pp. 33–48). Emerald Publishing Limited. doi:10.1108/978-1-78754-317-120181003 Peters, B. G. (2014). Is governance for everybody? Policy and Society, 33(4), 301–306. doi:10.1016/ j.polsoc.2014.10.005 Putkonen, R.,  & Tikkanen, U. (2019, August  21). MaaS, EVs and AVs: How Helsinki became a transport trendsetter. Intelligent Transport. www.intelligenttransport.com/transport-articles/86384/ maas-ev-av-helsinki-trendsetter/ Rhodes, R. A. (1997). Understanding governance: Policy networks, governance, reflexivity and accountability. Open University Press. Rogers, E. M. (1995). Diffusion of innovations. Simon and Schuster. Ruter. (2015). M2016: Fra dagens kollektivtrafikk til morgendagens mobilitetsløsninger. Ruter. https://m2016. ruter.no/content/uploads/2015/08/RUTE0057_M2016_10.08.15_Low.pdf Ruter. (2020). Målbilde for bærekraftig bevegelsesfrihet. Ruter. https://ruter.no/globalassets/dokumenter/ ruterrapporter/malbilde-barekraftig-bevegelsesfrihet-2020.pdf Smith, G. (2020). Making mobility-as-a-service: Towards governance principles and pathways [Doctoral thesis, Chalmers University of Technology]. Smith, G.,  & Hensher, D. A. (2020). Towards a framework for mobility-as-a-service policies. Transport Policy, 89, 54–65. doi:10.1016/j.tranpol.2020.02.004 Smith, G., Sochor, J., & Karlsson, I. C. M. (2017). Procuring mobility as a service: Exploring dialogues with potential bidders in West Sweden. Presented at the 24th World Congress on Intelligent Transportation Systems. Smith, G., Sochor, J.,  & Karlsson, I. C. M. (2018). Mobility as a service: Development scenarios and implications for public transport. Research in Transportation Economics, 69, 592–599. doi:10.1016/ j.retrec.2018.04.001 Smith, G., Sochor, J., & Karlsson, I. C. M. (2019a). Adopting mobility-as-a-service: An empirical analysis of end-users’ experiences (pp. 86–98). ICoMaaS 2019 Proceedings. Smith, G., Sochor, J., & Karlsson, I. C. M. (2019b). Public–private innovation: Barriers in the case of mobility as a service in West Sweden. Public Management Review, 21(1), 116–137. doi:10.1080/14719 037.2018.1462399 Smith, G., Sochor, J., & Karlsson, I. C. M. (2020). Intermediary MaaS integrators: A case study on hopes and fears. Transportation Research Part A: Policy and Practice, 131, 163–177. doi:10.1016/j.tra.2019.09.024 Smith, G., Sochor, J., & Sarasini, S. (2018). Mobility as a service: Comparing developments in Sweden and Finland. Research in Transportation Business & Management, 27, 36–45. doi:10.1016/j.rtbm.2018.09.004 Smith, G.,  & Theseira, W. (2020). Workshop 5 report: How much regulation should disruptive transportation technologies be subject to? Research in Transportation Economics, 83(100915). doi:10.1016/ j.retrec.2020.100915 Sochor, J., Arby, H., Karlsson, I. C. M.,  & Sarasini, S. (2018). A  topological approach to mobility as a service: A proposed tool for understanding requirements and effects, and for aiding the integration of societal goals. Research in Transportation Business & Management, 27, 3–14. doi:10.1016/ j.rtbm.2018.12.003

44

Mobility as a Service Sochor, J., Karlsson, I. C. M., & Strömberg, H. (2016). Trying out mobility as a service: Experiences from a field trial and implications for understanding demand. Transportation Research Record: Journal of the Transportation Research Board, 2542, 57–64. doi:10.3141/2542-07 Sochor, J., Strömberg, H., & Karlsson, I. C. M. (2015). An innovative mobility service to facilitate changes in travel behavior and mode choice. Presented at the 22nd World Congress on Intelligent Transportation Systems. Sørensen, E., & Torfing, J. (2009). Making governance networks effective and democratic through metagovernance. Public Administration, 87(2), 234–258. doi:10.1111/j.1467-9299.2009.01753.x Sørensen, E., & Torfing, J. (2011). Enhancing collaborative innovation in the public sector. Administration & Society, 43(8), 842–868. doi:10.1177/0095399711418768 Sørensen, E., & Torfing, J. (2016). Metagoverning collaborative innovation in governance networks. The American Review of Public Administration, 47(7), 826–839. doi:10.1177/0275074016643181 Strömberg, H., Karlsson, I. C. M., & Sochor, J. (2018). Inviting travelers to the smorgasbord of sustainable urban transport: Evidence from a MaaS field trial Transportation, 45(6), 1655–1670. doi:10.1007/ s11116-018-9946-8 Veeneman, W. (2019). Chapter three  – public transport in a sharing environment. In E. Fishman (Ed.), Advances in transport policy and planning (Vol. 4, pp.  39–57). Academic Press. doi:10.1016/ bs.atpp.2019.10.002 Vento, I. (2019). Hands-off or hands-on governance for public innovation? A comparative case study in the EU cohesion policy implementation in Finland. International Journal of Public Administration, 1–11. doi:10.1080/01900692.2019.1665065 VGR. (2014). Sammanträde med kollektivtrafiknämnden den 4 December  2014. https://alfresco.vgregion. se/alfresco/service/vgr/storage/node/content/workspace/SpacesStore/52b55bec-0833-4265-9fbff6e29194ff15/Kollektivtrafiknämnden%202014-12-04%20Protokoll.pdf?a=false&guest=true Yang, X. H., Cheng, Z., Chen, G., Wang, L., Ruan, Z. Y., & Zheng, Y. J. (2018). The impact of a public bicycle-sharing system on urban public transport networks. Transportation Research Part A: Policy and Practice, 107, 246–256. doi:10.1016/j.tra.2017.10.017

45

4 PUBLIC TRANSPORT USE The ‘soft’ side of the story Veronique Van Acker, Sazkia Sandoval and Mario Cools

Introduction Understanding the determinants of the demand for public transport is essential in designing attractive public transport systems that can offer an alternative to the dominating car use in many regions and countries (see also Chapter 21). But these determinants have been researched from many different perspectives (Polat, 2012), particularly the economic determinants such as price. In times when public transport is often deregulated, decentralised and privatised, it is not surprising that many studies seek to understand how sensitive demand for public transport is with respect to fare changes, as these processes are often associated with price increases (see, for example, Nijkamp et al., 2000, for an overview). But economic factors are not the only determinants of public transport demand. Other studies have also used structural determinants such as demographic and geographic factors. For example, in their longitudinal study of public transport use in 62 urban areas in France, Bresson et al. (2004) found a negative income elasticity, as with other economic studies. But by extending their analysis to include structural determinants such as urban sprawl and the level of car ownership, they noticed how the ‘income effect’ is actually a ‘motorisation effect’. As income increases, it seems to be associated with urban sprawl and higher car ownership, which eventually discourages public transport use as well. Compared to the economic determinants, less is known about how people really think and feel about public transport. Although there is literature on people’s satisfaction with public transport from a consumer perspective (see, for example, van Lierop et al., 2018, for an overview), relatively less is known about people’s opinion and attitudes towards public transport. The latter presupposes a broader behavioural perspective where the use of public transport is situated within people’s daily life. The aim of this chapter is, therefore, to present insights into this ‘soft’ side of public transport use. The chapter specifically focuses on the role of attitudes in explaining public transport use while taking into account the interplay with other behavioural decisions related to residential location, vehicle ownership and activity behaviour. Moreover, it will also illustrate how these attitudes are impacted by personal values.

Background This section explains basic concepts and their interrelationships with public transport use, as depicted in the model structures in Figure  4.1. This figure is based on three key elements: 46

Public transport use

Figure 4.1 Model structures summarising complex relationships between values, attitudes and behaviours related to public transport use

values, attitudes and behaviour, with behaviour not only related to public transport use but also to residential location, car ownership and daily activity patterns. The theoretical basis for interactions between these key elements can be found in travel behaviour studies considering a choice hierarchy (e.g., Ben-Akiva, 1973; Salomon & Ben-Akiva, 1983; Van Acker et al., 2010) on the one hand and extended by value-attitude-behaviour studies (e.g., Homer  & Kahle, 1988) on the other. The public transport literature reviewed in this section is therefore limited to studies using a behavioural perspective and those empirical studies using survey data about individuals’ travel behaviour.

Public transport use as part of a hierarchy of decisions A behavioural perspective on public transport use positions daily travel behaviour within a wider set of decisions (see grey shaded box in the centre of Figure 4.1). It starts with activitybased studies acknowledging travel behaviour as a derived demand from the activities in which someone wants to participate, which generates the need to travel (see for example McNally & Ridt, 2007, for an overview). Complex activity patterns with multiple activities combined in one trip chain, known as trip chaining, are usually considered a barrier to public transport use. As trip chaining becomes more complex, public transport does not usually offer the same flexibility as a car. Using data from Sydney, Australia, Hensher and Reyes (2000) noticed how these barriers to using public transport are even stronger in households owning multiple vehicles. They found a strong negative influence of vehicle ownership on public transport use not only for complex trip chains structured around work activity but also for simple home-to-work trip 47

Van Acker, Sandoval and Cools

chains. However, the effect of vehicle ownership did not appear as strong for non-work trip chains. Paulley et al. (2006) concluded that increasing car ownership might decrease demand for public transport, especially in the bus market. Consequently, once people own a vehicle, they will seek to use it. Currie and Delbosc (2011) also pointed out that complex activity patterns do not necessarily constitute a barrier to public transport use, especially when you distinguish between the train, tram and bus. Using data from Melbourne, Australia, they found trip chains, especially non-work trip chains, to be more complex for rail and tram than for car. Such nonwork trip chains by train or tram were particularly common in the central city, where a large range of services and activities are clustered, facilitating the use of public transport in complex activity patterns and traffic congestion and parking difficulties limiting car use. This finding illustrates how the built environment is also important in understanding public transport use. In their meta-analysis, Ewing and Cervero (2010) concluded that the built environment has a modest effect on travel behaviour, with bus and train use most strongly influenced by proximity to public transport and street network design variables and second by land-use diversity. Using data from ten communities in Tyne and Wear, northeast England, UK, Aditjandra et al. (2016) also found that accessibility is one of the most important features of the built environment explaining changes in public transport use following a residential relocation. Findings like these suggest an interconnection between public transport use, activity behaviour, vehicle ownership and the built environment. Ben-Akiva (1973) and Salomon and BenAkiva (1983), for example, argued how a hierarchy of decisions exists ranging from long-term lifestyle decisions to medium-term decisions on vehicle ownership, residential and workplace location and short-term decisions on daily activities and travel (such as activity type, activity duration, destination, route and transport mode). They explain how, within each time block, decisions are made jointly, but decisions in the lower time block are made conditional on those in the upper time block.

Public transport use and the influence of attitudes Decisions in the previously mentioned hierarchy are also determined by, among others, reasoned influences such as perceptions, attitudes and preferences [see arrow (1) in Figure 4.1]. While perceptions refer to general beliefs about various aspects of a subject, such as the built environment, vehicle ownership, activities and travel, attitudes refer to a positive, negative or neutral evaluation of these aspects. Preferences are subsequently developed based on these attitudes and perceptions (Van Acker et al., 2010). For example, someone might believe that using public transport is beneficial for the environment but at the same time find public transport stressful, unpleasant and uncomfortable. The combined perceptions of all these aspects might eventually result in a negative attitude towards public transport and a low preference to use it. Attitudes might have an important influence on behaviour, as explained by the theory of planned behaviour (Ajzen, 1991). This also applies to the use of public transport. For example, based on a cluster analysis of a wide variety of attitudinal statements (related to norms, environmental awareness, perceived behavioural control and habits), Anable (2005) found six different segments among visitors of National Trust sites in the northwest of the United Kingdom. Segments characterised by a positive attitude towards car alternatives, less strong attachment to the car, stronger moral norms and greater perceived control also exhibited stronger intentions to use alternative transport modes to the car. Attitudes may even have a more important role to play for public transport use compared to car use. For example, using data from urban regions in Norway,

48

Public transport use

Şimşekoğlu et al. (2015) found how a positive attitude towards public transport remains important in explaining public transport use, even when controlling for travel habits, while this was not the case for car use. Van Acker et al. (2020) also found how a pro-travel attitude – characterised by statements such as ‘I like travelling by public transport’ and ‘Getting stuck in traffic does not bother me too much’ – is one of the most important determinants of public transport use in Sydney, Australia. Recent studies have also signalled a significant indirect effect of travel attitudes on public transport use. People who prefer using public transport might also self-select themselves into a residential neighbourhood with access to good quality public transport services. This means that travel attitudes not only have a direct influence on public transport use (as already discussed previously) but also indirectly through the residential location choice (see Figure 4.1a). This is known as the question of ‘residential self-selection’ (Cao et al., 2009; Naess, 2009). For example, Ettema and Nieuwenhuis (2017) found that, compared to other transport modes, train users are most likely to choose a residential neighbourhood that is conducive to using their preferred mode of transport. An earlier study by van Wee et al. (2002) reached similar conclusions: transport mode preference plays a role in residential location choices, especially for public transport lovers. Moreover, adding transport mode preferences increased the explanatory power of the analyses, and this increase was found again to be the largest for public transport. However, residential location choices can be constrained by many other factors, and not everybody can choose a residential neighbourhood that meets their travel attitudes. In that case, ‘residential dissonance’ exists (Kamruzzaman et  al., 2016). However, after living somewhere and experiencing the residential neighbourhood for some time, this dissonance might become weaker. Travel attitudes might eventually change and come into line with the built environment attributes of that residential location. This illustrates how the built environment also impacts on travel attitudes (see Figure 4.1b). This process is known as ‘residential determination’ (Lin et al., 2017; De Vos et al., 2018) and is sometimes found to be more important than residential self-selection (Ewing et al., 2016). Based on a sample of recent movers in Ghent, Belgium, De Vos et al. (2018) found how moving to a more urbanised neighbourhood eventually resulted in a more positive attitude towards public transport and also active travel.

Public transport use and the influence of values While public transport attitudes may thus change over time due to processes like residential determination, values are considered deeply rooted and enduring beliefs. As mentioned before, attitudes are an evaluative response and refer to how someone likes or dislikes things, people and objects. Values are much more abstract and determine what someone considers right or wrong. These abstract cognitions then serve as prototypes from which attitudes and behaviour are constructed, finally resulting in a value-attitude-behaviour hierarchy [see arrow (2) in Figure 4.1]. Homer and Kahle (1988) first introduced this hierarchy in their study on natural food shopping. Since then, this hierarchy has been studied in many areas, but applications in travel behaviour research remain limited. One exception is the study by Paulssen et al. (2014). Using data from a sample of German commuters, they found that personal values of power, hedonism and security affect personal travel attitudes towards flexibility, comfort and convenience and ownership, which in turn influence the decision to use public transport over cars (or vice versa). This eventually resulted in an indirect effect of values on mode choice. The indirect effect of values towards security was negligible, but values towards hedonism and power both had strong negative indirect effects on public transport use.

49

Van Acker, Sandoval and Cools

Research design Each of the two model structures in Figure 4.1 could be considered a series of regression equations for which a structural equation model (SEM) is advanced in this chapter.

Methodology: structural equation model SEM is useful where one specific variable is an explanatory variable in one equation (e.g., travel attitudes influencing public transport use) and at the same time an outcome variable in another equation (e.g., travel attitudes influenced by values). Instead of estimating these equations one by one using, for example, separate regression analyses, SEM simultaneously estimates this set of related equations. Instead of using ‘independent’ and ‘dependent’ variables as in a regression analysis, SEM uses concepts of ‘exogenous’ and ‘endogenous’ variables. Exogenous variables influence other variables and are not influenced by any other variable in the model (and thus resemble the independent variables of a regression analysis). Endogenous variables, on the other hand, are impacted by these exogenous variables (like the dependent variable of a regression analysis) but can at the same time also impact other endogenous variables in the model (like the independent variables of a regression analysis). Furthermore, variables in a SEM can be ‘manifest’ or ‘latent’. Manifest variables are directly observed and measured, while latent variables are not. Latent variables can only be indirectly measured by their underlying indicators (Byrne, 2010; Kline, 2015). For example, ‘built environment’ in Figure 4.1 will be defined as a latent variable measured by the two indicators of having a bus stop within 500 m of the residence and having a train station within 2000 m. A SEM is estimated by matching the empirically based covariance matrix of the data with the resulting model-based covariance matrix. Maximum likelihood (ML) is often used as the estimation technique, although this assumes a multivariate normal distribution of all endogenous variables in the model. However, many of the endogenous variables in Figure 4.1 violate this assumption. Consequently, ML with bootstrapping was used instead, as this has proven to be a good alternative for analysis with non-normally distributed data (Byrne, 2010), as well as in travel behaviour research (Ory & Mokhtarian, 2010). Bootstrapping uses random sampling with replacement. It draws multiple subsamples of the same size as the original one and provides data for hypothesis testing. For each model in Figure 4.1, all relationships were estimated simultaneously using the software package IBM SPSS AMOS 22 Graphics. Only significant relationships were retained (in this case defined as relationships with p < 0.10). After refitting the reduced model, modification indices were considered. Modification indices specify the reduction in the overall model fit chi-square for each possible relationship that can be added to the model. AMOS always suggested many additional relationships, but only those that made sense theoretically were added.

Data For this study, data were used from a 2016 Internet survey on values and travel behaviour with respondents from Brussels, Belgium. Respondents were recruited using a snowball approach based on the distribution of flyers in the city of Brussels, Belgium (e.g., on KU Leuven campus in Brussels). The survey not only included questions on values but also spatial and mobility attitudes. After data cleaning, a final sample of 334 respondents was retained for further analysis 50

Public transport use Table 4.1 Descriptive sample statistics Frequency Gender Education Professional occupation Student Income Partner Car driving license Season ticket for bus, tram, metro Season ticket for train

Age Number of cars per household

43% male – 57% female 16% low – 84% high 49% professionally active – 51 not professionally active 32% yes – 68% no 83% low – 17% high 52% yes – 17% no 83% yes – 17% no 57% yes – 43% no 11% yes – 89% no Min.

Max.

Mean

Std. dev.

13 0

79 6

37.3 1.2

16.61 1.04

(see Table 4.1). The sample is characterised by a high level of education and income. But despite sampling on campus, it is not dominated by students.

Key variables Values Values were measured using the portrait values questionnaire (PVQ) developed by Schwartz (2003). This is a well-established 21-item measure of human values also used in, for example, the European Social Survey (ESS). The PVQ includes 21 short items describing a person’s goals, aspirations or wishes. Respondents are asked to compare the item stated to themselves (‘How much is this person like you?’) and to indicate the extent to which it is indeed applicable on a 6-point Likert scale (with 1 = very much like me, and 6 = not like me at all). Table 4.2 summarises how these 21 items can be combined to measure 10 basic human values. Scores on the different items belonging to a specific value are averaged out. The last column of Table 4.2 shows the mean score per value of the sample used in this chapter. The strongest values in this sample, indicated by the lowest scores, are related to benevolence and universalism.

Spatial and travel attitudes In addition to values, the survey also included questions on spatial and travel attitudes. Respondents were asked to rate on a 7-point Likert scale (1 = not important at all, 7 = extremely important) how important various aspects are in their residential location choices and transport mode choices. These aspects were found to be highly correlated with each other, and the number of aspects could therefore be reduced by means of factor analysis (principal axis factoring with promax rotation). The number of factors was determined based on the interpretability of the factors, the interpretation of the scree-plot and the eigenvalues larger than 1. The factor analysis resulted in four spatial attitudes (i.e., pro-safe and attractive environment, pro-social interaction, pro-accessibility of mandatory activities, pro-accessibility of non-mandatory activities) and four travel attitudes (i.e., pro-privacy and comfort, pro-time saving, pro-sustainability, pro-weather protection). Results of these two-factor analyses are summarised in Table 4.3. 51

Van Acker, Sandoval and Cools Table 4.2 Human values based on Schwartz’s PVQ Human value

Definition

Specific item from 21-item instrument

Mean score (std. dev.)

Power

Social status and prestige, control or dominance over people and resource Personal success through demonstrating competence according to social standards Pleasure and sensuous gratification for oneself Excitement, novelty, and challenge in life Independent thought and action – choosing, creating, exploring Understanding, appreciation, tolerance, and protection for the welfare of all people and for nature Preservation and enhancement of the welfare of people with whom one is in frequent personal contact Respect for, commitment to, and acceptance of the customs and ideas that traditional culture or religion impose on the self Restraint of actions, inclinations, impulses likely to upset or harm others and to violate social expectations or norms Safety, harmony and stability of society, of relationships and of self

Wealth, tell others

3.84 (1.067)

Show abilities, successful

3.15 (1.163)

Good time/spoil self, fun/pleasure New experience, risk/ excitement Creativity/originality, free/own decisions

2.59 (1.026)

Achievement

Hedonism Stimulation Self-direction

Universalism

Benevolence

Tradition

Conformity

Security

2.81 (1.117) 2.25 (0.865)

Equality for all, understand/listen, behave properly

2.16 (0.757)

Help others, loyal/ devoted

2.16 (0.806)

Inconspicuous/modest, tradition

3.19 (1.003)

Follow rules, behave properly

3.35 (1.012)

Secure surroundings, state protect

3.20 (1.194)

Source: Schwartz & Bardi, 2001; this study’s sample Table 4.3 Pattern matrix with factor loadings of four spatial attitudes and four travel attitudes Spatial attitudes (explained variance: 47.8%) Pro-safety and attractiveness Social safety, low crime Traffic safety

Pro-social interaction Pro-accessibility of mandatory activities

0.757 0.713

52

Pro-accessibility of non-mandatory activities

Public transport use

Spatial attitudes (explained variance: 47.8%) Pro-safety and attractiveness Neatness, tidiness Sufficient parking Appearance of buildings, architecture Quietness Good contact with neighbours Frequent contact with neighbours Presence of bike paths Presence of green areas Presence of sidewalks Close to public transport Close to shops Close to work/school Close to family and friends Close to leisure activities

Pro-social interaction Pro-accessibility of mandatory activities

Pro-accessibility of non-mandatory activities

0.550 0.527 0.473 0.411 0.772 0.771 0.601 0.433 0.415 0.896 0.683 0.383 0.634 0.493 Travel attitudes (explained variance: 39.6%) Pro-privacy and Pro-time saving comfort

Privacy-offering Image Comfortable Relaxing Time-saving Reliable Flexible Healthy Environment-friendly Cheap Safe Clothing Weather

Pro-sustainability Pro-weather protection

0.734 0.518 0.516 0.430 0.745 0.638 0.496 0.706 0.628 0.317 0.285 0.740 0.690

Residential location, vehicle ownership and activity behaviour Next to values and attitudes, other key variables in the different model structures in Figure 4.1 refer to the built environment of the residential location, vehicle ownership and activity behaviour. The built environment of the residential location was not questioned directly in the survey. However, the survey included two questions that could be used as indicators of the built 53

Van Acker, Sandoval and Cools

environment of the residential location. Consequently, residential location is considered in this chapter as a latent variable. Respondents indicated the distance from their residence to the nearest public transport stop. Of the respondents, 17.7% indicated living within 500 m of a bus, tram or metro stop, and 47.9% indicated living within 2000 m of a railway station. These two spatial variables are used as indicators of an ‘urban residential location’. Regarding vehicle ownership, the survey asked respondents about the number of cars owned in their household. On average, households owned 1.2 cars (see also Table 4.1). With respect to activity behaviour, respondents were asked how many work, school, leisure, shopping, service, business, drop-off and pick-up and touring activities they did on a weekly basis. Based on this, the weekly total number of activities was calculated. Respondents had, on average, 15 activities per week.

Public transport use The final outcome variable in this SEM analysis refers to public transport use. Respondents were not only asked about their weekly activities but also about their weekly number of trips by various transport modes. Based on this, the weekly total number of trips was calculated, as well as the percentages for each transport mode. Respondents make, on average, 16 trips per week. The majority of these trips are by car as a driver (32.1%), followed by walking (25.3%) and bus, tram and metro (19.2%). The share of other transport modes is remarkably lower (10.5% car as a passenger, 7.7% bicycle, 4.1% train, 1.1% moped/motorcycle). The SEM analysis in this chapter focuses on explaining the share of public transport (bus/tram/metro and train).

Results This section summarises the results of two SEMs representing the two model structures in Figure 4.1, measuring the effect of attitudes and values on public transport use while also accounting for residential self-selection (see Figure 4.2) and residential determination (see Figure 4.3), respectively. First, central in Figures 4.2 and 4.3 is a choice hierarchy with long-term urban residential location and mid-term vehicle ownership impacting daily activity behaviour and public transport use. This choice hierarchy is very simple for train use: residing in an urban built environment directly encourages train use. Other aspects of this choice hierarchy (vehicle ownership, activity behaviour) do not have a significant influence on train use. In contrast to train use, the choice hierarchy is much more complex for the use of bus, tram and metro. Residing in an urban built environment discourages vehicle ownership, and low vehicle ownership in turn is associated with high bus, tram and metro use. Complex activity patterns, on the other hand, discourage bus, tram and metro use. This is consistent with findings of earlier studies (Hensher & Reyes, 2000; Paulley et al., 2006; Currie & Delbosc, 2011). Second, attitudes are important predictors in this choice hierarchy. Three spatial attitudes have a direct influence on the choice of an urban residential location. Respondents with a positive attitude towards having access to mandatory activities such as work and school are also more likely to reside in an urban built environment. The reverse holds for respondents with a positive attitude towards social interaction and safety and attractiveness. Because of its interaction with the urban residential location, spatial attitudes eventually also have an indirect effect on public transport use. This indirect effect is significant but generally small compared to other variables. This confirms earlier research on modal choices in Flanders, Belgium (Van Acker et al., 2011). Only the positive attitude towards accessibility has a considerable indirect effect in the models 54

Public transport use

Figure 4.2 SEM diagram with unstandardised path coefficients – model structure including residential self-selection

for bus, tram and metro use (with standardised coefficients of 0.110 in Figure 4.2 and 0.142 in Figure 4.3). The effect of travel attitudes on public transport use is generally more important. Standardised coefficients of most travel attitudes are larger than 0.100. Similarly to other studies like Anable (2005) and Gardner and Abraham (2008), travel attitudes are found to be important explanatory variables. Respondents who prefer privacy, comfort and/or time savings are less likely to use public transport. Only respondents with a positive attitude towards sustainable travel options are more likely to use bus, tram and metro. The latter attitude has an indirect effect on bus, tram and metro use as well. Respondents who prefer sustainable travel options to some extent self-select themselves in an urban residential location, causing an indirect effect on bus, tram and metro use. This indirect effect is small but significant. Third, there is not only evidence of residential self-selection, as already mentioned, but also of residential determination. As the model fit of Figure 4.3 with residential determination is similar to the model fit of Figure 4.2 with residential self-selection, it seems that both processes are equally important. So, a sustainable travel attitude not only favours residing in an urban residential location (see Figure 4.2), but the reverse is also true. Urbanites might also develop a liking for sustainable travel options after experiencing living in an urban built environment for 55

Van Acker, Sandoval and Cools

Figure 4.3 SEM diagram with unstandardised path coefficients – model structure including residential determination

a while (see Figure 4.3). Moreover, urbanites are also more likely to develop a liking for saving travel time. Fourth and finally, values have a significant influence on spatial and travel attitudes and thus offer insights into the origins of these attitudes. Because of the interaction between values and attitudes, several significant indirect effects of values on public transport use exist. As could be expected, bus, tram and metro use were found to be higher among respondents with values of conformity (referring to proper behaviour and respecting rules) and lower among respondents with values of self-direction (referring to independent thinking and action) and achievement (referring to personal success). The positive indirect effect of conformity runs via a path along travel attitudes. Respondents who value proper behaviour and respecting rules are less likely to have a positive travel attitude towards privacy and comfort (probably associated with car use instead), which eventually encourages the use of bus, tram and metro. The negative indirect effects of self-direction and achievement, on the other hand, run via a path along residential attitudes and urban residential location choices. For example, self-direction is positively associated with a positive residential attitude towards social interaction, but this residential attitude does 56

Public transport use

not favour an urban residential location and also therefore not the use of bus, tram and metro. While bus, tram and metro use are discouraged by values of self-direction and achievement, the opposite holds for train use. Respondents who value self-direction and achievement are less likely to be in favour of travel time savings, which eventually encourages train use. This positive indirect effect via travel attitudes is large enough to compensate for the previously mentioned negative indirect effect of self-direction and achievement via residential attitudes. Furthermore, train use is also positively associated with the values of benevolence (referring to preserving and enhancing the welfare of people with whom one has close contacts), something that could be expected for public transport use. Rather surprisingly, the use of public transport (be it bus, tram and metro or train) was found to be negatively associated with values of universalism (referring to equality for all) and positively with values of power (referring to social status and prestige). One would expect the opposite, as has been found before in studies like Jaśkiewicz and Besta (2014). The negative indirect effect of universalism originates from two directions. First, values of universalism in this study do not encourage a positive residential attitude towards accessibility and therefore do not favour an urban residential location choice, eventually explaining the lower use of public transport. Second, values of universalism apparently do not translate themselves into sustainable travel attitudes, explaining the lower use of bus, tram and metro again. Similarly, the positive indirect effect of power originates from two directions via residential attitudes and travel attitudes. Respondents who value power are also more likely to have a positive residential attitude towards accessibility and as such reside in an urban residential location, explaining the higher use of public transport. Furthermore, values of power are negatively associated with travel attitudes of ‘pro-privacy and comfort’, which eventually discourages the use of bus, tram and metro, and positively associated with travel attitudes of travel time savings, which eventually discourages train use. Hence, the interaction between values, attitudes, and behaviour provides additional insights into different segments of public transport users. Nevertheless, although this interaction results in significant indirect effects of values and therefore is not to be neglected, other variables often have a stronger impact on public transport use. Travel attitudes were already mentioned as having a strong influence on public transport use, but the most important variable explaining public transport use is the possession of a season ticket (with standardised coefficients above 0.400 in both SEMs).

Conclusion Past studies on the determinants of public transport tend to focus on economic and structural factors. Compared to this, less is known about ‘soft’ factors like attitudes. By using data from an online survey organised in Belgium, this book chapter has provided detailed insights into the role of spatial and travel attitudes in explaining public transport use. Moreover, it has situated these attitudes not only in a choice hierarchy of residential location, vehicle ownership and activity behaviour but also in another hierarchy of values-attitudes-behaviours. Attitudes, and especially travel attitudes, were found to be important determinants of public transport use. Pro-sustainable travel attitudes encourage the use of bus, tram and metro, but positive attitudes towards privacy, comfort and time savings discourage public transport use. It indicates that public transport should not only be promoted as a sustainable way of travelling. Efforts are also needed to improve comfort, punctuality and reliability so that these negative attitudes towards public transport could be reversed. Furthermore, travel attitudes might be drivers of residential self-selection, but the reverse process of residential determination is equally important. This finding illustrates how important 57

Van Acker, Sandoval and Cools

the built environment of the residential location is for public transport use. Living in an urban residential neighbourhood with easy access to public transport not only has a direct effect (especially on train use), it also strengthens a positive attitude towards sustainability, resulting in an indirect effect (especially on the use of bus, tram and metro). However, living in an urban neighbourhood also strengthens a positive attitude towards time savings, which eventually drives people away from public transport use. Residents of urban areas might experience loss of travel time due to congestion, public transport stuck in traffic and being delayed. This finding indicates how important it is for cities to invest in dedicated bus lanes and traffic circulation systems that do not interfere with public transport. Adding attitudes but also considering the origins of attitudes in terms of personal values has proven to help in understanding the heterogeneity in travel choices made by individuals. It also offers avenues for sustainable mobility policies. For example, public transport seems to be positively associated with values of benevolence. Promoting the use of public transport as an act of caring for others might therefore be effective, especially in times of climate change. In addition, values of achievement were found to discourage public transport. Nevertheless, public transport could be promoted as a way to achieve personal goals. Although this might seem counterintuitive at first, it might be possible to link public transport use to personal goals related to health. For example, various studies pointed out how a shift from car to public transport increases health (Rissel et al., 2012; Rojas-Rueda et al., 2012; Stevenson et al., 2016). Consequently, by adding ‘soft’ factors to the models, additional opportunities for stimulating public transport use could be identified which otherwise remain unidentified. While values are considered stable, attitudes can change over time. This chapter illustrated how travel attitudes could change due to the process of residential determination. The current COVID-19 pandemic may have an accelerating effect on this. In many regions and countries, people have been encouraged to work from home, limit their travel behaviour to essential trips only and maintain social distancing. This resulted in a large reduction in mobility (Warren & Skillman, 2020) and a reduction of activity spaces (Klein et al., 2020). People might thus have experienced their residential neighbourhood in a completely different way, and, due to residential determination, they might eventually change their travel attitudes. This calls for more research on the stability of travel attitudes.

References Aditjandra, P. T., Cao, X., & Mulley, C.(2016). Exploring changes in public transport use and walking following residential relocation. A British case study. Journal of Transport and Land Use, 9(3), 77–95. Ajzen, I. (1991). The theory of planned behavior. Organizational Behavior and Human Decision Processes, 50, 179–211. Anable, J. (2005). “Complacent car addicts” or “aspiring environmentalists”? Identifying travel behavior segments using attitude theory. Transport Policy, 12, 65–78. Ben-Akiva, M. (1973). Structure of passenger travel demand models. MIT, Department of Civil Engineering. Bresson, G., Dargay, J., Madre, J. L., & Pirotte, A. (2004). Economic and structural determinants of the demand for public transport: An analysis on a panel of French urban areas using shrinkage estimators. Transportation Research Part A: Policy and Practice, 38, 269–285. Byrne, B. M. (2010). Equation modeling with AMOS: Basic concepts, applications, and programming. Taylor and Francis Group. Cao, X., Mokhtarian, P. L., & Handy, S. L. (2009). Examining the impacts of residential self-selection on travel behaviour: A focus on empirical findings. Transport Reviews, 29, 359–395. Currie, G., & Delbosc, A. (2011). Exploring the trip chaining behavior of public transport users in Melbourne. Transport Policy, 18, 204–210. De Vos, J., Ettema, D., & Witlox, F. (2018). Changing travel behaviour and attitudes following a residential relocation. Journal of Transport Geography, 73, 131–147.

58

Public transport use Ettema, D.,  & Nieuwenhuis, R. (2017). Residential self-selection and travel behaviour: What are the effects of attitudes, reasons for location choice and the built environment? Journal of Transport Geography, 59, 146–155. Ewing, R., & Cervero, R. (2010). Travel and the built environment. A meta-analysis. Journal of the American Planning Association, 76(3), 265–294. Ewing, R., Hamidi, S., & Grace, J. B. (2016). Compact development and VMT – environmental determinism, self-selection, or some of both? Environment and Planning B: Urban Analytics and City Science, 43(4), 737–755. Gardner, B., & Abraham, C. (2008). Psychological correlates of car use: A meta-analysis. Transportation Research Part F: Traffic Psychology and Behaviour, 11(4), 300–311. Hensher, D., & Reyes, A. J. (2000). Trip chaining as a barrier to the propensity to use public transport. Transportation, 27, 341–361. Homer, P. M., & Kahle, L. R. (1988). A structural equation test of the value-attitude-behavior hierarchy. Journal of Personality and Social Psychology, 54(4), 638–646. Jaśkiewicz, M., & Besta, T. (2014). Heart and mind in public transport: Analysis of motives, satisfaction and psychological correlates of public transportation usage in the Gdańsk–Sopot–Gdynia Tricity agglomeration in Poland. Transportation Research Part F: Traffic Psychology and Behaviour, 26, 92–101. Kamruzzaman, M. D., Baker, D., Washington, S., & Turrell, G. (2016). Determinants of residential dissonance: Implications for transit-oriented development in Brisbane. International Journal of Sustainable Transportation, 10(10), 960–974. Klein, B., LaRock, T., McCabe, S., Torres, L., Privitera, F., Lake, B., Kraemer, M. U. G., Brownstein, J. S., Lazer, D., Eliassi-Rad, T., Scarpino, S. V., Chinazzi, M., & Vespignani, A. (2020). Assessing changes in commuting and individual mobility in major metropolitan areas in the United States during the COVID-19 outbreak. Northeastern University. Network Science Institute. Retrieved July  8, 2020, from www. networkscienceinstitute.org/publications/assessing-changes-in-commuting-and-individual-mobilityin-major-metropolitan-areas-in-the-united-states-during-the-covid-19-outbreak Kline, R. B. (2015). Principles and practice of structural equation modeling. Guilford Press. Lin, T., Yang, D., & Guan, X. (2017). The built environment, travel attitude, and travel behavior: Residential self-selection or residential determination. Journal of Transport Geography, 65, 111–122. McNally, M. G., & Ridt, C. R. (2007). The activity-based approach. In D. A. Hensher & K. J. Button (Eds.), Handbook of transport modelling (Vol. 1, pp. 55–73). Emerald Publishing Limited. Naess, P. (2009). Residential self-selection and appropriate control variables in land use: Travel studies. Transport Reviews, 29, 293–324. Nijkamp, P., Rietveld, P., & Spierdijk, L. (2000). A meta-analytic comparison of determinants of public transport use: Methodology and application. Environment and Planning B: Planning and Design, 27, 893–903. Ory, D. T., & Mokhtarian, P. L. (2010). The impact of non-normality, sample size and estimation technique on goodness-of-fit measures in structural equation modeling: Evidence from then empirical models of travel behaviour. Quality & Quantity, 44, 427–445. Paulley, N., Balcombe, R., Mackett, R., Titheridge, H., Preston, J., Wardman, M., Shires, J., & White, P. (2006). The demand for public transport: The effects of fares, quality of service, income and car ownership. Transport Policy, 13, 295–306. Paulssen, M., Temme, D., Vij, A., & Walker, J. L. (2014). Values, attitudes and travel behavior: A hierarchical latent variable mixed logit model of travel mode choice. Transportation, 41, 873–888. Polat, C. (2012). The demand determinants for urban public transport services: A review of the literature. Journal of Applied Sciences, 12, 1211–1231. Rissel, C., Curac, N., Greenaway, M., & Bauman, A. (2012). Physical activity associated with public transport use – a review and modelling of potential benefits. International Journal of Environmental Research and Public Health, 9, 2454–2478. Rojas-Rueda, D., de Nazelle, A., Teixidó, O., & Nieuwenhuijsen, M. J. (2012). Replacing car trips by increasing bike and public transport in the greater Barcelona metropolitan area: A health impact assessment study. Environment International, 49(15), 100–109. Salomon, I., & Ben-Akiva, M. (1983). The use of the lifestyle concept in travel demand models. Environment and Planning A: Economy and Space, 15, 623–638. Schwartz, S. (2003). A proposal for measuring value orientations across nations. Retrieved July 2, 2020, from www.europeansocialsurvey.org/docs/methodology/core_ess_questionnaire/ESS_core_questionnaire_ human_values.pdf

59

Van Acker, Sandoval and Cools Schwartz, S., & Bardi, A. (2001). Value hierarchies across cultures: Taking a similarities perspective. Journal of Cross-Cultural Psychology, 32(3), 268–290. Şimşekoğlu, Ö., Nordfjærn, T., & Rundmo, T. (2015). The role of attitudes, transport priorities, and car use habits for travel mode use and intentions to use public transportation in an urban Norwegian public. Transport Policy, 42, 113–120. Stevenson, M., Thompson, J., Hérick de Sá, T., Ewing, R., Mohan, D., McClure, R., Roberts, I., Tiwari, G., Giles-Corti, B., Sun, X., Wallace, M., & Yoodcock, J. (2016). Land use, transport, and population health: Estimating the health benefits of compact cities. The Lancet, 388(10062), 2925–2935. Van Acker, V., Ho, L., Stevens, L., & Mulley, C. (2020). Quantifying the effects of childhood and previous residential experiences on the use of public transport. Journal of Transport Geography. doi:10.1016/j. jtrangeo.2020.102759 Van Acker, V., Mokhtarian, P. L., & Witlox, F. (2011). Going soft: On how subjective variables explain modal choices for leisure travel. European Journal of Transport and Infrastructure Research, 11(2), 115–146. Van Acker, V., van Wee, B.,  & Witlox, F. (2010). When transport geography meets social psychology: Toward a conceptual model of travel behaviour. Transport Reviews, 30(2), 219–240. van Lierop, D., Badami, M. G., & El-Geneidy, A. M. (2018). What influences satisfaction and loyalty in public transport? A review of the literature. Transport Reviews, 38(1), 52–72. van Wee, B., Holwerda, H., & van Baren, R. (2002). Preferences for modes, residential location and travel behaviour: The relevance for land-use impacts on mobility. European Journal of Transport and Infrastructure Research, 2(3–4), 305–316. Warren, M. S., & Skillman, S. W. (2020). Mobility changes in response to COVID-19. arXiv preprint arXiv: 2003.14228v1.Retrieved January 30, 2021 from https://arxiv.org/pdf/2003.14228.pdf

60

5 INTERCITY MODAL COMPETITION Roger Vickerman

Introduction and background The market for intercity travel has received much less attention than those for urban or regional travel markets. In part, this is because it is much more diverse and complex, ranging from travel between two adjacent towns or cities less than 20  km apart to longer-distance travel between major conurbations, including those involving international travel, that may be hundreds or thousands of kilometres apart. Understanding the demand for such travel is more difficult, as journeys are typically made less frequently than regular commuting trips that dominate urban travel markets and are more likely to involve multi-modal trips where access to, for example, major rail stations or airports becomes a key element in journey planning. Historically, the markets for intercity travel were highly regulated, with, in many countries, the state ownership of railways and airlines protected by strict licensing of, for example, longdistance coach services (see also Chapters 1, 13, 14 and 15). The growth of intercity highway networks from the 1950s followed by deregulation of airlines and coach services from the 1980s and more recently liberalisation of the market for rail services have changed patterns of provision and competition. This has come alongside social change that has made families more mobile and the growth of a long-distance market for visiting family and friends at the same time as a growth in longer-distance weekly commuting as labour markets have become more open. Against this background, this chapter explores the extent of changes in intercity travel markets over recent decades and how regulation and competition have dealt with these changes in a range of countries, including the United States, United Kingdom and other European countries. Economic pressures through an emphasis on agglomeration and social pressures through increased mobility have increased the focus on intercity transport and the associated large-scale infrastructure projects such as major highways, high-speed rail and airport development. This focus underlies most of the discussion in this chapter, as it provides the basis for much of the change in the regulatory structures that govern the various modes of transport. Some recent work has questioned the appropriate scale of such developments, suggesting that evidence in the United Kingdom points to the importance of densification within larger city regions rather than linking more distant centres (Arbabi et al., 2019)

61

Roger Vickerman

Trends in intercity travel Analysis of the US 1995 American Travel Survey by McGuckin (2013) was used as the basis of a comprehensive report into interregional travel by the Transportation Research Board (2016). This showed clearly how distance, trip purposes and travelling party composition all had significant impacts on both trip generation and mode choice. Overall, the average number of such trips per year was 7, with typically 2 to 2.5 trips for business purposes and just under 5 for other purposes. Single households were the least travelled (5.9 trips per year) and couple households the most frequent travellers (7.5 trips per year), closely followed by families with all children 5 years or older (7.3 trips). Age also played a part, with those aged 50–64 being the most frequent travellers and those aged 80 and over the least. Trip frequency also increased steadily with income. Trips by one adult were 41% for business, but those by two or more adults were 13% for business, and those with at least one child only 7%. Distance affected both trip generation and mode share. Almost 80% of all trips were between 100 and 500 miles; 11% were over 1000 miles. Up to 600 miles, the use of private vehicles dominated, though falling from around 95% for trips less than 200 miles to around 60% for those between 500 and 600 miles. Over 600 miles, the use of air increased from around 55% of those trips 600–700 miles to around 70% for those between 1000 and 1100 miles and almost 90% for those 1500 miles and more. Adults travelling alone were much more likely to use public transport (air, bus or train) than either a group of adults with no children or those travelling with children. The last group were the most likely to use a private vehicle and the least likely to use air or especially bus (see also Chapters 13, 14 and 15). Long-term analysis of the UK National Travel Survey (Department for Transport, 2018) shows that the average distance travelled per person per year rose by 46% between 1972/73 and 2018 from 4,476 miles to 6,530 miles, although there had been a 9% fall since the highest point of 7,211 miles in 2003. Over the whole period, the number of trips only rose by 3%, and the implied average trip length therefore rose from 4.7 miles to 6.6 miles, an increase of 41%. Looking at individual modes of transport, car driver trips fell by around 10% and annual mileage by around 12% between 2002 and 2018, with average trip lengths falling a small amount from 8.43 miles to 8.23 miles. Public transport bus trips outside London fell by 29% between 2002 and 2018, but mileage fell by less, with average trip lengths rising from 4.6 miles to 5.3 miles. On the other hand, rail trips increased by 64% and mileage by 41%, so average trip lengths fell from 35.8 miles to 30.9 miles. The number of trips of over 100 miles has remained fairly constant at between 6 and 8 per year since 2002, whilst those of between 50 and 100 miles have fallen slightly from 14 to 12 and those between 10 and 50 miles stayed fairly constant at between 140 and 150 trips per year. Trips of between 1 and 10 miles showed a 9% fall from 658 to 577, but those of less than 1 mile remained fairly constant at around 250. Looking only at trips of over 50 miles, around 80% of trips were made by car for journeys under 150 miles. Over 150 miles, the proportion of trips by car fell steadily to less than 50% over 350 miles; those by rail increased from around 15% of trips of 50–75 miles to around 25% of trips of between 250 and 350 miles, but then fell again, as air dominated non-car trips of over 250 miles, taking a 33% share. Whilst rail travel has shown a remarkable increase in many countries over the past two decades, this growth appears to have slowed or even reversed in the most recent years. To some extent, this mirrors a decline in commuting into large rail-served cities, although rail’s share of this market has held up or even increased. It is difficult to separate intercity travel from total non-work travel, but there are indications that off-peak rail travel has been falling in cities such 62

Intercity modal competition

as London, New York and San Francisco (Rail Safety and Standards Board, 2019). Visit Britain data suggests that this is in part due to a decline in the attractiveness of city centre attractions such as museums and art galleries, whilst visits to historic properties, farms and gardens showed an increase (Rail Safety and Standards Board, 2019). The reasons for this apparent change are varied. Large cities have become less attractive destinations, especially for families, because, it is suggested, of the fear of crime and terrorism. The previous growth is often attributed to changes in lifestyle with increased mobility and the increased likelihood of extended family members living in different urban areas. The cost of housing in major cities coupled with the availability of urban public transport has made car ownership less likely for single adults and couples without children, thus increasing the propensity for rail usage (Independent Transport Commission, 2018a; Rail Safety and Standards Board, 2019). But this change may prove to have been a shift in such relationships rather than the start of an ongoing trend. One further dimension to this is the shift toward more telecommuting. The decline in peak-period travel (Independent Transport Commission, 2018b) is indicative of this trend and is coupled with an increase in longer-distance commuting as commuters seek cheaper housing and an improved lifestyle when they don’t need to make daily commuting journeys. This affects rail market planning as season tickets, based on an assumption of daily return journeys, lose their attractiveness, although, as Mokhtarian (2003) has pointed out, there is evidence of a strong degree of complementarity between increased telecommunications and travel overall.

Regulatory frameworks Historical background Historically, intercity travel was highly regulated in most countries. In Europe, there was a general requirement to protect state-owned railways from predatory competition. In part, this was seen as necessary to protect the public service requirement to maintain little-used links that often used cross-subsidies. The development of aviation was usually promoted by state-owned flag-carrier airlines that were protected by restrictive entry barriers and serving largely state or local authority owned airports (see also Chapter 15). Longer-distance bus services or express coach services were not common in all countries. Often the operators were at least part owned by railway companies, as in the United Kingdom, and strictly regulated by a licensing system that allowed objections from potential competitors. Deregulation and the encouragement of competition started in the 1980s (Zhang et  al., 2011). The 1978 Airline Deregulation Act in the United States started the deregulation of aviation that led to the growth of the low-cost, no-frills operators (see also Chapter 15). These operators developed new business models that took away complex interlining operations and focused on the most profitable links using hub and spoke operations but often without guaranteed interchange. The deregulation of aviation followed in Europe, where the emphasis on competition of the European Single Market allowed for the development of third country and cabotage operation instead of complex bilateral agreements and the creation of a single European airspace coordinated air traffic control. These changes coincided with the changing markets for both business and leisure travel, leading to a remarkable upsurge in air travel. In the EU-27, air travel in terms of numbers of passengers grew by 36% between 2007 and 2018 (Eurostat, 2019). For those countries where a longer time series is available (such as Germany, France and the United Kingdom), we can see a more than three-fold increase in air passengers 63

Roger Vickerman

between 1993 and 2018. Movements between the United Kingdom and other EU countries increased by this same factor over this period and by all modes by 7.1% between 2008 and 2018. At the same time, the development of high-speed rail from 1981 in France and 1992 in Spain, with slower and less coordinated developments in Germany, Italy and some other countries, led to a displacement of air travel, particularly in the range of 400–600 km, by the new modes, which also gained substantial inroads in some longer-distance routes. Total passenger-kilometres by high-speed rail increased more than seven-fold from 1990, mainly though the addition of new lines in more countries, but in France, the original European high-speed network carried 234% more passengers by 2015 than in 1990. But high-speed rail was not just a competitor for air; it became an asset to airlines, as rail could be used to substitute for shorter-distance movements, acting as a feeder service and freeing up less profitable airport slots for more lucrative international and intercontinental routes. Crozet (2013) shows how at excess times over air of up to 2 hours, rail typically takes a 50% or greater share of the market, falling quite rapidly to 20% or less at a 3 hours excess time over air. Nash (2013) quotes data showing that with rail station-to-station journey times of up to about 4 hours, rail typically captures a 45% or greater market share, and up to about 2 hours and 30 minutes, the rail share is typically 80% or greater.

Bus and express coach markets The intercity bus or coach market has been one of the more interesting developments in recent years (see also Chapter 13). This market was very heavily regulated in most countries, such that there was virtually no such market in some, such as France (Blayac & Bougette, 2017). In Europe, the United Kingdom was the first to deregulate the market in 1980 (White & Robbins, 2012). There was already a significant network of long-distance coach services, mainly operated under the name National Express, owned by the nationalised National Bus Company (NBC) and operated by its subsidiary companies with standardised liveries and national marketing. This network was in turn inherited from a network operated through complex joint operations of the regional bus companies that had been brought together in 1968 to form NBC. This covered all of England and Wales, but a similar network was operated by the nationalised Scottish Bus Group in Scotland, which also operated Anglo-Scottish services in collaboration with NBC. All of these services operated under a licensing system that had originally been introduced in 1930. This served in part to protect the railways from direct competition, as they could object to proposals to run parallel services. Generally, coach services were cheaper but slower than rail services and therefore effectively differentiated the market, except where coach services were provided directly on routes that were not served by direct trains. This was particularly the case for summer services to seaside resorts. The 1950s marked the heyday for this type of traffic before the increase in car ownership removed large parts of the market. The big change was the development of the motorway network during the 1960s that allowed the acceleration of journey times and increased the potential competition for rail in journeys between large cities such as London to Birmingham. The 1980 deregulation changed all that. Licensing of specific intercity routes was abolished, and operators were free to run services with only a simple advance notification, although quality control through operator licensing was enhanced. Other than an operator licence, there are few barriers to entry into the express coach market, so the expectation was of a big increase in competition. The main problem new entrants often faced was access to bus stations that were often controlled either by local authorities or incumbent bus operators. For example, the main Victoria coach station in London was owned by National Express until 1988, when it was taken over by London Transport. National Express continued to operate services, mainly on a 64

Intercity modal competition

franchised basis for individual services, but serving as a national marketing agency to maintain the idea of a national network. The lack of this network of booking agencies in the pre-internet era also hampered the growth of independent operators. Very quickly, the era of competition was replaced by consolidation as National Express asserted its dominance, though, interestingly, one of the new entrants, Stagecoach, rapidly became a major group for local bus operation following the deregulation and privatisation of NBC in 1986. Stagecoach also re-emerged as a coach operator with its Megabus operation focusing on a specific segment of the market, students, using online booking and providing direct services to university campuses. Some specific routes also saw the development of high-frequency coach services not provided by National Express. These developed in particular where rail service was relatively poor, such as between London and Oxford, where two companies provided competing services running several times an hour and offering on-board services such as refreshments and Wi-Fi. However, improvements to the rail service making it more reliable and competitive and increasing road congestion that caused unreliability in coach operations led to one of the operators leaving the market in 2019. Competition between the two coach operators has, however, continued in direct services to airports, where rail connections are poorer. A similar pattern emerged in Germany after deregulation between 2009 and 2013 (Dürr & Hüschelrath, 2017). A number of new operators entered a market that had previously largely been controlled by a bus-operating subsidiary of Deutsche Bahn (DB), the state-owned rail operator. Unlike in the United Kingdom, a number of these new operators were foreign owned, including some of the post-privatisation UK bus operating groups (including Stagecoach’s Megabus), although DB also participated in the form of ICBus. Gradually, however, the market has consolidated with one operator, Flixbus, dominating after the takeover of Megabus and merger with the early leader MeinFernbus. Flixbus has also developed international routes, building on the success of the joint venture Eurolines, and entered the French and US markets. The market has continued to show some instability and is now threatened by a German government plan to reduce value-added tax (VAT) on rail tickets, designed to promote the environmental benefits of rail, which may lead to cuts to more marginal services. The French market was much less well developed, and rail also offered a lower-priced and less speedy alternative to the full-service train à grande vitesse over key routes through its lowcost operation called Ouigo, which focused on the market often taken by long-distance coach. However, there has been a gradual introduction of competing long-distance coach services since deregulation in 2015 (Blayac & Bougette, 2017). Interestingly one of the main entrants was the national rail operator SNCF, using a low-risk franchised model under the name Ouibus competing with the German operator Flixbus and the largely state-owned Transdev operating as Isilines alongside its existing international operations Eurolines. These three operators control over 85% of the French market, but the Transdev operators have by far the largest share. The situation in the United States has been rather different (Augustin et  al., 2014). The generally poorer quality of long-distance passenger rail services away from a few key corridors such as the Northeast Corridor between Washington and Boston, and the earlier completion of a national interstate highway network, led to a well-developed coach network mainly provided by two operators, Greyhound and Trailways. This was regulated in a similar way to that in the United Kingdom prior to 1980, requiring operators to prove that any new service was not detrimental to existing services and to avoid predatory pricing. Deregulation, or reregulation as Augustin et al. (2014) call it, occurred in the Bus Regulatory Reform Act of 1982. This allowed for freedom of entry and exit and greater freedom on fares but retained the right of existing operators to object to new services. Early price wars destabilised the incumbents such as Greyhound and Trailways as new entrants were attracted. The US coach market had also suffered 65

Roger Vickerman

from the rise of the low-cost airlines competing for the low-income students and older persons cohort. The perceived attractiveness of this market led to attempts at entry by UK-based groups such as Stagecoach (through its Megabus subsidiary) and FirstGroup (which had a 50% stake in BoltBus). But these proved to be difficult to sustain, and the US operations of Stagecoach were sold in 2019, although the Megabus brand was retained. Hence the pattern of initial aggressive competition followed by consolidation and the emergence of a single dominant operator seems to have occurred in most countries.

Intercity rail markets The development of the rail market in Europe has been dominated by the European Union’s various railway packages and a move towards privatisation and competition (Nash, 2011, 2015; Preston & Robins, 2013). The EU policies have had two main strands in the passenger market, vertical unbundling and the introduction of competition (European Commission, 2016). Vertical unbundling involves the separation of track management and the provision of services. This may involve separation of ownership, as in the United Kingdom, or simply a clear accounting separation with the objective of providing equal terms of access to any operator removing the presumption of privileged access to an incumbent operator (Cantos et al., 2010). Competition can either be for the market though a franchising system, typically involving a group of regionally coherent services, or in the market through the right of open access (Nash, 2011; Competition and Markets Authority, 2016). Competition was initially seen as the key to allowing rail to compete against airlines in the international market through a requirement for interoperability. This has been achieved through various joint ventures between the national rail companies such as Eurostar or Thalys, creating direct services between major cities over routes that favour rail over air such as those in the Paris, Brussels, Cologne, Amsterdam and London region of northwest Europe. These joint ventures have gradually been consolidated into a single entity controlled by SNCF, the French railway operator. There are similar joint operations in transalpine routes. Only in the United Kingdom has there been a wholesale privatisation of the entire passenger service. In other countries such as Germany, regional services have been franchised, and in France, such services have been placed under the control of regional authorities. But in the United Kingdom, all services, both long-distance intercity services and regional services, have been franchised since 1996. The franchise system has been revised several times since 1996 (Nash, 2015) and is again under review. This has mainly affected such elements as contract length and the allocation of revenue risk between the government and the operator. The main intercity routes have proved the most susceptible to contractual problems, including default, and one, the East Coast Main Line between London, Yorkshire, the Northeast and Edinburgh, has had to be taken back into public ownership on more than one occasion, most recently in 2018 (British Broadcasting Corporation, 2018). All franchises have tended to be subject to the winner’s curse problem where overoptimistic revenue forecasts have led to excessive bids that have not allowed for potential changes in the macroeconomic environment. Intercity rail demand has proved particularly difficult to forecast, and although a basic relationship with gross domestic product (GDP) growth was the main driver in the past, this relationship has also become less stable in recent years, and the most recent franchise awarded has attempted to allow for risk sharing on this basis (Department for Transport, 2019). Interestingly the UK franchises have encouraged international participation, including stateowned rail operators from other EU countries. Although open access is allowed for, there has been relatively little open access entry; what has occurred has focused on the provision of 66

Intercity modal competition

direct services to destinations (typically middle-sized towns and cities) off the mainline routes rather than direct competition for the core markets. With minor exceptions, where franchises compete over historic alternative routes, there has been little attempt to provide differentiated services for segmented markets using secondary routes or terminals. The United Kingdom has been slower than other countries in developing new high-speed rail routes (Givoni, 2006). In part, this has been due to the relatively higher speeds achieved on conventional routes, and these have been successful in helping the renaissance of rail without the levels of investment seen, for example, in France or Spain (Chen  & Hall, 2011). In the case of the one relatively short dedicated high-speed line between London and the Channel Tunnel, HS1, the international services are totally segregated from the domestic ones, and the regional services that use the line are integrated into the regional franchise, so there is no real competition. The current plans for the proposed HS2 network between London, Birmingham, Manchester and Leeds also foresee the high-speed services as being integrated in the relevant franchise. The announcement for this new franchise for the West Coast main line makes provision for the incumbent operator to assume responsibility for HS2 services in due course (Department for Transport, 2019). The rationale for this is in part the importance of integrating all long-distance services so that cities away from the new line can benefit from direct services using the line. It also avoids the potential problems of traffic, and hence revenue, abstraction from an existing franchise holder. The provision of direct services off a high-speed line has been seen as important in growing markets in France (Bonnafous, 1987; Vickerman, 1997). Only in Italy has there been open access use of new high-speed lines with a private operator, NTV, competing with the state-owned operator, Trenitalia, over core routes, albeit using secondary terminals in some cities (Croccolo, 2013). Whilst some countries such as Sweden have encouraged competition for the incumbent operator (SJ in Sweden) over longer distance routes, others, such as Germany, have kept the main intercity network for the national operator (there DB), using franchising for regional or secondary routes. Germany is demonstrating, however, how a national rail operator can be used as an instrument of wider public policy by reducing the VAT rate on train tickets from 19% to 7% from January 2020 as a means of lowering the price of rail travel relative to other modes to promote environmental policies (BBC, 2019). This is a relatively rare example of trying to promote intermodal competition; most regulatory and competition policies have been intrarather than intermodal. Intercity rail travel in the United States, except for some long-distance commuting lines around major metropolitan areas, is now relatively undeveloped compared to the European situation, in contrast to the important role of the railways in the early development of the United States (Fogel, 1964). This is in part due to geography, with much greater distances and relatively lower population densities such that air travel has a greater competitive advantage than, for example, in Europe. Higher levels of car ownership and the development of the interstate highway network from the 1950s onwards (Altshuler & Luberoff, 2003, Chapter 4; Friedlander, 1965) also provided strong competition for railroads over shorter distances. But it also reflects the greater dominance of freight traffic on US railroads, with freight operators owning and having priority use of tracks. It also reflects the earlier moves to remove subsidy in the interests of creating a level playing field for transport, although Amtrak was established as a governmentowned operator of longer-distance passenger services (Winston, 2006). Only in the northeast corridor between Washington and Boston does the geography lend itself to a reasonably frequent intercity rail service that can compete with air, and although speeds have been increased, this is not high-speed rail under the normal definition of segregated track capable of speeds of 250 km/h or greater. There are similar linear networks in Florida and California, both of which 67

Roger Vickerman

have developed plans for high-speed rail, although these remain controversial and have been slow to come to fruition (Perez Henriquez & Deakin, 2017).

Competition and regulation in intercity transport markets: a synthesis In some respects, intercity transport markets demonstrate the greatest potential for competition between modes of all transport markets. Over shorter distances of up to around 200 km, car, bus and rail can all provide reasonably competitive offerings. The balance will shift according to such factors as the ease of access to rail stations or the existence of uncongested highways offering bus and car faster journey times. Over longer distances of up to 600 km (or more), rail (especially high-speed rail) and air will be the main competitors. The level of interaction between city-pairs depends on city size and intervening distance, and that will determine the likely density of service offered and the number of viable competitors that can be supported. The traditional development of transport services, often with the protection of some degree of regulation, allowed operators to cross-subsidise between more and less profitable routes. This was justified on the basis of maintaining at least a minimum level of service to all communities. In some cases, this may also require some form of public subsidy justified under a public service obligation (Ponti, 2011; Quinet & Vickerman, 2004, pp. 181–184). This goes back to the early days of rail, where early private rail companies in the United Kingdom were required, as part of being granted the rights to develop a rail route, to maintain a minimum level of service often referred to as ‘the Parliamentary train’. As noted earlier, regulation of bus services was often prompted by the need to maintain this basic rail service level. The first real challenge to this system came in the United Kingdom with the publication of the Beeching Report into British Railways in 1963 (British Transport Commission, 1963). Growing deficits prompted the government of the day to question the public service obligation ethos and treat the railways as a business, of which the less profitable or loss-making parts should be closed. Rail was also seen as an old-fashioned mode likely to be replaced by road over shorter distances and air over longer distances and therefore not something to be a priority for investment. This was similar to the situation in the United States but contrasts with the decisions made first in Japan and then in France to invest heavily in new forms of rail using new dedicated tracks and allowing for higher speeds. It could be argued that these were in fact completely new modes of transport replacing traditional rail rather than just an investment in rail, but the level of integration with existing rail networks was important in determining the success of high-speed rail. This coincided with the move to deregulate the key competing modes of bus and air, removing the protection of the railways from, in some cases, any competition. Allowing free competition from modes that have lower barriers to entry than rail, as they do not require dedicated infrastructure, leads to cherry-picking of the most profitable routes. The first effect is one of aggressive price or service-level competition between rival operators within each mode. This leads to the emergence of a dominant operator within that market after which prices rise and service levels fall. This battle to become the dominant modal operator has consequences for the other modes in the market, which also see profit margins squeezed, affecting their ability to invest to compete. The question here is the extent to which transport markets are genuinely contestable, the criterion for effective competition (Gagnepain et al., 2011; Small & Verhoef, 2007, pp. 205–208). From the perspective of public authorities and regulatory agencies, the focus is normally on the situation within each mode, ensuring free and fair competition between operators, but this 68

Intercity modal competition

can have unintended consequences on other modes in the market. Often in city regions, this effect is minimised by the creation of an overall transport authority that seeks to avoid wasteful competition by allowing competing operators to compete for the market by, for example, a franchising system rather than within the market. This allows for the development of an integrated multi-modal transport network and a comprehensive multi-modal ticketing system. The multiplicity of public authorities often makes this more difficult in the intercity context. Whilst some wider regional systems do exist, such as the various ‘Tarifverbund’ in Germany, these are more concerned with integrating the smaller towns in the hinterland of a major city rather than the links between major cities. The creation of smart tickets such as the Oyster card in London allows for a similar integrated ticketing across a wide area that increasingly includes journeys to towns in the employment catchment area as well as services wholly within the metropolitan region. There are very few examples of nationwide ticketing systems that integrate both urban and interurban trips, one such being in the Netherlands, where the OV-chipkaart can be used on all public transport. In the United Kingdom, the ‘plus bus’ ticket allows rail travellers to purchase an add-on that can be used to complete a journey by bus in selected towns and cities. Ultimately, however, intercity transport has not been subject to the same degree of integration as in urban areas. The desire to promote competition within modes and reduce the cost of public subsidy has led to a piecemeal approach in most countries. It is only in cases where there is an overwhelming need to integrate services, due mainly to geography (Switzerland and Norway being obvious examples), that there is an attempt to provide a national transport service using the most appropriate mode for each link on a network. Attempts to coordinate timetables have, for example, been seen as anticompetitive by the Competition and Markets Authority in the United Kingdom. But, as argued previously, this reflects the fact that much less is known about the pattern of demand for intercity travel, making it more difficult both to forecast future demands and understand the potential for substitution and complementarity between modes. That intercity journeys are more likely to involve a degree of intermodality increases the problem of designing appropriate regulation and overall policy. Two areas for further research are identified as improving the coverage of such trips in regular national travel surveys and deriving better information on the cross-elasticities of demand between modes. Traditional diary methods are an inefficient way of collecting detailed data on less frequent longer-distance trips. Harnessing big data sources such as mobile phone records is one way of tracking mobility that is being experimented with (see also Chapter 24). Evidence on responsiveness to travel costs usually comes additionally from expenditure data and ticket revenues. These are less effective in measuring elasticities when journeys involve multiple modes which are charged for separately or when separate services are bundled and sold through integrating agencies. Simple cross-elasticities do not measure the complementarity between modes, which is essential to ensure that regulatory practices keep pace with the changing markets for travel. Furthermore, the growth of online selling with variable fares under yield management makes linking expenditure and actual travel more difficult. Much remains to be done in really understanding this particular aspect of transport. COVID-19 has had a major impact on intercity travel markets and thrown into sharp relief some of the underlying issues in these markets. The lockdowns imposed in many countries came with, if not an outright ban on travel, an exhortation to avoid all but essential travel, to work from home where possible and to conduct business by means of electronic conferencing. The closure of hotels and recreational facilities virtually stopped all business and leisure travel. The imposition of social distancing rules also reduced effective capacity in public transport vehicles to less than 20% of normal. Ridership fell to around 5% of normal. Rail timetables were drastically reduced, and many interurban bus services were suspended. Entire fleets of aircraft 69

Roger Vickerman

were grounded and international rail services cancelled or drastically reduced as borders were closed to all but essential travel. Most governments offered some financial support to transport operators; in the United Kingdom, rail franchises were suspended and replaced by management contracts that compensated operators for services run, and bus operators were given support grants that compensated on the basis of a fixed sum per kilometre actually run. As life gradually returned to something approaching normality at the end of the strictest period of lockdown, public transport services faced continuing problems. Social distancing meant that effective capacities on trains and buses remained limited, and most countries mandated the wearing of face masks. Some governments even maintained an official position that public transport should be avoided where possible, leading to worrying rises in car commuting. Whilst financial support has in most cases remained in place, transport, and especially less essential intercity transport, is competing with many other claims on public budgets. It seems likely that any return to the status quo ante will take a long time for transport, probably two to three years, if not longer. It seems unlikely that a sticking plaster solution of the type of financial support used in the initial stages to preserve some basic public transport service can be maintained over such a period. It is more likely that private-sector operators will continue the retrenchment that had already been seen prior to the pandemic. This suggests that the public sector will need to make a wholesale reappraisal of the way public transport is provided, and that could lead to the end of the deregulated competitive model that had become the norm in many countries. The emergency occasioned by the pandemic could be a catalyst for much more fundamental changes. Understanding the demand for intercity transport in this changed situation will become even more essential.

References Altshuler, A.,  & Luberoff, D. (2003). Mega projects: The changing politics of urban public investments. The Brookings Institution. Arbabi, H., Mayfield, M., & McCann, P. (2019). On the development logic of city-regions: Inter- versus intra-city mobility in England and Wales. Spatial Economic Analysis, 14, 301–320. Augustin, K., Gerike, R., Martinez Sanchez, M. J., & Ayala, C. (2014). Analysis of intercity bus markets on long distances in an established and a young market: The example of the US and Germany. Research in Transportation Economics, 48, 245–254. Blayac, T., & Bougette, P. (2017). Should I go by bus? The liberalization of the long-distance bus industry in France. Transport Policy, 56, 50–62. Bonnafous, A. (1987). The regional impact of the TGV. Transportation, 14, 127–137. British Broadcasting Corporation. (2018). East Coast train line to be put into public control. Retrieved January 24, 2020, from www.bbc.co.uk/news/business-44142258 British Broadcasting Corporation. (2019). Germany plans €54bn climate deal amid 500 protests. Retrieved January 24, 2020, from www.bbc.co.uk/news/world-europe-49767649 British Transport Commission. (1963). The reshaping of British railways. The Stationery Office. Cantos, P., Pastor, J. M., & Serrano, L. (2010). Vertical and horizontal separation in the European railway industry and its effects on productivity. Journal of Transport Economics and Policy, 44, 139–160. Chen, C. L., & Hall, P. (2011). The impacts of high-speed trains on British economic geography: A study of the UK’s intercity 125/225 and its effects. Journal of Transport Geography, 19, 689–704. Competition and Markets Authority. (2016). Competition in passenger rail services in the UK: A policy document. Retrieved January 24, 2020, from https://assets.publishing.service.gov.uk/media/56ddc41aed915d0 37600000d/Competition_in_passenger_rail_services_in_Great_Britain.pdf Croccolo, F. (2013). New entry in the Italian high-speed market (Discussion Paper 2013-29). International Transport Forum, OECD. Crozet, Y. (2013). Performance in France: From appraisal methodologies to ex-post evaluation (Discussion Paper 2013-26). International Transport Forum, OECD Publishing.

70

Intercity modal competition Department for Transport. (2018). National travel survey, England 2018. Department for Transport. Retrieved January 24, 2020, from https://assets.publishing.service.gov.uk/government/uploads/sys tem/uploads/attachment_data/file/823068/national-travel-survey-2018.pdf Department for Transport. (2019). West Coast marks new partnership model for rail. Retrieved January 24, 2020, from www.gov.uk/government/news/west-coast-marks-new-partnership-model-for-rail Dürr, N. S., & Hüschelrath, K. (2017). Patterns of entry and exit in the deregulated German interurban bus industry. Transport Policy, 59, 196–208. European Commission. (2016). Fourth railway package. Retrieved January 24, 2020, from https://ec.europa. eu/transport/modes/rail/packages/2013_en Eurostat. (2019). Air passenger transport statistics. Retrieved January 24, 2020, from https://appsso.eurostat. ec.europa.eu/nui/submitViewTableAction.do(avia-paoc) Fogel, R. M. (1964). Railroads and American economic growth: Essays in economic history. Johns Hopkins University Press. Friedlander, A. F. (1965). The interstate highway system: A study in public investment. North Holland. Gagnepain, P., Ivaldi, M., & Muller-Vibes, C. (2011). Industrial organization of competition in local bus services. In A. de Palma, R. Lindsey, E. Quinet, & R. Vickerman (Eds.), A handbook of transport economics (pp. 744–762). Edward Elgar Publishing. Givoni, M. (2006). Development and impact of the modern high-speed train: A review. Transport Reviews, 26, 593–611. Independent Transport Commission. (2018a). Wider factors affecting the long-term growth in rail travel. Independent Transport Commission. Independent Transport Commission. (2018b). What is the contribution of peak and off-peak travel to the urban economy. Independent Transport Commission. McGuckin, N. (2013). Intercity travel market analysis. Committee for a Study of Intercity Passenger Travel Issues and Opportunities. Retrieved January 24, 2020, from https://travelbehavior.us/documents/65.pdf Mokhtarian, P. L. (2003). Telecommunications and travel: The case for complementarity. Journal of Industrial Ecology, 6, 43–57. Nash, C. (2011). Competition and regulation in rail transport. In A. de Palma, R. Lindsey, E. Quinet, & R. Vickerman (Eds.), A handbook of transport economics (pp. 763–778). Edward Elgar Publishing. Nash, C. (2013). When to invest in high-speed rail (Discussion Paper 2013-25). International Transport Forum. OECD. Nash, C. (2015). Rail. In C. Nash (Ed.), Handbook of research methods and applications in transport economics and policy (pp. 359–370). Edward Elgar Publishing. Perez Henriquez, B. F., & Deakin, E. (2017). Institutional evolution and the politics of planning HSR in California. In B. F. Perez Henriquez & E. Deakin (Eds.), High-speed rail and sustainability: Decisionmaking and the political economy of investment. Routledge. Ponti, M. (2011). Competition, regulation and public service obligations. In A. de Palma, R. Lindsey, E. Quinet,  & R. Vickerman (Eds.), A handbook of transport economics (pp.  661–683). Edward Elgar Publishing. Preston, J., & Robins, D. (2013). Evaluating the long-term impacts of transport policy: The case of passenger rail privatisation. Research in Transportation Economics, 39, 14–20. Quinet, E., & Vickerman, R. (2004). Principles of transport economics. Edward Elgar Publishing. Rail Safety and Standards Board. (2019). Understanding the drivers that impact travel behaviour. Rail Safety and Standards Board. Small, K. A., & Verhoef, E. (2007). The economics of urban transportation. Routledge. Transportation Research Board. (2016). Interregional travel: A new perspective for policy-making. Transportation Research Board Special Report 320, TRB. Vickerman, R. (1997). High-speed rail in Europe: Experience and issues for future development. Annals of Regional Science, 31, 21–38. White, P., & Robbins, D. (2012). Long-term development of express coach services in Britain. Research in Transportation Economics, 36, 30–38. Winston, C. (2006). The US: Private and deregulated. In J. Gomez-Ibanez & G. de Rus (Eds.), Competition in the railway industry (pp. 135–152). Edward Elgar Publishing. Zhang, A., Zhang, Y., & Clougherty, J. A. (2011). Competition and regulation in air transport. In A. de Palma, R. Lindsey, E. Quinet, & R. Vickerman (Eds.), A handbook of transport economics (pp. 797–821). Edward Elgar Publishing.

71

6 PUBLIC TRANSPORT INTEGRATION Lucy Budd and Stephen Ison

Introduction In the lexicon of daily conversation, ‘integration’ means to bring disparate ideas or components together to form a co-ordinated, functional, and unified whole to enable systems or processes to work more efficiently and effectively to the benefit of their users. Within the context of urban transport policy and land use planning, ‘integrated transport’ has long been promoted as a means of achieving sustainable travel outcomes and reducing reliance on the private car. However, while the goals of delivering economically viable public transport that is of a reliable quality, meeting the needs of a diverse portfolio of users, and achieving widespread public acceptance are strategically important, the challenges to achieving them are considerable. Facilitating truly ‘integrated’ transport thus remains an aspiration rather than a reality for many places and cities around the world. This chapter begins by examining the concept of transport integration. The challenges of planning and delivering it are then discussed before specific vignettes of schemes which have delivered an integrated transport system are presented. Examples of where innovations in public transport technologies have facilitated greater integration are provided before the conclusions are presented.

The concept of transport integration Transport integration involves co-locating different modes of public and private transport (including road, rail, air, maritime, and active transport infrastructure and vehicles) at dedicated or distributed interchanges to create accessible, affordable, reliable, sustainable, convenient, comfortable, safe, and seamless journeys for passengers (see Campaign for Better Transport, 2018; Nielsen et al., 2005) (see also Chapter 7). The provision of attractive, convenient, and safe interchanges not only influences human travel behaviour but can also have a positive effect on the public realm and built environment by making transport hubs attractive places in which to meet, eat, conduct business, and invest. Depending on their location, capacity, and design features, transport interchanges can facilitate multi-modal journeys and provide temporally efficient connections to a wide number of spatially dispersed destinations. This reduces the interchange penalty and the generalised cost of a journey by public transport, thereby making them more convenient and attractive propositions. 72

Public transport integration

Integrated transport that is convenient, reliable, and attractive to users is essential for incentivising modal shift away from private cars, reducing carbon emissions, and improving local air quality. However, transport infrastructure has often been constructed in an ad hoc manner over time with little by way of co-ordinated planning or control and an emphasis on the needs of individual transport companies rather than the end users. There is now a recognition that the requirements for integrated public transport must be viewed from the perspective of the user. As a general rule, passengers require: • • • •

• • •

The ability to travel to where they want to go at a convenient time and with a trip duration comparable (or ideally preferable) to that which could be achieved by private transport; Convenient and straightforward connections where a change of transport mode is required; Accurate, accessible, and reliable real-time information; Provision of an intuitive single (and preferably universal) end-user payment system involving contactless technology that is identical on every vehicle and mode, regardless of vehicle type or operator (in some jurisdictions, individual providers have separate end-user payment systems with little if any overlap); Safe and comfortable facilities and vehicles; High service frequency and regularity, including ‘turn up and go’ provision on highdemand routes; Assistance in the event of delay or disruption and the ability to take an alternative mode of transport to the intended destination without penalty, if required.

The components of an integrated transport system thus include not only spatial characteristics (the ability to access a wide range of destinations and enjoy convenient interchanges that involve minimal walking) and temporal attributes (convenient, reliable, and quick services) but also practical considerations concerning the provision of facilities at interchanges including seating areas, toilets, step-free access, travel information screens and help desks, and a single (and preferably universal) end-user payment system. Best practice suggests that this needs to be led and co-ordinated by a single agency which is responsible for developing public transport policy, planning system extensions and enhancements, devising pricing and fare structures, and operating and overseeing the system.

Achieving transport integration There are arguably four prerequisites to achieving integrated transport systems (based on May et  al., 2006): integrated policy, integrated planning, integrated infrastructure, and integrated operations. 1

2

Integrated policy – Transport policy is often developed at a variety of levels: national, regional, and local (see also Chapter 2). The three often have competing needs and priorities and thus do not always fully align. The involvement of politicians and governments, who often have short tenures, means that the creation of long-term transport policy can be problematic, and cost pressures mean that pragmatism rather than long-term progress is often prioritised. Integrated planning – The second challenge is ensuring that the multifarious public and private actors who are involved in planning and delivering transport networks and services co-ordinate their endeavours and engage in joint working to ensure new infrastructure and services are developed in an integrated manner for the benefit of end users. Co-ordinating 73

Lucy Budd and Stephen Ison

planning in this way should help to ensure that all the different transport modes readily connect (both spatially, temporally, and practically) at interchanges to minimise transfer times and provide a safe and positive customer experience. Although planning is often considered a long-term exercise, planning is also required for short-term or one-off events that will lead to a surge in demand (or a significant change in the normal pattern of demand) for public transport. Major sporting events, cultural and religious festivals, national celebrations, and political marches would fall into this category, and all may result in a sudden upturn in demand (Currie  & Shalaby, 2012). Co-ordinated planning between transport providers and network controllers is thus required to ensure adequate capacity (which may involve the use of additional or larger vehicles as well as the introduction of extra services) is provided. 3 Integrated infrastructure – Individual transport modes require dedicated infrastructure, such as tracks, signalling, roadways, and ramps, and, for reasons of public safety, security, and operational expediency, often need to be physically segregated from one another. At the same time, however, there is a need to ensure that passengers can change from one mode of transport to another in a safe, convenient, and seamless way that minimises delays to their journey. Public transport interchanges need to ensure safe and secure seamless physical connections between pavements, cycleways, trams and railway platforms, bus stops, car parks, jetties, and airports. The infrastructure also needs to extend to public facilities, principally toilets and waiting areas, as well as retail concessions.   It is important to note the different types of transport infrastructure integration and how it varies by transport mode and geographic location. National railways, for example, often exhibit a type of vertical integration in which the track, signals, station, and rolling stock are all owned and operated by one provider. This system emerged because, historically, railways developed as integrated firms which owned both the track and the trains and which had regional or national monopolies. Since the 1990s, some countries have privatised the provision of train services and created a complex system of franchising and track access agreements. In the case of the United Kingdom, maintenance of the track and lineside infrastructure is provided by one agency – Network Rail (which in turn outsources many maintenance functions to third-party contractors) – while the operation of stations and rolling stock is provided by a range of private train operating companies (TOCs) that have been awarded the franchise to operate passenger services on a particular part of the network. This separation of track from trains was designed to improve the productivity of the railway network and reverse the decline in patronage. It does, however, make the provision of integrated operations (see subsequently) arguably more challenging and complex, as more stakeholders are involved in the production of public services. 4 Integrated operations – As well as integrating the built infrastructure of transportation, transport services need to be co-ordinated to ensure seamless connections between services and between modes. The challenge is that waiting times are often perceived to be two to three times longer than they actually are, and so ‘turn up and go’ services, which reduce reliance on a timetable, are a good option on high-demand routes such as urban metro networks. The Moscow Metro, for example, can provide a service once every 95 seconds, while the Victoria Line on the London Underground offers a peak-time service frequency of a train every 100 seconds. These routine high-frequency services help to further reduce the interchange penalty. Crucially, all the different modes (and different lines within a single network) must be co-ordinated and complement each other in terms of service frequency, capacity, and customer service in order to prevent congestion and avoid potentially damaging and disruptive competition. 74

Public transport integration

  Service and timetable co-ordination, particularly where multiple operators are using the same infrastructure (for example, multiple train operating companies operating from the same railway stations and platforms), is essential. Integrated ticketing and fares are also key to delivering a seamless connection between services and modes. It is important that no additional cost be levied for transferring between modes. Integrated operations must also extend to the provision of travel information and customer support. Real-time service information and the provision of customer service agents to support passengers who are unfamiliar with the network or who are experiencing disruption are also needed to help users navigate the network and make informed decisions about their travel options,   Integrated operations also extend to the oversight and supervision of the network. Here, close control and co-ordination between drivers, customer service assistants, signalling centres, area traffic control, the police, and the security services is key to keeping the transport system safe and secure.

The challenges of, and opportunities for, public transport integration So far, this chapter has explored the concept of public transport integration and identified its key components. The reality, however, is that achieving true integration is fraught with practical and political difficulties, some of which are a legacy of past planning decisions and modes of operation. Examples from the United Kingdom are presented later in this chapter by way of illustration. May et al. (2006) detail the principles of integration, albeit in terms of urban transport strategy, identifying the barriers to integration as legal and institutional, financial, political and cultural, practical, and technological. Historically, private individuals and enterprises were responsible for building much of the early public transport infrastructure. Over the last 150 years or so, the network in the United Kingdom has variously been in private and public hands. As these owners have different priorities and financial resources, the pattern of service provision that has emerged is highly complex, and this has important implications for public transport integration. This point is illustrated with reference to the development of the UK railway and London underground network.

UK mainline rail/bus integration The Victorian-era mainline railway network was financed and constructed by private companies. These firms surveyed the routes; purchased the land; constructed the track, station, and communications infrastructure; and operated the trains. Frequently these companies faced opposition from wealthy landowners who did not want new railway tracks constructed on or near their land. As a result, the railway companies often had to compromise not only on the route their tracks took but also on the location of their stations and termini. For example, London’s major railway stations – Kings Cross, St Pancras, Euston, Marylebone, Paddington, Waterloo, Victoria, and Liverpool Street – each of which initially served a different company and route network, had to be constructed on the periphery of the old city, as the railways were not permitted to penetrate its core. Beyond London, too, the independent railway companies built their own station facilities with the result that passenger rail services were fragmented between different sites. Over time, the railway companies bought up competitors to increase their market share and to take advantage of regional or national monopolies. Although this process of consolidation naturally led to the closure of some lines and stations, the legacy of multiple stations remained, and examples include, in the United Kingdom’s second-largest city of Birmingham, the three 75

Lucy Budd and Stephen Ison

mainline stations of Birmingham New Street, Moor Street, and Snow Hill. Outside the railway stations, private companies, who do not necessarily share the same payment systems or ticket types as the TOCs, provide local and regional bus services. Although customers booking rail tickets online have been offered a ‘PlusBus’ option since 2002, this is not available at every railway station, and coverage of the scheme is patchy. PlusBus is a discount price travelcard for unlimited bus and tram travel around town at the start or end of a journey on the UK national rail network. From an initial offering of 35 cities, over 290 are now in the scheme. The most notable exception is London (see later in this chapter), a city which has its own distinctive transport geography and forms of transport integration and governance. Between 1923, when the railway companies were rationalised, and nationalisation in 1947, the ‘Big Four’ (as they were known) private railway companies – the Great Western Railway (GWR); the London, Midland and Scottish Railway (LMS); the London and North Eastern Railway (LNER); and the Southern Railway (SR) – dominated British railway operations. The creation of the new national British Railways in 1948 brought all track, stations, and rolling stock under state control. This model existed until 1994, when British Rail (as it was then known) was privatised over a period of 3 years. Responsibility for the track and lineside infrastructure initially passed to the private company Railtrack (before subsequently being bought back under public control as Network Rail in 2002 following a series of high-profile accidents), while responsibility for operating the trains passed to private companies (see also Chapter 14). The process of UK railway privatisation was completed in 1997 (Harris  & Godward, 1997; Shaw, 2000; Preston, 2017). As of early 2020, 16 separate franchises and 5 open access operators provided passenger train services in the country. Each operator pays access charges (which are regulated by the UK government’s centralised Office for Road and Rail) to Network Rail for using the infrastructure. The track access charge is levied per vehicle mile travelled and varies according to the type of locomotive or multiple unit that is being used. The passenger train operating companies, in turn, lease much of their rolling stock from third-party leasing companies, including Porterbrook and Angel Trains. As the TOCs are not generally able to make any interior or structural changes to the vehicles they lease, they cannot adapt the vehicles to local needs (for example, by removing kitchen areas from catering cars to increase the seating capacity of their services). For the average passenger, however, such administrative and managerial complexities are irrelevant. Passengers simply want a safe, reliable, and cost-effective service that meets their travel needs. The idiosyncrasies of service provision and rail franchising are not important and only manifest themselves in the event of service disruption, when passengers need to know which TOC they were travelling within in order to claim compensation for any service delays they experienced. The fragmentation of service provision and service delivery is not simply a UK phenomenon. In the San Francisco Bay Area in California, US, 27 different transit agencies were providing public transportation services in early 2020. One consequence of separate companies providing public transport services is that there is often little or no commonality between fares, ticket types, concessionary arrangements, end-user payment systems, cartographic styles or timetable co-ordination. Indeed, at one provincial railway station in the United Kingdom, an entirely empty bus leaving the railway station two minutes before the morning peak intercity rail service arrives from London is a regular occurrence. At the next railway station, 8 minutes further north, the £25.5 million East Midlands Parkway station featured inconveniently timed services with 2 trains an hour in each direction timed 7 minutes apart (northbound services) and 10 minutes apart (southbound services). Furthermore, despite being located less than 5 miles away from an international airport, rail-air connections are only provided by an on-demand (and relatively expensive) minibus shuttle service to the terminal which passengers are obliged 76

Public transport integration

to pre-book. The railway station that was built in 1971 adjacent to Durham Tees Valley airport in northeast England is even less well served, with only one train in each direction calling per week. Unsurprisingly, it is one of the United Kingdom’s least-used stations, with only 74 passengers using the site in 2017–2018 (BBC, 2020). More positively, owing to its location near the north-south M1 motorway and the fact that the station was relatively underutilised, from 30th March  2009, East Midlands Parkway was used as an interchange station for combined multi-modal journeys using Megabus-branded services run by Stagecoach (which at the time also operated the East Midlands Trains franchise). The MegabusPlus services were designed to transport passengers from/to cities in the north of England to the East Midlands Parkway station, where they could then transfer to the Megatrain rail service for the journey to/from London. Routes operated under the MegabusPlus brand included London St Pancras International to/from Hull and Bradford. Although Abellio replaced Stagecoach as the franchise holder for East Midlands Trains (renamed East Midlands Railway from August 2019), Megatrain services continue to operate from the station.

Vignettes of public transport integration This section provides two vignettes relating to public transport integration, namely urban public transport integration and airport ground (or surface) access. Park and Ride is another example (see also Chapter 7).

Urban public transport integration – London Underground and Transport for London As was the case with mainline railway development in the United Kingdom, the construction of the London Underground in the mid-nineteenth century was also influenced by commercial and political concerns. Tracks and stations were often constructed opportunistically wherever land was available, and private companies vigorously competed for custom. Pragmatism and politics were thus significant factors in the development of London’s early urban railway. As with the national railway network, private companies developed the early underground lines from the 1860s onwards. It was not until 1902 that most lines came under the unified control of the Underground Electric Railway Company of London. In 1933, following the creation of the London Passenger Transport Board, the Underground came under state control and was managed alongside the city’s other urban railways, buses, trams, trolleybuses, and coaches. This arguably made timetable co-ordination, service provision, and the creation of a centralised ticketing system (based on a series of zones which radiated out from central London) more straightforward. The subsequent integration of London Overground lines, Transport for London (TfL) rail, the Docklands Light Railway, trams, river buses, the Air Line cable car, and bus services (which are provided by 20 different private bus operators) within a single centralised jurisdiction has further improved the provision of integrated transport in the capital. The authority responsible for overseeing London’s public transport and managing its major roads is TfL, an integrated transport authority created in 2000. In 2003, TfL introduced the Oyster card, a contactless smart card payment system in which users tap in and tap out of the network to ensure the correct fee for the journey is charged (see also Chapter 33). The Oyster card can be used to pay for individual trips (on a pay-as-you-go basis) or be programmed to carry various different travel cards. A daily maximum charge ensures travellers never pay more than the nearest equivalent Day Travelcard. In addition to using an 77

Lucy Budd and Stephen Ison

Oyster card, passengers can now pay for travel using contactless debit and credit cards. The contactless payment system was first introduced on London’s buses in December 2012 and later extended to Underground and TfL rail services from September 2014. The system automatically calculates the best fare for a customer’s specific journey history and then charges them that amount at the end of the day. By mid-2017, more than one billion pay-as-you-go journeys had been made by contactless payment cards, and 2 million journeys a day (accounting for 40% of pay-as-you-go travel) were being undertaken using it. The contactless payment system was further expanded to accept payment using Apple Pay, Google Pay, and Samsung Pay. By 2017, almost 1 in 10 contactless transactions was made using a mobile device, and over 31 million journeys were conducted using mobile phones (Transport for London, 2017). The addition of a free Oyster and contactless app further allow travellers to top up pay-as-you-go (PAYG) credit, receive notifications when PAYG credit runs low, check journey history, buy certain types of Travelcards and travel passes, receive notifications before a Travelcard or travel pass expires, and manage multiple season tickets and cards. Future features will enable it to be used by customers who receive free or discounted travel. The new contactless payment mechanisms have substantially changed the way customers pay for their travel around the city. One key advantage is that it is open and accessible to multiple currencies and does not require customers to buy specific tickets or travel passes. This makes public transport a much more convenient and attractive option for overseas tourists and visitors to the capital. Additional convenience and incentives to use public transport in the capital have been generated through the comprehensive redevelopment of many of the original London railway termini into major destinations (or ‘Destin-Stations’ as some commentators have termed them) in their own right. In 2007, the neo-Gothic station at St  Pancras, which had been designed by the Midland Railway and opened in 1863, was unveiled after an £800-million redevelopment project. The project involved remodelling the station’s layout, creating new platforms for domestic national rail services, and creating space for Eurostar train services from Europe. St  Pancras International is also directly connected to six London Underground lines. The station’s interior was reimagined into a major retail space featuring four zones – Rendezvous, Market, Circle, and Arcade – supporting a range of luxury shops, dining options, lifestyle brands, and, in the case of Market, a daily farmer’s market. The station is used by over 45 million passengers a year. Kings Cross station on the former LNER line to northeast England and Scotland has undergone a similar transformation. The £550-million redevelopment was opened in 2012 and included placing a geodesic steel and glass dome over the top of a London Underground ticket hall, redesigning existing platforms, enlarging the existing concourse, and incorporating a range of shops and restaurants. Onward transport is provided through the combined Kings Cross/ St Pancras London Underground station, local buses, taxis, and bike hire facilities, and the station is used by over 47 million passengers a year.

Airport ground access Airports are major generators of surface/ground access traffic. Passengers, staff, and visitors need to access and egress airports 24 hours a day, and a lot of surface access is undertaken by private car. This impacts congestion and local air quality. Major airports offer the traffic density that is required to make significance investment in public transport provision and provide an integrated multi-model hub. London Heathrow airport is a major generator of traffic, but as well as servicing over 475,000 air traffic movements in 2018, it is also the site of a major public transport interchange (CAA, 2019). The central terminal area features the busiest bus and coach station 78

Public transport integration

in the United Kingdom, with over 1,600 services each day to over 1,000 destinations (Europa. eu, 2019). The site is also connected to the London Underground, TfL rail, and the Heathrow Express rail services. Public transport access focusing on integrated air-rail links and connection to local and regional bus networks has been encouraged as a way of dealing with the issue of surface transport congestion and environmental degradation (Budd et al., 2011). Air-rail integration, such as that found at Heathrow, Birmingham, and Stansted airports in the United Kingdom, is impacted by the: • • •

numbers of passengers allowing for costs to be covered and for a frequency of service to be delivered; the level of regional service so as to facilitate the ease of connection, in essence, a hub-andspoke network in the delivery of rail provision; and the level of difficulty in accessing an airport by private car (Budd et al., 2011).

Airports attract a number of users, most notably passengers, employees, and visitors (such as meeters and greeters), all with different ground access requirements including the need for frequency of service, reliability, convenience, and price. As such, various rail services are more suited for some of these users than others. For example, the Heathrow Express, a direct but more expensive service from Paddington Station in central London to Heathrow airport, is costly and is primarily used by air passengers. The local rail service, on the other hand, tends to be less expensive and provide a more dispersed trip origin and as such is clearly a focus for Heathrow employees. Bus-airport integration has been a significant component of ground access, there being a range of provision, be it local scheduled services or long-distance coach services, such as National Express in the United Kingdom. In this situation, airport bus stations are acting as hub points providing heated and/or air-conditioned waiting areas including food and drink provision, information points, signage, and real-time passenger information (RTPI). The provision of public transport networks within an airport location provides a facility in which passengers can readily transfer between modes or indeed the same mode, in which passengers simply use the public transport provision with the intention of taking a journey via air. As with rail provision, local scheduled services appeal more to airport employees than airline passengers given their frequency of stops and poor provision of luggage facilities. Long-distance coach provision can be slower than journeys undertaken by rail and can also be subject to road congestion (Kazda & Caves, 2015). This can be addressed to some extent by the provision of bus lanes aimed at increasing speed and reliability and thus impacting ridership. Ridership is also a function of familiarity with detailed information about public transport provision. However, the environmental imperative to increase public transport use conflicts with the commercial incentive to generate non-aeronautical revenue through the provision of extensive car parking facilities (see Ison et al., 2009, 2014).

Innovations in public transport As cities become more crowded and polluted, the need for innovative public transport solutions grows as the link between economic performance and seamless mobility becomes ever more apparent. Transport has always been innovative. Changing patterns of passenger demand have, over time, led to the introduction of new public transport and communications technologies from the stagecoach to the train to bus and automated rail. Recent decades have seen the 79

Lucy Budd and Stephen Ison

widespread uptake of real-time travel information and service updates as well as contactless payment systems, on-demand mobility, and increased electrification and automation. Some of these changes have been transformative, such as the introduction of contactless payment for TfL services in London, while others have been more subtle but nonetheless have sought to make public transport more convenient and attractive to potential users. Co-ordinating timetables; synchronising payment systems; and co-locating different transport modes within a single, clean, and safe transport hub are all important. Mainline train services in Switzerland and the Netherlands, for example, routinely feature digital displays which, when the train is on approach to a station, provide details of connecting rail services (time of departure, destination, and platform number) and local bus and tram networks to make the mode transfer as seamless and straightforward as possible. Other transit operators have sought to broaden their appeal to potential users by turning stations into centres for artistic exhibitions and employing leading urban designers and architects to develop new stations and transform the urban realm. The Hungerbergbahn funicular railway in Innsbruck, Austria, for example, features stations by Zaha Hadid, while the architect Sir Norman Foster designed stations on the Bilbao metro. In this way, form and function can be combined to create a space that not only meets the needs of users from the perspective of transport and mobility but which also creates pleasant transfer spaces which have often become tourist attractions in their own right. Elsewhere, major interchanges in cities including Madrid, Paris, and London act not only as transport hubs but also provide many of the shops and services that support daily urban living. Likewise, the landside shopping plaza at Schiphol airport is used not only by airline passengers and airport staff but also by local residents who arrive by train to do their regular shopping. Even smaller interchanges have the potential to offer a limited number of additional services, such as shelters, newspaper stalls, vending machines, and lavatories. In this way, mobility and retail services can be spatially integrated and co-located to provide timesaving convenience to users.

Conclusion Public transport integration is far from a new concept, but it is an idea that has not always been done well owing to the competing (if not commercially conflicting) priorities, interests, and involvement of multiple different public, private, and individual stakeholders. Stakeholder management, when the stakeholders are as varied in their needs and working practices as local authorities/planning committees, national government, private operators, business groups, environmental organisations, and users, is all important. Indeed, public transport integration works most effectively when there is good communication between all parties and all stakeholders are incentivised to work towards a common goal that benefits everyone. There is a compelling need to make public transport work by making it easier, cheaper, and more convenient to access. The goal of reducing private vehicles will not only lead to reductions in congestion and improve local air quality, it will also help countries meet increasingly urgent goals on climate change by helping to decarbonise their transport systems. The COVID-19 pandemic is likely to have a major impact on the passenger use of public transport interchanges. Future research will be required in relation to the use of such hubs as attractive places to meet and to facilitate multi-modal journeys by air, rail, tram, bus, coach, and so on, journeys that from a public perspective are unlikely to be risk free. There will be a reluctance to undertake journeys involving hubs in the short term post-COVID-19, and research focusing on the role of planning and service providers will be all important so as to address legitimate user fears, not least since the spread of the pandemic in the first instance was 80

Public transport integration

facilitated by air passenger movements (see Budd  & Ison, 2020). In addition, at the time of writing, the advice of many governments is to avoid the use of public transport if at all possible. Transport infrastructure has been constructed in an ad-hoc manner, and clearly uncoordinated planning is unlikely to be sufficient in the post-COVID environment. Research will be required as to the correct response of planning, in terms of the both the aftermath of the pandemic and preparing for future reoccurrences, involving all stakeholders but the user in particular. While a high-quality, frequent, convenient integrated transport service is required, a safe environment will be paramount.

References BBC News. (2020). Little-used Teesside airport railway station gets £6m boost. BBC News. Retrieved February 1, 2020, from www.bbc.co.uk/news/uk-england-tees-51340195 Budd, L. C. S., & Ison, S. G. (2020, February 6). Air travel restrictions won’t protect us from the coronavirus. The Conversation. https://theconversation.com/air-travel-restrictions-wont-protect-us-from-the-coro navirus-131237 Budd, T., Ison, S. G., & Ryley, T. (2011). Airport ground access: Issues and policies. Journal of Airport Management, 6(1), 80–97. CAA (Civil Aviation Authority). (2019). Retrieved March 9, 2020, from www.caa.co.uk/uploadedFiles/ CAA/Content/Standard_Content/Data_and_analysis/Datasets/Airport_stats/Airport_data_2018_ annual/Table_03_1_Aircraft_Movements.pdf Campaign for Better Transport. (2018). Integrated transport a new generation of interchanges. Retrieved May 17, 2020, from https://bettertransport.org.uk/sites/default/files/research-files/integrated-transport-anew-generation.pdf Currie, G., & Shalaby, A. (2012). Synthesis of transport planning approaches for the world’s largest events. Transport Reviews, 32(1), 113–136. Europa.eu. (2019). United Kingdom-Hounslow: Transport services (excl. waste transport) 2018/S 054-12 0558 prior information notice services. Retrieved February 17, 2020, from https://ted.europa.eu/TED/ notice/udl;JSESSIONID=93D30052CAEED081E61B37AE03FF61AE.backend-b2?uri=TED: NOTICE:120558-2018:TEXT:EN:HTML Harris, N. G., & Godward, E. W. (1997). The privatisation of British rail. The Railway Consultancy Press. Ison, S. G., Francis, G., Humphreys, I., & Rye, T. (2009). UK airport car parking management. Journal of Airport Management, 3(2), 164–175. Ison, S. G., Merkert, R., & Mulley, C. (2014). Policy approaches to public transport at airports – some diverging evidence from the UK and Australia. Transport Policy, 35, 265–274. Kazda, A., & Caves, R. E. (2015). Airport design and operation (3rd ed.). Emerald Publishing Limited. May, A. D., Kelly, C., & Shepherd, S. (2006). The principles of integration in urban transport strategies. Transport Policy, 13(4), 319–327. Nielsen, G., Nelson, J. D., Mulley, C., Tegnér, G., Lind, G., & Lange, T. (2005). HiTrans best practice guide 2: Public transport – planning the networks. Civitas Consultants. Retrieved April 30, 2020, from http:// civitas.no/assets/hitrans2publictransportplanningthe-networks.pdf Preston, J. (2017). The impact of regulatory reform on public transport markets. In J. Cowie & S. G. Ison (Eds.), The Routledge handbook of transport economics (pp. 121–140). Routledge. Shaw, J. (2000). Competition, regulation and the privatisation of British rail. Ashgate. Transport for London. (2017). One billion journeys made by contactless payment on London’s transport network. Retrieved May  17, 2020, from https://tfl.gov.uk/info-for/media/press-releases/2017/july/ one-billion-journeys-made-by-contactless-payment-on-london-s-transport-network

81

7 INTERMODAL CONNECTIONS Bianca Eiben and Jianqiang Cui

Introduction Around the globe, major cities are faced with the challenge of increasing mobility while simultaneously reducing car dependency, congestion and pollution (Hernandez & Monzon, 2016). With rapid population growth contributing to urban sprawl, distances travelled and travel times have increased (Hernandez  & Monzon, 2016; Zito  & Salvo, 2009). Longer travel times and distances have pushed many public transport users to combine multiple modes of transport to complete their journeys (Hernandez & Monzon, 2016; Zito & Salvo, 2009). This, however, has decreased the attractiveness of public transport for users as a result of additional wait periods and transfers needed to complete a trip (Hernandez & Monzon, 2016; Iseki & Taylor, 2009). The added inconvenience compared to private modes results in a decrease in user satisfaction due to inefficiency and discourages use of public transport (Daudén et al., 2014; Hernandez & Monzon, 2016). To reduce the impacts of intermodal transfers and increase the attractiveness of public transport, intermodal connections need to adopt the principle of intermodality, which is a planning approach that aims to deliver a seamless journey for passengers using different modes of transport in a combined trip chain (Daudén et al., 2014). Integrating multiple modes of transport is becoming more and more complex, with growing populations placing pressure on transport services to balance supply and demand for public transport users. In order for public transport to remain a viable and competitive option in comparison to private modes, providing seamless connectivity is essential for encouraging public transport use (Hernandez et al., 2016). While determining the factors that discourage use is significant, identifying and applying new policy to resolve issues is also important. A Park and Ride scheme is one such example which has become a favourable option to combat some of the common issues surrounding interchange for intermodal connections. Park and Ride schemes have been implemented globally to encourage public transport use, combat congestion in inner-city areas and provide a convenient interchange between public and private modes (Aros-Vera et al., 2013; Lam et al., 2001; Rosli et al., 2012). This chapter explores the issues surrounding intermodal connections for both public and private transport users, examining the role of interchange and Park and Ride schemes in public transport networks. It begins with a review of the relevant literature detailing the key areas of importance for intermodal connections, including key issues for public transport users as well as 82

Intermodal connections

identifying the role of Park and Ride. A case study of the Gold Coast Light Rail Park and Ride in Australia is then presented, followed by a discussion of key issues and implications for public transport users. This chapter concludes with a summary of key points.

Literature review Intermodal connections Intermodal connections are the connections between different modes of transport, that is, between active, public and private modes or, for example, the connections between buses and rail (Pitsiava-Latinopoulou  & Iordanopoulos, 2012). For most public transport users, intermodal connections are a necessary part of a journey, with long distances travelled requiring transfers between modes (Cheng & Tseng, 2016). However, with these additional transfers and sometimes insufficient services which decrease satisfaction, users are shifting behaviour away from public transport. While most intermodal connections adopt the planning principle of intermodality to provide better services, it is often challenging to provide positive experiences for all users and encourage public transport use (Beirão  & Cabral, 2007). Intermodality can be defined most simply as combining different modes of transport to create a seamless travel experience. As such, reducing wait periods and transfers to deliver seamless connectivity were highlighted as important goals in many policies (Liu et al.,1997). But it is often hard to define what seamless connectivity entails. A seamless journey can be defined as having a fully integrated intermodal approach between all transport services, and needs to be the best option for all individuals to get to their destination as quickly and comfortably as possible (Chowdhury & Ceder, 2016). According to the OECD 2016 policy paper, providing a seamless transport system is a process of adapting infrastructure, operations, fare structures and payment systems while also providing the essential information that delivers a convenient travel experience. Therefore, to achieve seamless connectivity, transport networks need to consider the planning and design of terminals to better accommodate passenger needs and the operational management between modes, as well as understanding the travel behaviour of users and non-users to identify gaps and areas for improvement (Beirão & Cabral, 2007; Hernandez & Monzon, 2016; Pitsiava-Latinopoulou & Iordanopoulos, 2012).

Travel behaviour Throughout the research conducted on travel behaviour, it has been highlighted that user experiences play a significant role in modal choice. This is because creating a reoccurring behaviour requires a satisfactory experience to first take place to provide feedback that reinforces the association with a particular goal (Aarts et al., 1998; Ajzen, 1985). Better experiences therefore link to stronger associations and increase the likelihood of repeating a certain behaviour, or in this instance encouraging public transport use. Keeping this in mind, an unsatisfactory experience produces the opposite and may cause a shift in behaviours (Bolles, 1972; Hull, 1943). Identifying the motives for selecting a specific mode is also important to consider when understanding travel behaviour, as they can help identify issues in providing seamless connectivity in transport systems and user experiences. The literature confirms that factors such as cost, convenience and design are emphasised as key issues affecting modal choice (Aros-Vera et al., 2013; Keijer & Rietveld, 1999; Loutzenheiser, 1997; O’Sullivan & Morrall, 1996; Van Acker et al., 2010). Aspects such as an individual’s 83

Bianca Eiben and Jianqiang Cui

socioeconomic demographic, the cost of parking, public transport fares and fuel prices were all identified as attributes considered in mode selection regarding costs of travel (Rosli et al., 2012; Rye & Koglin, 2014; Seik, 1997; Song et al., 2017; Ying & Xiang, 2009; Zhu et al., 2013). How convenient a mode was perceived as being was also identified as having a substantial influence. Convenience is defined by aspects such as speed, reliability, travel time, distance, purpose of travel, accessibility and transfers needed, and can be used to compare the same attributes on different modes (Olaru et al., 2014; Parkhurst, 2000; Rosli et al., 2012). The design of the facilities being used also has influences on mode choice and includes factors such as location, lighting, weather protection, safety and security, ease of wayfinding during transfer and seating availability (Cheng & Tseng, 2016). Whilst considering design for intermodal connections, it is notable that differences in terminal design may affect user experiences and are a crucial aspect to consider when trying to achieve seamless connectivity.

Planning and design of intermodal terminals In planning and designing intermodal terminals, there are a number of issues highlighted in the literature. These include terminal design, choice of location and surrounding land uses (Pitsiava-Latinopoulou & Iordanopoulos, 2012). These are important issues to address, as they significantly impact achieving seamless connectivity as well as influencing the travel behaviour of public transport users. Whilst considering terminal design, Pitsiava-Latinopoulou and Iordanopoulos (2012) distinguished five specific types of intermodal terminals: intercity terminals, commuter transit centres,1 interchanges, Park and Ride and on-street facilities. This chapter focuses on interchange and Park and Ride. Each terminal type has specific design standards that should be met, as poor planning, incorrect location choice and inadequate operational management may cause disruptions to transport systems, as well as discouraging use (PitsiavaLatinopoulou & Iordanopoulos, 2012). Although there are different types of terminals, each terminal should have six essential functions: providing reliable service, adequate facilities, lowcost travel, accessibility for all users, competitive travel time compared to trips without transfer and direct access to different modes serviced at the same terminal (Pitsiava-Latinopoulou  & Iordanopoulos, 2012). Providing these six functions during the design of intermodal connections improves facilities and the potential for positive experiences, and aids in creating seamless connectivity between modes. The location of intermodal terminals, whilst dependent on the type, holds significance in planning for facilities. Incorrect planning for the location and type of terminal may affect the capture of new users and impact current passengers (Pitsiava-Latinopoulou & Iordanopoulos, 2012). Keeping this in mind, the location must consider surrounding land uses to optimise connectivity. Land use planning has become a significant aspect in the planning of transport networks and achieving sustainable development, as there are many benefits to well-planned neighbourhoods, such as reduced travel times, enhanced non-motorised modes and greater accessibility and efficiency of public transport (Geerlings & Stead, 2003; Lah, 2019). For many transport systems, implementing policies such as Transit Oriented Development (TOD) optimises user catchments given the co-ordination between surrounding land uses and transport. TOD has seen success in encouraging public transport use through implementing mixed-use development of residential, business and leisure activities (Carlton, 2009). Intermodal connections between active modes such as walking and public transport have seen beneficial outcomes in developing positive user experiences through more convenient accessibility by adopting landuse planning approaches such as TOD; thus, it is important to consider surrounding land uses whilst planning for and designing intermodal terminals (Carlton, 2009). 84

Intermodal connections

Operational management The operational management of intermodal connections has also been identified as a key aspect of interest for intermodality. The literature outlines that key issues in operational management include the need to overcome poor co-ordination between stakeholders, information provision to passengers and poor governance (Chowdhury & Ceder, 2016; Hernandez et al., 2016). A lack of co-ordinated planning processes between land uses, transport systems and environmental considerations results from poor co-ordination between key stakeholders, and may lead to a segmented approach to policy making and prevent the development and implementation of integrated plans, which in turn impacts achieving seamless connectivity (Geerlings & Stead, 2003). Furthermore, a lack of co-ordination in urban travel and land use policy between the various stakeholders can lead to organisational problems and inefficiencies (Geerlings & Stead, 2003). This can significantly impact the operational management of intermodal connections and may cause issues in timetabling, information provision and providing transport services (Cheng & Tseng, 2016; Geerlings & Stead, 2003).

Interchange As mentioned by Pitsiava-Latinopoulou and Iordanopoulos (2012), an interchange is one example of an intermodal terminal (see also Chapter 6). An interchange can be defined as either an intermodal facility found at connection points of different modes of a transport network or a connection point between different routes for the same mode (Pitsiava-Latinopoulou & Iordanopoulos, 2012). They are normally established in central districts or commercial areas where most public transport routes pass and are easily accessible by active modes such as walking or cycling (Pitsiava-Latinopoulou  & Iordanopoulos, 2012). Interchanges should be designed to meet the needs of frequent users and reduce transfer time (Pitsiava-Latinopoulou  & Iordanopoulos, 2012). By doing so, the attractiveness of public transport can be increased through reduced walking and waiting periods (Hernandez & Monzon, 2016). When designing interchanges, usability principles (including accessibility, ease of navigation, comfort and amenities, information, safety, local area integration and community ownership and activity) were developed and considered by the Station User Panel in 2011 for the Victorian government railway system to improve user experiences (Hernandez & Monzon, 2016). These principles have been identified in various studies as key areas of interest for interchange (e.g. Lois et al., 2018; Lucietti et al., 2016; Monzon et al., 2017), with the most relevant issues being seen as information provision (see also Chapter 25), transfer conditions, accessibility and safety (Hernandez & Monzon, 2016). Whilst considering the issues for interchanges, it is acknowledged that there are two types of issues for users, functional and psychological. The two main functional issues hindering interchange were indicated to be information provision and accessibility, while psychological aspects included safety, security and comfort (Hernandez & Monzon, 2016). Other issues identified in the literature included the layout of the interchange, internal connections, the ease of movement, waiting areas, open spaces, ventilation, lighting, functional and aesthetic integration with the surrounding urban area and additional services (Hernandez & Monzon, 2016; Lucietti et al., 2016). Of the two functional issues, accessibility has been well researched throughout the literature and is highlighted as a key factor impacting intermodality. Functional issues regarding accessibility predominantly revolve around connectivity between modes, transfer times and the overall ease of movement. Numerous articles have discussed ways of reducing the inconvenience of transfers through interchange design (Cheng & Tseng, 2016; Geerlings & Stead, 85

Bianca Eiben and Jianqiang Cui

2003; Pitsiava-Latinopoulou & Iordanopoulos, 2012). For interchanges which involve significant transfer distances such as changing between terminals not within the immediate vicinity, improving pedestrian access via zebra-striped crossings, traffic lights activated by pedestrians, pedestrian islands for crossing large roads and adding traffic calming measures not only reduced transfer time but also increased safety, thus, ensuring a likely positive experience for users (Zito & Salvo, 2009). Information provision at stations is also highlighted as a key area of interest for interchange design. Providing users with more information regarding services, delays and amenities available significantly increased user experiences (Hernandez & Monzon, 2016; Zito  & Salvo, 2009). For example, providing well-signed pedestrian routes in terminal areas reduced time spent on transferring. Another area of interest is ticketing. In some instances, users are required to purchase multiple tickets to transfer between modes, which again adds additional time for transfers. In a number of cases, lack of kiosks further delayed transfer due to the need for queuing. Terminals should be designed with sufficient ticketing machines available or alternative options such as e-tickets, smart cards, the sale of on-board tickets and monthly or yearly passes should be considered (Dorbritz et al., 2009).

Park and Ride The development of Park and Ride since the 1930s has gained significant interest in transportplanning policies for its potential in alleviating congestion, reducing the need for parking in inner city areas and, in some circumstances, preventing private vehicle intrusion into historic cities (Aros-Vera et al., 2013; Lam et al., 2001; Parkhurst, 1995; Rosli et al., 2012). It works on the principle that commuters combine public and private modes of travel, driving part of their journey to a Park and Ride facility that is usually situated on a urban fringe and then taking public transport (either bus, train or tram) to a central location (Aros-Vera et al., 2013; Olaru et al., 2014). This method allows commuters to save on a higher cost of parking and fuel and avoid congestion during peak hours (Aros-Vera et  al., 2013). Used in conjunction with limiting available parking, implementing time restrictions and high-cost paid parking in central locations, Park and Ride has shown promise in relieving congested arterial roads as well as encouraging public transport use (Rye & Koglin, 2014; Seik, 1997; Song et al., 2017; Zhu et al., 2013). The location of Park and Ride should take into account the surrounding land uses to the Park and Ride location, including the proximity to residential housing, additional services nearby, accessibility to other modes of transport and distance to Park and Ride and final destination (Maat et al., 2005). According to the literature reviewed (e.g. Meek et al., 2008; Mingardo, 2013; Parkhurst, 2000), there are three general areas in which Park and Ride facilities are normally situated. These are: 1 2 3

In remote areas, with the purpose of collecting drivers at the beginning of their journey. These are usually located in suburban residential areas; On the urban fringe, with the purpose of intercepting drivers just before their final destination and; Local Park and Ride, normally situated in non-residential suburban areas along main transport corridors. These have the purpose of intercepting drivers somewhere along their trip between the origin and the destination (Mingardo, 2013).

Of these three locations, remote Park and Ride locations are preferred and should be considered to maximise the reduction in car use, as not only would the unintended effects be lower, 86

Intermodal connections

but there would be an overall reduction in car use due to the early intercept of motorists. This method should be favoured to increase public transport use and reduce congestion (Meek et al., 2008; Mingardo, 2013; Parkhurst, 2000). The design of Park and Ride facilities plays a significant role in the perceptions and attitudes towards Park and Ride use. This takes into account the location and Park and Ride attributes, as well as policy implementation. Of the issues for the design of Park and Ride, poorly planned locations were one of the most considered factors found to influence use. This is because when selecting a mode of transport, commuters consider the shortest path and speed needed to arrive at a destination and often select the mode which is most convenient (Mingardo, 2013). Therefore, poor planning of Park and Ride locations that increased the distance to a transfer location reduced the probability to use public transport (Keijer & Rietveld, 1999; Loutzenheiser, 1997; O’Sullivan & Morrall, 1996; White, 2016). The potential time savings was found to be highly valued by users and in some cases more valued than cost savings (Dickins, 1991; Pas, 1998). Therefore, poorly locating a Park and Ride facility can significantly impact the use and hinder the desired outcomes. It must be acknowledged that while the introduction of Park and Ride may bring many benefits to a transport network, several articles reviewed reported an overall increase in private vehicle use because of the introduction of Park and Ride. It appears that public transport users who previously relied on services became aware of the convenience of travelling partway by private vehicle and therefore adjusted their travel behaviour (Wiseman et  al., 2012, p.  39). Traffic redistribution, trip generation and abstraction of users from public transport occurred due to the introduction of these facilities (Dickins, 1991; Hamer, 2010; Meek et al., 2008; Parkhurst, 1995). As this literature review has discussed, this highlights the need for careful consideration towards implementing the usability principles and land use integration when designing intermodal connections. The development of Park and Ride for the Gold Coast Light Rail, discussed as a case study in the next section, is one example of how travel behaviours may change when introducing new intermodal connections into a transport network.

A case study of the Gold Coast Light Rail Park and Ride The following case study of the Gold Coast Light Rail (GCLR) Park and Ride provides an example of how Park and Ride can be used in intermodal connections as well as demonstrating how design can be used to address some of the issues of interchange, as discussed previously. On the Gold Coast, rapid population growth and a lack of competitive public transport options has led to high car dependency and congestion (Aurecon, 2015; Yigitcanlar et  al., 2008). With increasing population and limited scope for the continued expansion of road networks; considerable development is required for public transport infrastructure on the Gold Coast to provide viable alternatives to cars (Aurecon, 2015; Goodwin, 1999). In a bid to shift the travel behaviours of commuters and encourage the use of public transport, the introduction of new services was proposed to increase connectivity. The light rail was implemented and has since gained popularity with commuters, evidenced by the reported reprieve in congestion, with a reduction in vehicle traffic by 13.9% for the Gold Coast Highway, as well as an overall increase in public transport use of 27% since it began operation in 2014 (City of Gold Coast, 2017). A stage two extension has been initiated which incorporated Park and Ride facilities and linked stage one of the light rail to a major existing interchange which serves heavy rail and bus. Park and Ride facilities were proposed at two locations (Helensvale and Parkwood) and were designed to increase intermodality and provide convenient services for existing and potential 87

Bianca Eiben and Jianqiang Cui

users (Department of Transport and Main Roads, 2015). Through upgrading existing facilities at the Helensvale interchange, users were provided with greater modal choice at the terminal and better connectivity to inner-city areas as a result of reduced travel times and fewer transfers compared to bus. This increased seamless connectivity and the attractiveness of public transport. Moreover, constructing new Park and Ride facilities further along the stage two alignment adjacent to the Smith Street Motorway and the Pacific Motorway as part of a local Park and Ride design enabled the potential for capturing more passengers (Department of Transport and Main Roads, 2015). In addition to implementing new transport services on the Gold Coast, changes were also made to policies regarding high-cost paid parking in inner-city areas and time restrictions around public transport facilities and high-traffic areas. These changes aimed at producing negative associations with the use of private transportation by increasing the cost of travel and inconveniencing private transport users with the idea of financial penalty if they exceed parking time limits, thus, increasing the attractiveness of public transport. Improvements to the Gold Coast transport network also aimed at increasing usability across all modes of public transport through a number of means such as optimising information provision both online and at terminals, providing users with sufficient ticketing machines and various options for paying fares and making public transport cost effective. Providing users with apps such as Journey Planner, which enables users to plan their journeys beforehand by time and mode, as well as detailing service numbers, schedules and information about potential delays, better accommodates user convenience. In addition to this, all light rail terminals are fitted with electronic scheduling boards which provide the time until the next departure. Terminals and transport systems on the Gold Coast were also designed to provide users with a range of options for purchasing tickets, which provides better services for users. Tickets include a range of paper tickets, on-board ticketing for buses and smart cards (Go Cards), as well as discounted tickets for concession, students and children. For commuters using multiple modes of transport, Go Cards are most convenient, as they can be used across all modes of public transport, are easily purchased at any self-service kiosk and can be linked to either a credit card or debit card, reducing the need to queue for top ups. Alternatively, Go Cards can also be topped up at self-service kiosks or other convenience stores. To further encourage the use of public transport, users who own a Go Card are provided additional discounts to fares. Using a Go Card is at least 30% cheaper than a single paper ticket, and travel within off-peak hours is an additional 20% cheaper, therefore providing a cost-effective alternative to private modes (TransLink, n.d.). Further, Park and Ride facilities are free to use for those travelling on public transport and offer 24/7 closed-circuit television (CCTV) surveillance to provide users with additional security.

Discussion As evidenced in many transport studies, understanding the travel behaviour of users and nonusers is an essential part in developing successful public transport systems. Determining what factors are valued by users, as well as understanding the attitudes and perceptions of a particular mode, are necessary areas to comprehend when planning for and designing transport infrastructure. Failure to identify and provide for the needs of commuters can negatively impact travel behaviour and the prospects for achieving intermodality. Taking into consideration the GCLR Park and Ride example, it is evident that transport planners aimed to fill the gap in available, reliable and efficient public transport provision through the introduction of new services while also addressing the key issue of congestion and high car dependency. By introducing new modes of transport for commuters on the Gold Coast network, transport planners were able 88

Intermodal connections

to incorporate usability principles by increasing functionality on the network which in turn improves intermodality. The implementation of the GCLR can be suggested as a good example of how to implement intermodal connections in a growing transport system. As identified in the literature review, several aspects needed to be considered to increase intermodality, which included the travel behaviour of users, terminal design and the operational management of the transport system. Whilst considering how transport planners were able to shift behaviours towards public transport for the GCLR Park and Ride study, the travel behaviour of commuters needs to be examined. As identified in the literature review, factors such as cost, convenience and design are highlighted as key influencers in mode selection. For the GCLR study, transport policies targeted cost, convenience and accessibility as key aspects to improve for intermodal connections, as they are identified as highly valued factors in modal selection. Policies took into account that most commuters want to use a mode of transport that is convenient, cost effective, reliable, comfortable and accessible. Acknowledging that one of the main issues with transport systems on the Gold Coast was a lack of public transport options, the introduction of new services instantly became appealing for commuters just by being available for use. Acknowledging (from the review) that the two key issues for users of interchange are functional and psychological, improving these factors in existing facilities and applying previous learning in similar modes can prove beneficial in achieving intermodality across a network. When planning for and designing intermodal connections, policies must also consider the factors surrounding the cost of travel, as this is a significant aspect considered by commuters in modal selection. Rising fuel prices and increasing cost of vehicle maintenance and registration, as well as additional parking expenses in central areas for those using private modes, further influence modal selection and travel behaviours, with commuters seeking cheaper options. Ensuring that Park and Ride facilities are free to use without time constraints and that public transport fares are affordable for all users increases the appeal of public transport, as seen in the GCLR Park and Ride example. Through providing users with the option of further cost savings via the use of the Go Card, which can be used across all transport modes, users gained additional incentive to use public transport. Notably, to further encourage and promote public transport on the Gold Coast, during major events such as New Year’s Eve or sporting events, public transport is free to use. This allows commuters to not only test the convenience and reliability of the service without having to sacrifice financially but also aims to provide users with the opportunity to receive positive feedback that may increase the likelihood of continued use. The design of new terminals for the GCLR was also considered in order to optimise convenience for commuters and expand the catchment for existing and potential passengers. As highlighted in the literature review choice of location, terminal design and surrounding land uses all impact providing users with seamless connectivity. For the GCLR Park and Ride, this was mediated by selecting locations on the urban fringe that would capture users commuting through areas of high congestion for the Parkwood station and by selecting locations with accessibility to other modes of transport and additional services at Helensvale. Furthermore, terminals were designed with consideration to disability access and ease of movement between the Park and Ride and the light rail stations, as well as to other amenities at Helensvale, with signalled crossing to shops and access to other modes within the same terminal, which reduces transfer penalties. This highlights the need for careful consideration with surrounding land use integration when planning for intermodal terminals to optimise user catchments and the need to consider the operational management within existing transport networks. 89

Bianca Eiben and Jianqiang Cui

The literature review also identified that the operational management of intermodal connections needs to address poor co-ordination between stakeholders, information provision to passengers and poor governance. The GCLR Park and Ride example demonstrates how integrating planning networks can successfully provide satisfactory services for users. This is achieved by co-ordinating public transport networks through one main agency (in this case the Queensland Department of Transport and Main Roads), which contracts work to private stakeholders, and using universal ticketing across all modes, as well as developing a system of information provision across all light rail networks both online and at terminals via electronic notice boards. This demonstrates a good example of how to provide users convenient and upto-date information on delays and scheduled services to increase connectivity and provide users with a seamless journey.

Conclusion This chapter has discussed how interchange and Park and Ride schemes are used in intermodal connections and examined the GCLR Park and Ride model that was recently introduced. The chapter outlined key areas of interest in intermodal connections and discussed the importance of intermodality and what measures can be taken to achieve seamless connectivity for public transport users. Key takeaways from this chapter include the need for transport policies to understand the travel behaviour of users in order to optimise user experiences and encourage public transport use through cost-effective, reliable, accessible and convenient services. When designing intermodal connections, public transport systems need to thoroughly consider terminal design, location and surrounding land uses to encourage the use of public transport. In order for public transport networks to remain a competitive and viable alternative to private modes, it is evident that the introduction of new policies that hinder the use of private transportation is necessary in conjunction with improving existing and new public transport facilities. This chapter highlights several key areas for further research, including investigating better strategies for mitigating increasing private vehicle use when introducing new modes, researching strategies to optimise terminal design suitable for the population densities they serve and investigating best practice for community engagement prior to planning for new infrastructure to improve the delivery of intermodality on transport networks.

Note 1 Commuter transit centres mainly serve travel to and from urban centres, are located in central areas of the cities and give high-frequency hourly services with access to numerous modes – so basically a multimodal hub (Pitsiava-Latinopoulou & Iordanopoulos, 2012).

References Aarts, H., Verplanken, B., & Van Knippenberg, A. (1998). Predicting behavior from actions in the past: Repeated decision making or a matter of habit? Journal of Applied Social Psychology, 28(15), 1355–1374. Ajzen, I. (1985). From intentions to actions: A theory of planned behavior. In Action control (pp. 11–39). Springer. Aros-Vera, F., Marianov, V., & Mitchell, J. E. (2013). p-Hub approach for the optimal park-and-ride facility location problem. European Journal of Operational Research, 226(2), 277–285. Aurecon. (2015). Gold coast high capacity public transport corridor studies. City of Gold Coast. Retrieved May  16, 2020, from www.goldcoast.qld.gov.au/documents/bf/gc-high-capacity-public-transportpurpose-needs.pdf

90

Intermodal connections Beirão, G.,  & Cabral, J. S. (2007). Understanding attitudes towards public transport and private car: A qualitative study. Transport Policy, 14(6), 478–489. Bolles, R. C. (1972). Reinforcement, expectancy, and learning. Psychological Review, 79(5), 394. Carlton, I. (2009).  Histories of transit-oriented development: Perspectives on the development of the TOD concept (Working Paper No. 2009, 02). Institute of Urban and Regional. Cheng, Y. H., & Tseng, W. C. (2016). Exploring the effects of perceived values, free bus transfer, and penalties on intermodal metro–bus transfer users’ intention. Transport Policy, 47, 127–138. Chowdhury, S., & Ceder, A. A. (2016). Users’ willingness to ride an integrated public-transport service: A literature review. Transport Policy, 48, 183–195. City of Gold Coast. (2017). Light rail corridor 2017 status report. Retrieved May 18, 2020, from www.gold coast.qld.gov.au/documents/bf/light-rail-status-report-2017.PDF Daudén, F. J., Carpio-Pinedo, J., & García-Pastor, A. (2014). Transport interchange and local urban environment integration. Procedia-Social and Behavioral Sciences, 160, 215–223. Department of Transport and Main Roads. (2015). Gold Coast light rail stage 2: Fact sheet. Retrieved May  18, 2020, from www.tmr.qld.gov.au/-/media/Projects/G/Gold-Coast-Light-Rail-Stage-2/gclr 2factsheet1115.pdf?la=en Dickins, I. S. (1991). Park and ride facilities on light rail transit systems. Transportation, 18(1), 23–36. Dorbritz, R., Lüthi, M., Weidmann, U.,  & Nash, A. (2009). Effects of onboard ticket sales on public transport reliability. Transportation Research Record, 2110(1), 112–119. Geerlings, H., & Stead, D. (2003). The integration of land use planning, transport and environment in European policy and research. Transport Policy, 10(3), 187–196. Goodwin, P. (1999). Transformation of transport policy in Great Britain. Transportation Research Part A: Policy and Practice, 33(7–8), 655–669. Hamer, P. (2010). Analysing the effectiveness of park and ride as a generator of public transport mode shift. Road & Transport Research: A Journal of Australian and New Zealand Research and Practice, 19(1), 51. Hernandez, S., & Monzon, A. (2016). Key factors for defining an efficient urban transport interchange: Users’ perceptions. Cities, 50, 158–167. Hernandez, S., Monzon, A.,  & de Oña, R. (2016). Urban transport interchanges: A  methodology for evaluating perceived quality. Transportation Research Part A: Policy and Practice, 84, 31–43. Hull, C. L. (1943). Principles of behavior (Vol. 422). Appleton-Century-Crofts. Iseki, H., & Taylor, B. D. (2009). Not all transfers are created equal: Towards a framework relating transfer connectivity to travel behaviour. Transport Reviews, 29(6), 777–800. Keijer, M. J. N., & Rietveld, P. (1999). How do people get to the railway station: A spatial analysis of the first and the last part of multimodal trips (Research Memorandum, No. 1999-9). Faculty of Economics and Business Administration, Vrije Universiteit Amsterdam. Lah, O. (2019). Sustainable urban mobility in action. In O. Lah (Ed.), Sustainable urban mobility pathways (pp. 133–282). Elsevier. Lam, W. H., Holyoak, N. M., & Lo, H. P. (2001). How park-and-ride schemes can be successful in Eastern Asia. Journal of Urban Planning and Development, 127(2), 63–78. Liu, R., Pendyala, R. M., & Polzin, S. (1997). Assessment of intermodal transfer penalties using stated preference data. Transportation Research Record, 1607(1), 74–80. Lois, D., Monzón, A., & Hernández, S. (2018). Analysis of satisfaction factors at urban transport interchanges: Measuring travellers’ attitudes to information, security and waiting. Transport Policy, 67, 49–56. Loutzenheiser, D. R. (1997). Pedestrian access to transit: Model of walk trips and their design and urban form determinants around Bay area rapid transit stations. Transportation Research Record, 1604(1), 40–49. Lucietti, L., Hoogendoorn, C., & Cré, I. (2016). New tools and strategies for design and operation of urban transport interchanges. Transportation Research Procedia, 14, 1240–1249. Maat, K., Van Wee, B.,  & Stead, D. (2005). Land use and travel behaviour: Expected effects from the perspective of utility theory and activity-based theories.  Environment and Planning B: Planning and Design, 32(1), 33–46. Meek, S., Ison, S., & Enoch, M. (2008). Role of bus-based park and ride in the UK: A temporal and evaluative review. Transport Reviews, 28(6), 781–803. Mingardo, G. (2013). Transport and environmental effects of rail-based park and ride: Evidence from the Netherlands. Journal of Transport Geography, 30, 7–16. Monzon, A., Alonso, A., & Lopez-Lambas, M. (2017). Joint analysis of intermodal long distance-last mile trips using urban interchanges in EU cities. Transportation Research Procedia, 27, 1074–1079.

91

Bianca Eiben and Jianqiang Cui OECD. (2016). Intermodal connectivity for destinations. OECD Publishing. Retrieved May 16, 2020, from www.oecd.org/industry/tourism/2016%20-%20Policy%20paper%20on%20Intermodal%20Connec tivity%20for%20Destinations.pdf Olaru, D., Smith, B., Xia, J. C., & Lin, T. (2014). Travellers’ attitudes towards park-and-ride (PnR) and choice of PnR station: Evidence from Perth, Western Australia. Procedia-Social and Behavioral Sciences, 101–110. O’Sullivan, S., & Morrall, J. (1996). Walking distances to and from light-rail transit stations. Transportation Research Record, 1538(1), 19–26. Parkhurst, G. (1995). Park and ride: Could it lead to an increase in car traffic? Transport Policy, 2(1), 15–23. Parkhurst, G. (2000). Influence of bus-based park and ride facilities on users’ car traffic.  Transport Policy, 7(2), 159–172. Pas, E. I. (1998). Time in travel choice modeling: From relative obscurity to center stage. In Theoretical foundations of travel choice modeling. Emerald Publishing Limited. Pitsiava-Latinopoulou, M., & Iordanopoulos, P. (2012). Intermodal passengers terminals: Design standards for better level of service. Procedia-Social and Behavioral Sciences, 48, 3297–3306. Rosli, N. S., Adnan, S., Alyia, S. A., Ismail, F. D., Hamsa, K., & Azeez, A. (2012). A theoretical review on sustainable transportation strategies: The role of park and ride facility as a generator of public transport mode shift. Proceedings of the 23rd EAROPH World Congress the Eastern Regional Organisation for Planning and Human Settlement. Rye, T., & Koglin, T. (2014). Parking management. In Parking issues and policies (transport and sustainability, Vol. 5). Emerald Publishing Limited. Seik, F. T. (1997). Experiences from Singapore’s park-and-ride scheme (1975–1996).  Habitat International, 21(4), 427–443. Song, Z., He, Y.,  & Zhang, L. (2017). Integrated planning of park-and-ride facilities and transit service. Transportation Research Part C: Emerging Technologies, 74, 182–195. Translink. (n.d.). About Go Card. Retrieved May 16, 2020, from https://translink.com.au/tickets-and-fares/ go-card/about-go-card Van Acker, V., Van Wee, B., & Witlox, F. (2010). When transport geography meets social psychology: Toward a conceptual model of travel behaviour. Transport Reviews, 30(2), 219–240. White, P. R. (2016). Public transport: Its planning, management and operation. Taylor & Francis. Wiseman, N., Bonham, J., Mackintosh, M., Straschko, O., & Xu, H. (2012). Park and Ride: An Adelaide case study. Road & Transport Research: A Journal of Australian and New Zealand Research and Practice, 21(1), 39. Yigitcanlar, T., Fabian, L., & Coiacetto, E. (2008). Challenges to urban transport sustainability and smart transport in a tourist city: The Gold Coast, Australia. The Open Transportation Journal, 2(1). Ying, H., & Xiang, H. (2009, October). Study on influence factors and demand willingness of park and ride. In  2009 second international conference on intelligent computation technology and automation  (Vol. 4, pp. 664–667). IEEE. Zhu, C., Jia, B., Han, L., & Gao, Z. (2013). Commuting pattern with park-and-ride option for heterogeneous commuters. Discrete Dynamics in Nature and Society, 2013. Zito, P.,  & Salvo, G. (2009). Random utility approach to re-designing the passengers interchange node. Urban Transport XV-Urban Transport and the Environment, 107, 51–62.

92

8 PUBLIC TRANSPORT AND THE ENVIRONMENT Ka Ho Tsoi and Becky P.Y. Loo

Introduction As private motorisation becomes pervasive across the world, transport has generated a wide range of negative environmental externalities. These include air pollution, noise pollution, water pollution, hydrologic impacts, habitat and ecological degradation, depletion of nonrenewable resources and climate change (Chapman, 2007; Litman & Burwell, 2006; OECD, 1996). Among these negative impacts, the most prolonged and significant impacts are the increasing amount of carbon emissions that contribute to climate change. The global transport carbon emissions have increased significantly from 4,608 million metric tonnes CO2 in 1990 to 8,145 million metric tonnes CO2 in 2018 (Crippa et al., 2019), representing a 77 percent increase. Transport is also one of the major sources of global carbon emissions. As of 2018, transport contributed around a quarter of the carbon dioxide emissions in the world (International Energy Agency, 2019). In this context, Loo and Banister (2016) examined the transport decoupling pathways of the 16 countries that make up 64  percent of the global GDP and 63 percent of global CO2 emissions. Their findings suggest that environmental decoupling in the transport sector has not made much progress in the past three decades. To reverse the escalating trend of transport carbon emissions, developing mass public transport has been advocated to suppress car growth and encourage mode shift. According to UITP (2017), there were around 243 billion public transport journeys made in 39 countries of the world (mainly the developed and newly industrialised countries) in 2017, which is an 18 percent increase when compared to the figure in 2000. Moreover, the role of public transport in reducing carbon footprints is fundamental. Asian Development Bank (2010) conducted a study of the savings in CO2 from its projects and concluded that railways, metro rail and bus rapid transit (BRT) can save over 2,000 tons of CO2 per lane per kilometre every year. In order to better understand how public transport can promote environmental sustainability, this chapter first highlights the three important environmental benefits of public transport, ranging from energy savings and greenhouse gas reductions to air quality improvement and climate change mitigation. It then examines the effective policy instruments in public transport that help decarbonise the transport sector. The successful global experience of public transport policies of (i) transit-oriented development, (ii) integrated public transport systems, (iii) public and active transport and (iv) public transport electrification is discussed. Figure 8.1 depicts the 93

Ka Ho Tsoi and Becky P.Y. Loo

Figure 8.1 The relationship between public transport and the environment

relationship between public transport and the environment. This helps policymakers and practitioners develop a holistic policy toolkit of effective public transport policies.

Environmental benefits of public transport Energy savings and greenhouse gas reductions One of the key advantages of developing public transport is to reduce energy consumption and greenhouse gas (GHG) emissions, in particular carbon dioxide (CO2). Multiple studies have quantified the energy consumption and GHG emissions of different transport modes. To allow for a consistent comparison across transport modes, these figures cannot be directly compared based on the absolute volume. Instead, specific indicators need to be evaluated with the normalisation of passenger-distance (e.g. passenger-kilometres and passenger-miles), fuel carbon intensity and energy intensity. Sims et al. (2014) consolidated the direct CO2 emissions of major passenger transport modes per passenger distance from a range of international experiences. The typical ranges of CO2 emissions in private and public transport modes are shown in Table 8.1. Overall, public transport generally has lower ranges of carbon emissions than private transport. As expected, rail has the lowest upper limit, largely because of its energy efficiency with higher capacity. Water passenger transport and BRT are similar, but the range between the lower and upper limits can be quite large (i.e. more than five times different). This suggests the number of passengers, choice of fuels and efficiency of engines can affect the level of direct CO2 emissions. In some cases, the upper limit of public transport can be lower than the lower limit of private transport. This further indicates that the environmental benefits of public transport can only be effectively delivered under certain circumstances, such as high ridership, the use of less carbonintensive fuels and better fuel efficiency. The average energy intensity of different passenger transport modes in the United States in three different years (i.e. 1990, 2000 and 2010) is shown in Figure 8.2. In most circumstances, the transport energy intensity of each transport mode per passenger mile has gradually decreased over time, largely due to the improvement in fuel-saving technologies. A lower energy intensity per passenger-mile indicates that less fuel is consumed for the same travel distance per passenger. When compared to other road transport modes, passenger cars have a higher energy 94

Public transport and the environment Table 8.1 Range of direct CO2 emissions of different passenger transport modes (g CO2/km) Transport modes

Lower range

Upper range

Private transport Light-duty vehicles (diesel, hybrid) Light-duty taxi (gasoline, diesel, hybrid) Motorcycle

80 140 80

220 >250 220

25 30 20 95

140 120 150 >250

Public transport Coach, bus and rapid transit Rail Water Air Source: Sims et al., 2014

Figure 8.2 Energy intensity of different transport modes in the United States Source: Bureau of Transport Statistics, 2018

intensity than other public transport modes such as bus and intercity rail (up to four times). Air travel traditionally possesses the highest energy intensity but has gradually decreased with more advanced fuel-saving technology. The energy intensity of public transport is quite similar to passenger cars, and public transport even had a higher energy intensity than passenger cars in 2000. The possible reason could be the lower ridership for public transport in the United States. This indicates that promoting public transport with sufficient ridership is key to reducing energy intensity and carbon emissions. Another common indicator of evaluating the environmental benefits of public transport is fuel economy. Fuel economy is the ratio of total passenger distance to the total fuel consumption. The higher the ratio, the higher the energy efficiency per passenger-distance. Generally, an increase in public transport ridership and a decrease in fuel consumption can lead to higher fuel economy. Figure 8.3 depicts the average fuel economy of different transport modes per passenger mile in the United States in 2018. The unit is passenger-mile per gasoline-gallon equivalents (GGE). In 2018, the fuel economy of intercity rail was the highest, followed by domestic air, 95

Ka Ho Tsoi and Becky P.Y. Loo

Figure 8.3 Average fuel economy of different transport modes in the United States Source: U.S. Department of Energy, 2018

transit rail, commuter rail and transit buses. It is noteworthy that public transport has a lower fuel economy than cars in the United States. The major reason is again largely due to the low ridership, which is usually less than 25 percent of the full capacity (U.S. Department of Energy, 2018). The demand response mode (e.g. taxis) usually has the lowest fuel economy because fuel is consumed to reach passengers. The potential savings in energy consumption by public transport also display similar patterns in the European context. According to ODYSSEE-MURE (2019), an energy database cofunded by the European Union, the fuel consumption measured in kg of oil equivalents (koe) per passenger-kilometre of public transport is also lower than that of the private transport in three years (i.e. 2000, 2007 and 2017). With more fuel-efficient technologies, there has been a gradual decrease in energy consumption in different transport modes, with the greatest drop in domestic air, followed by cars and railways. However, there is a slight increase in energy consumption for buses (from 0.015 koe/pkm to 0.02 koe/pkm). In 2017, domestic air had the highest energy consumption rates (0.068), followed by cars (0.038), buses (0.021) and rail (0.007). The energy consumption for buses and rail is far lower than that of passenger cars. The energy consumption of buses and rail is typically half to one-fifth of that of cars. This illustrates the benefits of public transport in reducing “direct” GHG emissions (i.e. fuel combustion in vehicle operations). Indeed, promoting public transport can also further reduce “indirect” energy consumption and hence maximise carbon savings in the lifecycle. Indirect carbon emissions from transport are composed of, but not limited to, vehicle manufacturing, maintenance and construction of infrastructure for storage. Overall, the difference between private vehicles and public transport is even more significant if we evaluate the environmental benefits from a lifecycle perspective. Chester and Horvath (2009) conducted a detailed analysis in the lifecycle energy and emission inventories of a wide range of transport modes in different cities of the United States. Their findings demonstrate that using public transport can reduce 96

Public transport and the environment

carbon emissions significantly in vehicle manufacturing, vehicle maintenance, vehicle insurance and fuel production. Although infrastructure construction (e.g. stations and terminals) and public transport operations (i.e. daily energy consumed for operating terminals and parking) also contribute towards energy consumption, GHG emissions are significantly lower than their private counterparts. Overall, the reduction of GHG emissions (measured in CO2 equivalent) is around two to four times. The reduction is particularly significant for carbon monoxide (CO), particulate matter of 10 micrometres or less in diameter (PM10), volatile organic compounds (VOCs) and nitrogen oxides (NOX), including nitrogen dioxide (NO2).

Air quality improvement A mode shift from private car to public transport usage is essential to improve air quality, in particular for areas of heavy traffic and dense population. The common assessment parameters include black carbon (BC), PM10 and ozone (O3). Although there is a theoretical debate as to the linkage between public transport supply and air pollution levels (Beaudoin et al., 2015; Harford, 2006), several empirical studies suggest that public transport can somehow alleviate air pollution. Lalive et al. (2018) investigated the impacts of increasing public transport supply on air quality. Overall, increasing rail service by 10 percent can reduce CO and NO by approximately 1 percent and 2 percent, respectively, but there is no direct link to other air pollutants such as sulphur dioxide (SO2) and ozone (O3). Titos et al. (2015) evaluated the effect of public transport on the air quality in Spain and Slovenia and indicated that the implementation of a new public transportation system (including bus route rationalisation and the use of higher-capacity fleets) can lead to a reduction of BC and PM10 by 37 percent and 33 percent, respectively. Chen and Whalley (2012) investigated the level of air pollutants after the introduction of a new metro line in Taipei, and the results illustrate a reduction of CO by 5 to 15 percent, although ground-level O3 does not have a clear relationship with the new metro service. Sun et al. (2019) conducted a similar study on the increase in public transport on air quality in Chinese urban areas. Their findings suggest that marginal improvements of public transport such as increasing the number of buses by 1 percent can lead to a reduction of air pollution index by 0.08 percent. Some studies examine the air quality during public transport disruption (e.g. public strikes) as a counterfactual scenario to demonstrate the benefits of public transport in reducing air pollution (i.e. the change of air quality without public transport service). Bauernschuster et  al. (2017) examined the public transport strikes in several major German cities from 2002 to 2011. According to their study, there is generally an increase of particular pollutants (NO2 and PM10) by 14 percent during the morning peak hour of a strike day. Even for a more robust model that considers weather controls and city indicators, there is a 4.3 percent increase of NO2 when compared to non-strike periods. Basagaña et al. (2018) examined the air quality impacts during public transport strikes (including bus and metro) in Barcelona from 2005 to 2016. Similarly, their findings after accounting for other types of strikes suggest that days of strike are associated with an increase of NOx and BC when compared to non-strike days, with more significant impacts in full-day strikes and metro strikes of multiple days. Moreover, increasing public transport accessibility is suggested to be a way of reducing traffic congestion and further improving air quality. Road congestion exacerbates air quality when idling engines consume fuel energy, and the associated acceleration and deceleration can also lead to higher amounts of fuel combustion. A wide range of studies suggest that road congestion can generate more concentrated air pollution on roads and even at a neighbourhood scale (Barth & Boriboonsomsin, 2008; Zhang & Batterman, 2013). Several studies investigated the relationship between traffic congestion and air pollutants and found that an increase in rail and 97

Ka Ho Tsoi and Becky P.Y. Loo

bus mileage can reduce congestion costs, with a larger reduction for rail service (Nelson et al., 2007; Winston & Langer, 2006). The previous studies align with some counterfactual scenarios where a disruption of public transport can lead to an average increase in traffic delays (Anderson, 2014). The diverging results are explained in Beaudoin and Lawell (2017) – that the association of public transport investment, congestion and air quality of a total of 96 urban areas in the United States between 1991 and 2011 differed in magnitude across cities. The spatial heterogeneity can be attributed to different population densities, availably of public transport networks, initial modal share and level of transport investment (Anderson, 2014; Baum-Snow et al., 2005; Beaudoin et al., 2015; Winston & Langer, 2006). The empirical evidence suggests that public transport can greatly contribute to air quality improvement, and it has significant social implications because it reduces residents’ exposure to air pollutants and addresses public health issues.

Climate change mitigation Encouraging public transport can also be an effective way to reduce carbon emissions per travel distance and contribute to climate change mitigation in the long run. In a study by Davis and Hale (2007), public transport was expected to reduce CO2 emissions by 6.9  million metric tonnes and save an additional amount of 340 million gallons of gasoline due to reduced congestion, which is equivalent to three million metric tonnes of CO2 emissions. Moreover, there was also a substantial decrease in other hazardous GHG such as hydrofluorocarbons (HFCs) and chlorofluorocarbons (CFCs) (Davis  & Hale, 2007). According to a large-scale study of Latin American cities (Wright & Fulton, 2005), it was estimated that only a 5 percent increase in BRT mode share, a 1  percent decrease of private cars and taxis and a 2  percent decrease of minibuses can potentially reduce CO2 emission by 4  percent. This can be essential, as a 1 percent reduction of car share can reduce around one million tonnes of CO2 in the 20-year projection period (Wright & Fulton, 2005). According to Woodcock et al. (2009), with a significant increase in the development of public and active transport (i.e. walking and cycling), the increasing trend of CO2 emission by 2030 can potentially slow down to only 235 percent above the 1990 level and 199 percent if there is a wider adoption of cleaner and fuel-efficient vehicles. Moreover, Shin et al. (2009) investigated the impacts of public transport on land consumption. Their findings suggest that developing a high-capacity public transport system to encourage mode shift is an effective policy to reduce the amount of land required for transport land use (i.e. road infrastructure). This helps protect rural land from being encroached, in particular for undeveloped green areas, forest reserves and natural habitats near cities. Climate change cannot be tackled with only one single policy (such as an improvement or diversion to public transport). There needs to be a holistic framework to integrate various policy instruments in paving the way for decarbonisation (Loo & Tsoi, 2018). In fact, public transport can generate a wide range of synergies (or co-benefits) with other policy instruments. Kwan and Hashim (2016) conducted a comprehensive review of the co-benefits of public transport in climate change mitigation from 2004 to 2015. The commonly identified scenarios of potential mitigation strategies can produce co-benefits, which include the promotion of public transport, limit of car traffic, land-use policies, fuel technology advancement and modal shift initiatives. Some empirical examples include public transport accessibility and carbon tax policies (Fu & Kelly, 2012), metro expansion combined with introducing hybrid buses (McKinley et al., 2005), improved fuel efficiency and expansion of BRT (Chavez-Baeza  & Sheinbaum-Pardo, 2014) and promoting modal shift and the use of non-motorised transport such as walking and cycling (Creutzig et al., 2012). 98

Public transport and the environment

Public transport policies and decarbonisation Transit-oriented development Transit-oriented development (TOD) is a community-based strategy highly dependent on public transport development. The key to TOD is to integrate dense and diverse land uses around a highly accessible public transport location with a walkable environment (Cervero, 2004; Curtis, 2012; Loo & Banister, 2016). The fundamental elements of TOD are the “3Ds” – density, diversity and design (Cervero & Kockelman, 1997). Overall, dense development, mixed land uses and a people-centric design of the built environment are favourable conditions. Essentially, TOD and public transport are used to be mutually reinforcing. On the one hand, TOD provides denser and more diverse opportunities around public transport stations, which can sustain public transport ridership and support its long-term viability, while, on the other hand, public transport provides sustainable, accessible and high-capacity travel modes to accommodate the travel demand of passengers and residents in TOD communities. To deliver the associated environmental benefits effectively, TOD needs to be supported by the use of sustainable transport alternatives (i.e. public and active transport) (Cervero & Sullivan, 2011; Loo & du Verle, 2017; Loo & Tsoi, 2018; Litman & Steele, 2017). In this context, Ewing and Cervero (2010) integrated two other components to make the “5Ds” (i.e. distance to transit and destination accessibility). “Distance to transit” and “destination accessibility” highlight the fundamental principle of developing an accessible public transport and pedestrian network for connecting to the transit location and nearby activity opportunities. The goal is to encourage public and active transport. Ogra and Ndebele (2014) also incorporated demand management to form the “6Ds”. “Demand management” refers to the ability to accommodate the existing and future demand of different transport modes, such as the allocation/expansion of public transport facilities and parking infrastructure. These three elements are largely related to the public transport dimension. The effectiveness of TOD in achieving transport decarbonisation has been examined by a wide range of researchers. The focus here is on the empirical studies of TOD and its association with transit ridership and travel distance. Multiple studies highlight that TOD can reduce travel by private vehicles, increase public transport ridership and promote active transport in different geographical contexts (Loo et al., 2010; Loo et al., 2017a; Kamruzzaman et al., 2013; Sung & Oh, 2011). Moreover, proximity of activity nodes to public transport locations can effectively reduce travel distance (Zhang, 2010; Lee et al., 2010; Bartholomew & Ewing, 2008), thus further reducing energy consumption of motorised travel and encouraging non-motorised access (i.e. walking and cycling). Loo et al. (2017a) investigated rail-based TOD (RTOD) communities in Hong Kong, and they found that RTOD is an effective policy instrument in increasing population and land-use density. Overall, the use of multi-modal rail and walking modes increased in both greenfield and brownfield sites after the opening of new railway stations (Loo et al., 2017a). However, it is noteworthy that TOD is largely a “neighbourhood” concept. In other words, TOD strategies are not identical across different communities. They have to be highly place based and context specific. Using Hong Kong as an example, TOD in different parts of Hong Kong displays distinctive characteristics in terms of the 3Ds (i.e. density, diversity and design). According to principal component analysis (Loo & du Verle, 2017), there are five types of TOD communities, CBD-type, integrated community, balanced-type, residential-type and stationtype. In these different neighbourhoods, the public transport settings and accessibility features can be very different. For example, CBD-type TOD neighbourhoods require the provision of 99

Ka Ho Tsoi and Becky P.Y. Loo

all hierarchies of public transport with a dense and well-connected street network. A stationtype TOD neighbourhood features proximity to other transport modes and a good design of the metro station, such as covered walkways and car-free environment. An integrated community TOD neighbourhood usually has higher residential density and denser coverage of metro exits. Though there can be, to some extent, spatial heterogeneity of public transport planning and TOD, there are several common policy features based on empirical evidence (Bertolini et al., 2012; Cervero & Sullivan, 2011; Loo et al., 2010; Loo et al., 2017a; Renne, 2009). First, integrating public transport and land-use planning at a neighbourhood scale is a fundamental principle. The essence is to provide attractive public transport and well-coordinated land use. Appropriate land zoning strategies such as high-density residential or employment land use and mixed commercial, recreation and retail opportunities around the public transport stops are commonly found in the more effective TOD communities. Second, developing a well-connected pedestrian and cycling network offering a pleasant travel experience (i.e. safety, comfort and convenience) is essential. Non-motorised access and a reduction of pedestrian-vehicular conflicts around public transport stations are duly paramount. Third, high-frequency, efficient and good-quality public transport services are important for the success of TOD. Developing a sustainable public transport hierarchy by integrating different levels of transport modes in the public transport hub is vital.

An integrated public transport system Developing an integrated public transport system is an important initiative to encourage modal shift and promote transport decarbonisation. An integrated public transport system integrates different public transport modes in a holistic network and offers fast, convenient, comfortable and accessible transfer between transport modes (Chowdhury & Ceder, 2016; Li & Loo, 2016; Ülengin et al., 2007) (see also Chapters 6 and 31). Offering attractive transfers within the same mode (e.g. metro lines, bus routes or tram routes) or multi-modal transfers (e.g. train-metro, train-bus or metro-bus) can help provide a door-to-door service and encourage modal shift. Earlier studies focus on the integration of air and road transport (Givoni & Banister, 2006; Li & Loo, 2016). If properly planned, the integration can foster a “complementary” arrangement between airlines and train operators, where trains can help serve areas that are not well connected by flight routes (Givoni & Banister, 2006). This suggests that a well-developed public transport interchange can play an important role in a hub-and-spoke network and foster seamless transfer between modes. Essentially, the transfer experience can affect the overall success of public transport to deliver the desired environmental benefits. As suggested by a wide range of literature (Iseki & Taylor, 2010; Guo & Wilson, 2011; Shrivastava & O’Mahony, 2009), transfers can generate different negative utilities such as uncertain waiting time and longer travel time. The negative factors can hinder the development of transport decarbonisation, as a low level of multi-modal integration can lead to lower willingness to use public transport and lower public transport ridership (see also Chapter 32). Essentially, an integrated public transport system is not only about the physical integration of different transport modes at the same location but rather the integration of a wide range of attributes that affect the overall experience during transfer (Li & Loo, 2016). A peoplecentric design is essential. Preston (2012) proposed the concept of the “integration ladder” and highlighted that integration is a multifaceted concept that incorporates horizontal integration (e.g. information, physical integration, fares and ticketing and infrastructure) and vertical integration (e.g. passenger and freight, transport authorities, transport and land-use

100

Public transport and the environment

planning and policies of transport and other sectors). This pinpoints the out-of-fleet experience during the journey which is conducive to modal shift and public transport ridership. Moreover, integration should consider the coordination between different transport modes so that the benefits of the large-capacity mass system can be delivered (Loo, 2020). The idea of integrating different mass transport modes into a comprehensive multi-modal “sustainable transport hierarchy” should be advocated. The hierarchy indicates the different orders of public transport, as defined by their carrying capacity and environmental impact per passenger. For example, commuters may take intercity trains for regional travel (higher-order) and take buses or paratransit (lower-order) as the last-mile journey to reach the final destination at local communities.

Integration of public and active transport The linkage between public and active transport is essential to further promote non-motorised travel and support TOD. First, TOD requires a well-connected pedestrian network linking the public transport locations to nearby activity nodes (see also Chapter  6). Encouraging active transport can produce significant environmental advantages, as increasing walkability in the neighbourhoods can alleviate environmental harms per unit of transport volume (Loo & Tsoi, 2018). It has been emphasised that proximity to activity locations from public transport locations is important; however, proximity does not necessarily mean a shorter walking distance, as the street connectivity, junction density, road-crossing facilities and size of street blocks are all relevant parameters in affecting walking distance. Most importantly, walkability is not merely about walking distance but different aspects, such as safety, convenience and comfort of the pedestrian experience (Loo & Lam, 2012). Walking experience is about how people interact with the built environment and street landscape (Loo et al., 2017b; Wang et  al., 2016). In relation, land-use diversity, residential density, street connectivity and aesthetics can affect walkability. Essentially, the walking experience between public transport locations or between activity points and public transport locations needs to be further investigated. The interface of the public transport and pedestrian fabrics is an important construct of sustainable mobility (Newman et  al., 2016). Hence, a microscale assessment of walkability within the public transport station and around the station is required to bridge the pedestrian network and public transport network. The common parameter of walkability evaluation around the public transport station is 500 metres (Loo et al., 2010, 2017a). Moreover, interchange stations sometimes involve multilevel structures, with bridges and tunnels connecting to the concourse, platforms, waiting areas and nearby buildings. Therefore, the walkability assessment should also integrate the three-dimensional network (i.e. at-grade, elevated walkway and underground walkway). Bicycle-train integration that strengthens the linkage between public and active transport is also a popular measure to sustain public transport ridership (see also Chapter 31). In a case study of the Netherlands (Geurs et al., 2016), it was found that bicycle-train integration (such as better bicycle routes and parking, travel time and cost reductions) was essential to train ridership and job accessibility, especially at large stations with multiple train lines. Zhao and Li (2017) also examined the integration of bicycles and train stations in Beijing and found that the number of public bicycles, proximity to the city centre and the presence of public parks were associated with a higher level of cycling to metro stations. The acceptable range of cycling between the public transport locations and the activity nodes was also found to be around 1 and 5 km in Beijing (Zhao & Li, 2017).

101

Ka Ho Tsoi and Becky P.Y. Loo

Public transport electrification Electric mobility has been a recent initiative in transport decarbonisation. Indeed, electrifying public transport fleets can offer potential carbon savings in public transport operations. Recently, there has been research exploring the relationship between electrified/hybrid public transport and carbon emissions. Sánchez et al. (2013) conducted a lifecycle assessment of different fuel types of buses, including fuel cell hybrid, hybrid-diesel-electric, battery electric and combustion ignition engine. The findings indicate that the fuel cell hybrid and battery electric buses can potentially reduce 26 and 28 percent of energy consumption, with an equivalent carbon reduction by 29 and 31 percent. Lajunen and Lipman (2016) investigated the lifecycle costs of CO2 emission of buses in two case scenarios of Finland and the United States and found that the energy efficiency of city buses can be improved by alternative powertrain technologies. When compared with traditional diesel buses, hybrid and full-electric buses can potentially reduce CO2 emission up to 75 percent (Lajunen & Lipman, 2016). Dreier et al. (2018) conducted a case study in the BRT system in Brazil. The findings indicate that the adoption of hybrid bus and plug-in hybrid city buses can reduce 30 and 75 percent of fuel energy per distance when compared to a traditional bus, which is equivalent to around 1.12 kg of CO2 well-to-wheel per kilometre (WTW/km) and 1.54 kg of CO2e WTW/km. Ribau et al. (2014) investigated the lifecycle impact difference between hybrid electric and plug-in hybrid electric cars. They suggest that fuel cell buses can reduce the overall energy consumption by 58 percent and produce a two-thirds decrease in the CO2 equivalent throughout the lifecycle. With the advancement of technologies, the electrification of public transport becomes more feasible. However, several challenges may hinder the development of such initiatives, including the lack of charging infrastructure, prohibitive costs for operators and uncertainty in investment (Gallo, 2016; Loo, 2018). Multiple policy instruments need to be considered for the success of promoting low- or zero-emission vehicles. In a European project, “Electrification of public transport in cities” under Horizon 2020, three thematic pillars were identified – integration of electric buses, energy storage and multipurpose use of infrastructure (Glotz-Richter & Koch, 2016). An electric bus demonstration study by Miles and Potter (2014) illustrated that a new company can be developed to purchase electric buses and chargers can be beneficial. The company is responsible for leasing the fleet to the bus operators and maintaining the charging infrastructure (Miles & Potter, 2014). This is to reduce the financial burden and the risks entirely shared by the operators and increase the feasibility of putting the electrified fleet into practice.

Conclusion In this chapter, the environmental implications of public transport on carbon emissions, air quality and climate change have been discussed. Developing a public transport system has been a common strategy to encourage sustainable travel in the past several decades. The common initiatives include transit-oriented development, integrated public transport systems, integration of public and active transport and public transport electrification. Though there are certain geographical discrepancies, these policy instruments can generally pave the way to transport decarbonisation by encouraging modal shift, reducing car travel distance and reducing transport energy intensities. To ensure that public transport can maximise its potential in carbon reductions, future research should examine the complementary effects of integrating different public transport modes in a network. Essentially, establishing a sustainable public transport hierarchy seems a promising trend – integrating different orders of public transport can help strengthen the mode shift initiatives and enhance the resilience of public transport networks. Moreover, 102

Public transport and the environment

the synergies provided by public transport and other policy instruments need to be further investigated. Since this chapter was devised, the COVID-19 pandemic outbreak has emerged and has impacted individual physical mobility which has had a knock-on effect on the environment. Indeed, reduced surface transport and fuel consumption, in particular by private vehicles, have manifested the potential environmental benefits. According to a global study of over 69 countries (Le Quéré et al., 2020), there was a reduction of activities in surface transport by 40 to 65 percent when compared to the pre-COVID-19 situation. The corresponding reduction of global emission ranges from 7.5 to 36 percent (equivalent to a decrease of 5.9 to 9.6 million metric tonnes CO2), which is the largest contributor to the total emission change (even higher than the aviation sector). However, the COVID-19 pandemic has also posed significant challenges to public transport. Since early 2020, public transport patronage has dropped significantly in different parts of the world. The Citymapper urban mobility index (Citymapper, 2020), which indicates the use of public transport, has rapidly declined in many big cities worldwide during COVID-19. Some cities had less than 10 percent of the mobility index they had previously. Even when some confinement measures have been loosened since mid-2020, the mobility index in most cities stayed at around one-third of the usual figures. With the COVID-19 associated health risk and uncertainty, adaptive strategies need to be implemented in public transport to enhance its resilience so that the environmental benefits can be delivered. In particular, this pandemic may encourage people to use private transport to reduce public interactions. Notwithstanding, a public transit society is not inconsistent with controlling public health risk, with public transport being a safe and sustainable travel option under a compact city environment (Cowling et al., 2020).

References Anderson, M. L. (2014). Subways, strikes and slowdowns: The impacts of public transit on traffic congestion. American Economic Review, 104(9), 2763–2796. Asian Development Bank. (2010). Reducing carbon emissions from transport projects. Asian Development Bank. Barth, M., & Boriboonsomsin, K. (2008). Real-world carbon dioxide impacts of traffic congestion. Transportation Research Record, 2058(1), 163–171. Bartholomew, K., & Ewing, R. (2008). Land use-transportation scenarios and future vehicle travel and land consumption: A meta-analysis. Journal of the American Planning Association, 75(1), 13–27. Basagaña, X., Triguero-Mas, M., Agis, D., Pérez, N., Reche, C., Alastuey, A., & Querol, X. (2018). Effect of public transport strikes on air pollution levels in Barcelona (Spain). Science of the Total Environment, 610, 1076–1082. Bauernschuster, S., Hener, T., & Rainer, H. (2017). When labor disputes bring cities to a standstill: The impact of public transit strikes on traffic, accidents, air pollution and health. American Economic Journal: Economic Policy, 9(1), 1–37. Baum-Snow, N., Kahn, M. E., & Voith, R. (2005). Effects of urban rail transit expansions: Evidence from sixteen cities, 1970–2000. Brookings-Wharton Papers on Urban Affairs, 147–206. Beaudoin, J., Farzin, Y. H., & Lawell, C. Y. C. L. (2015). Public transit investment and sustainable transportation: A review of studies of transit’s impact on traffic congestion and air quality. Research in Transportation Economics, 52, 15–22. Beaudoin, J., & Lawell, C. Y. C. L. (2017). The effects of urban public transit investment on traffic congestion and air quality. In H. Yagohoubi (Ed.), Urban transport systems (pp. 111–122). InTech. Bertolini, L., Curtis, C., & Renne, J. (2012). Station area projects in Europe and beyond: Towards transit oriented development? Built Environment, 38(1), 31–50. Bureau of Transport Statistics. (2018). National transportation statistics. Bureau of Transport Statistics. Cervero, R. (2004). Transit-oriented development in the United States: Experiences, challenges and prospects (Vol. 102). Transportation Research Board.

103

Ka Ho Tsoi and Becky P.Y. Loo Cervero, R., & Kockelman, K. (1997). Travel demand and the 3Ds: Density, diversity and design. Transportation Research Part D, Transport and Environment, 2(3), 199–219. Cervero, R.,  & Sullivan, C. (2011). Green TODs: Marrying transit-oriented development and green urbanism. International Journal of Sustainable Development and World Ecology, 18(3), 210–218. Chapman, L. (2007). Transport and climate change: A  review. Journal of Transport Geography, 15(5), 354–367. Chavez-Baeza, C.,  & Sheinbaum-Pardo, C. (2014). Sustainable passenger road transport scenarios to reduce fuel consumption, air pollutants and GHG (greenhouse gas) emissions in the Mexico City metropolitan area. Energy, 66, 624–634. Chen, Y., & Whalley, A. (2012). Green infrastructure: The effects of urban rail transit on air quality. American Economic Journal: Economic Policy, 4(1), 58–97. Chester, M., & Horvath, A. (2009). Life-cycle energy and emissions inventories for motorcycles, diesel automobiles, school buses, electric buses, Chicago rail and New York City rail. UC Berkeley Center for Future Urban Transport. Chowdhury, S., & Ceder, A. A. (2016). Users’ willingness to ride an integrated public-transport service: A literature review. Transport Policy, 48, 183–195. Citymapper. (2020). Citymapper mobility index. https://citymapper.com/cmi Cowling, B. J., Ali, S. T., Ng, T. W., Tsang, T. K., Li, J. C., Fong, M. W., Liao, Q., Kwan, M. Y. W., Lee, S. L., Chiu, S. S., Wu, J. T., Wu, P., & Leung, G. M. (2020). Impact assessment of non-pharmaceutical interventions against coronavirus disease 2019 and influenza in Hong Kong: An observational study. The Lancet Public Health, 5, 279–288. Creutzig, F., Mühlhoff, R., & Römer, J. (2012). Decarbonizing urban transport in European cities: Four cases show possibly high co-benefits. Environmental Research Letters, 7(4). doi:10.1088/1748-9326/7/4/044042 Crippa, M., Oreggioni, G., Guizzardi, D., Muntean, M., Schaaf, E., Lo Vullo, E., Solazzo, E., MonfortiFerrario, F., Olivier, J. G. J., & Vignati, E. (2019). Fossil CO2 and GHG emissions of all world countries – 2019 report. Publications Office of the European Union. Curtis, C. (2012). Delivering the “D” in transit-oriented development: Examining the town planning challenge. Journal of Transport and Land Use, 5(3), 83–99. Davis, T., & Hale, M. (2007). Public transportation’s contribution to US greenhouse gas reduction. Science Applications International Corporation. Dreier, D., Silveira, S., Khatiwada, D., Fonseca, K. V., Nieweglowski, R., & Schepanski, R. (2018). Wellto-wheel analysis of fossil energy use and greenhouse gas emissions for conventional, hybrid-electric and plug-in hybrid-electric city buses in the BRT system in Curitiba, Brazil. Transportation Research Part D: Transport and Environment, 58, 122–138. Ewing, R., & Cervero, R. (2010). Travel and the built environment: A meta-analysis. Journal of the American Planning Association, 76(3), 265–294. Fu, M., & Kelly, J. A. (2012). Carbon related taxation policies for road transport: Efficacy of ownership and usage taxes and the role of public transport and motorist cost perception on policy outcomes. Transport Policy, 22, 57–69. Gallo, J. B. (2016). Electric truck and bus grid integration, opportunities, challenges and recommendations. World Electric Vehicle Journal, 8(1), 45–56. Geurs, K. T., La Paix, L., & Van Weperen, S. (2016). A multi-modal network approach to model public transport accessibility impacts of bicycle-train integration policies. European Transport Research Review, 8(4), 25. Givoni, M., & Banister, D. (2006). Airline and railway integration. Transport Policy, 13(5), 386–397. Glotz-Richter, M., & Koch, H. (2016). Electrification of public transport in cities (horizon 2020 ELIPTIC project). Transportation Research Procedia, 14, 2614–2619. Guo, Z., & Wilson, N. H. (2011). Assessing the cost of transfer inconvenience in public transport systems: A  case study of the London Underground. Transportation Research Part A: Policy and Practice, 45(2), 91–104. Harford, J. D. (2006). Congestion, pollution and benefit-to-cost ratios of US public transit systems. Transportation Research Part D: Transport and Environment, 11(1), 45–58. International Energy Agency. (2019). CO2 emissions from fuel combustion highlights. OECD Publishing. Iseki, H., & Taylor, B. D. (2010). Style versus service? An analysis of user perceptions of transit stops and stations. Journal of Public Transportation, 13(3), 2. Kamruzzaman, M. D., Baker, D., Washington, S., & Turrell, G. (2013). Residential dissonance and mode choice. Journal of Transport Geography, 33, 12–28.

104

Public transport and the environment Kwan, S. C., & Hashim, J. H. (2016). A review on co-benefits of mass public transportation in climate change mitigation. Sustainable Cities and Society, 22, 11–18. Lajunen, A., & Lipman, T. (2016). Lifecycle cost assessment and carbon dioxide emissions of diesel, natural gas, hybrid electric, fuel cell hybrid and electric transit buses. Energy, 106, 329–342. Lalive, R., Luechinger, S., & Schmutzler, A. (2018). Does expanding regional train service reduce air pollution? Journal of Environmental Economics and Management, 92, 744–764. Lee, S., Lee, S., Park, D., & Lee, C. (2010). Analysis of travel patterns between road and transit-oriented development areas. Journal of Advanced Transportation, 44(1), 3–10. Le Quéré, C., Jackson, R. B., Jones, M. W., Smith, A. J., Abernethy, S., Andrew, R. M., De-Gol, A. J., Willis, D. R., Shan, Y., Canadell, J. G., Friedlingstein, P., & Peters, G. P. (2020). Temporary reduction in daily global CO2 emissions during the COVID-19 forced confinement. Nature Climate Change, 10, 647–653. Li, L., & Loo, B. P. Y. (2016). Towards people-centered integrated transport: A case study of Shanghai Hongqiao comprehensive transport hub. Cities, 58, 50–58. Litman, T., & Burwell, D. (2006). Issues in sustainable transportation. International Journal of Global Environmental Issues, 6(4), 331–347. Litman, T., & Steele, R. (2017). Land use impacts on transport. Victoria Transport Policy Institute. Loo, B. P. Y. (2018). Unsustainable transport and transition in China. Routledge. Loo, B. P. Y. (2020). Transport, urban. In A. Kobayashi (Ed.), International encyclopaedia of human geography (2nd ed., Vol. 13, pp. 457–462). Elsevier. Loo, B. P. Y., & Banister, D. (2016). Decoupling transport from economic growth: Extending the debate to include environmental and social externalities. Journal of Transport Geography, 57, 134–144. Loo, B. P. Y., Chen, C., & Chan, E. T. (2010). Rail-based transit-oriented development: Lessons from New York city and Hong Kong. Landscape and Urban Planning, 97(3), 202–212. Loo, B. P. Y., Cheng, A. H., & Nichols, S. L. (2017a). Transit-oriented development on greenfield versus infill sites: Some lessons from Hong Kong. Landscape and Urban Planning, 167, 37–48. Loo, B. P. Y., & du Verle, F. (2017). Transit-oriented development in future cities: Towards a two-level sustainable mobility strategy. International Journal of Urban Sciences, 21(Supp 1), 54–67. Loo, B. P. Y., & Lam, W. W. Y. (2012). Geographic accessibility around health care facilities for elderly residents in Hong Kong: A microscale walkability assessment. Environment and Planning B: Planning and Design, 39(4), 629–646. Loo, B. P. Y., Lam, W. W. Y., Mehandran, R., & Katagiri, K. (2017b). How is the neighborhood environment related to the health of seniors in Hong Kong, Singapore and Tokyo? Some insights for promoting aging-in-place. Annals of the American Association of Geographers, 107(4), 812–828. Loo, B. P. Y., & Tsoi, K. H. (2018). The sustainable transport pathway. Journal of Transport and Land Use, 11(1), 961–980. McKinley, G., Zuk, M., Höjer, M., Avalos, M., González, I., Iniestra, R., Laguna, I., Martínez, M. A., Osnaya, P., Reynales, L. M., Valdés, R., & Martínez, J. (2005). Quantification of local and global benefits from air pollution control in Mexico City. Environmental Science and Technology, 39(7), 1954–1961. Miles, J., & Potter, S. (2014). Developing a viable electric bus service: The Milton Keynes demonstration project. Research in Transportation Economics, 48, 357–363. Nelson, P., Baglino, A., Harrington, W., Safirova, E., & Lipman, A. (2007). Transit in Washington, DC: Current benefits and optimal level of provision. Journal of Urban Economics, 62(2), 231–251. Newman, P., Kosonen, L., & Kenworthy, J. (2016). Theory of urban fabrics: Planning the walking, transit/ public transport and automobile/motor car cities for reduced car dependency. Town Planning Review, 87(4), 429–458. ODYSSEE-MURE. (2019). Sectoral profile – transport. ODYSSEE-MURE. OECD. (1996). Towards sustainable transportation proceedings. OECD Publishing. Ogra, A., & Ndebele, R. (2014, October 31–November 2). The role of 6Ds: Density, diversity, design, destination, distance and demand management in transit-oriented development (TOD). Proceedings of the NeoInternational Conference Habitable Environment. Preston, J. (2012). Integration for seamless transport. International Transport Forum. Renne, J. L. (2009). From transit-adjacent to transit-oriented development. Local Environment, 14(1), 1–15. Ribau, J. P., Silva, C. M., & Sousa, J. M. (2014). Efficiency, cost and life cycle CO2 optimization of fuel cell hybrid and plug-in hybrid urban buses. Applied Energy, 129, 320–335. Sánchez, J. A. G., Martínez, J. M. L., Martín, J. L., Holgado, M. N. F., & Morales, H. A. (2013). Impact of Spanish electricity mix, over the period 2008–2030, on the life cycle energy consumption and GHG

105

Ka Ho Tsoi and Becky P.Y. Loo emissions of electric, hybrid diesel-electric, fuel cell hybrid and diesel bus of the Madrid transportation system. Energy Conversion and Management, 74, 332–343. Shin, Y. E., Vuchic, V. R., & Bruun, E. C. (2009). Land consumption impacts of a transportation system on a city: An analysis. Transportation Research Record, 2110(1), 69–77. Shrivastava, P., & O’Mahony, M. (2009). Modeling an integrated public transportation system-a case study in Dublin, Ireland. European Transport, 41, 28–46. Sims, R., Schaeffer, R., Creutzig, F., Cruz-Núñez, X., D’Agosto, M., Dimitriu, D., Figueroa Meza, M. J., Fulton, L., Kobayashi, S., Lah, O., McKinnon, A., Newman, P., Ouyang, M., Schauer, J. J., Sperling, D., & Tiwari, G. (2014). Transport. In O. Edenhofer, R. Pichs-Madruga, Y. Sokona, E. Farahani, S. Kadner, K. Seyboth, A. Adler, I. Baum, S. Brunner, P. Eickemeier, B. Kriemann, J. Savolainen, S. Schlömer, C. von Stechow, T. Zwickel, & J. C. Minx (Eds.), Climate change 2014: Mitigation of climate change. contribution of working group III to the fifth assessment report of the intergovernmental panel on climate change (pp. 599–670). Intergovernmental Panel on Climate Change. Sun, C., Zhang, W., Fang, X., Gao, X., & Xu, M. (2019). Urban public transport and air quality: Empirical study of China cities. Energy Policy, 135. doi:10.1016/j.enpol.2019.110998 Sung, H., & Oh, J. T. (2011). Transit-oriented development in a high-density city: Identifying its association with transit ridership in Seoul. Korea. Cities, 28, 70–82. Titos, G., Lyamani, H., Drinovec, L., Olmo, F. J., Močnik, G., & Alados-Arboledas, L. (2015). Evaluation of the impact of transportation changes on air quality. Atmospheric Environment, 114, 19–31. UITP (International Association of Public Transport). (2017). Statistics brief. UITP. Ülengin, F., Önsel, Ş., Topçu, Y. I., Aktaş, E., & Kabak, Ö. (2007). An integrated transportation decision support system for transportation policy decisions: The case of Turkey. Transportation Research Part A: Policy and Practice, 41(1), 80–97. U.S. Department of Energy. (2018). Average per-passenger fuel economy by travel mode. Retrieved March 1, 2020, from https://afdc.energy.gov/data/categories/fuel-consumption-and-efficiency Wang, Y., Chau, C. K., Ng, W. Y., & Leung, T. M. (2016). A review on the effects of physical built environment attributes on enhancing walking and cycling activity levels within residential neighborhoods. Cities, 50, 1–15. Winston, C., & Langer, A. (2006). The effect of government highway spending on road users’ congestion costs. Journal of Urban Economics, 60(3), 463–483. Woodcock, J., Edwards, P., Tonne, C., Armstrong, B. G., Ashiru, O., Banister, D., Beevers, S., Chalabi, Z., Chowdhury, Z., Cohen, A., Franco, O. H., Haines, A., Hickman, R., Lindsay, G., Mittal, I., Mohan, D., Tiwari, G., Woodward, A.,  & Roberts, I. (2009). Public health benefits of strategies to reduce greenhouse-gas emissions: Urban land transport. The Lancet, 374(9705), 1930–1943. Wright, L., & Fulton, L. (2005). Climate change mitigation and transport in developing nations. Transport Reviews, 25(6), 691–717. Zhang, K., & Batterman, S. (2013). Air pollution and health risks due to vehicle traffic. Science of the Total Environment, 450, 307–316. Zhang, M. (2010). Can transit-oriented development reduce peak-hour congestion? Journal of the Transportation Research Board, 2174, 148–155. Zhao, P., & Li, S. (2017). Bicycle-metro integration in a growing city: The determinants of cycling as a transfer mode in metro station areas in Beijing. Transportation Research Part A: Policy and Practice, 99, 46–60.

106

9 COMMUNITY PARTICIPATION IN PUBLIC TRANSPORT DECISION-MAKING Lake Sagaris

Introduction: sustainability, transport justice and citizen engagement Citizen involvement in public transport and other mobility planning decisions has evolved substantially in recent years and is likely to receive more attention in the wake of major protest movements triggered by mobility-related issues. These protests have included major national revolts against a bus fare increase (Brazil 2013), diesel taxes (Paris 2018), rising gasoline prices (Ecuador, 2019) and, most recently and dramatically, a rise in Metro fares, in Chile, 18 October 2018. The Chilean case catalyzed a “perfect storm” of diverse social movements, all frustrated by lack of action on specific issues, blocked by the rigidities of the 1980 constitution, created by military dictator Augusto Pinochet. Modest reforms over the past 30 years did little to change the underlying antidemocratic institutions generated by that constitution, leading to demonstrations. In all of these cases, mobility issues became the lightning rod for broad, national social movements and protests that went beyond specific mobility concerns, suggesting that what is at stake could be broader and more challenging than has been assumed to date in studies of citizen participation (the Latin American term) or citizen engagement (the term most frequently used in English). Many views of participation fit well within more general planning theories focusing on communications and collaboration (Healey, 1997, 2006; Innes & Booher, 1999), although critics note that power imbalances raise the potential for manipulation and exclusion (Cooke & Kothari, 2001). In one study, for example, although participatory methods came to the fore during efforts to take transport policy in new directions, “evidence of substantive impacts remained sparse” (Bickerstaff et  al., 2002, p.  61). A  closer look, moreover, found “a deeply problematic relationship between citizen involvement and established structures of democratic decision-making” (Bickerstaff & Walker, 2005, p. 2123), noting the problem of “consultation fatigue” (p. 2135). This chapter examines current thinking about citizen engagement (or participation) in the face of major challenges for both individual and sociopolitical transformation, associated with dealing effectively with climate change and other issues relating to sustainability. As the upsurge in social movements, mentioned previously, suggests, social justice may not be central to current 107

Lake Sagaris

definitions of social sustainability but is clearly high on citizens’ agendas, particularly when conflicts occur. The COVID-19 pandemic has further raised the stakes, shortening timeframes for response and requiring combined actions that favor social practices (distancing, masks) and management changes (improving air circulation in public transport, integrating cycling more fully as a feeder mode). These make adequate participatory systems, rather than one-off, projectcentered events, increasingly central to the rapid detection of problems and the formulation of responses acceptable to the relevant user groups. Based on attempts to innovate in citizen participation in two Chilean cities (Santiago and Temuco), this chapter uses Flyvbjerg’s categories of knowledge to consider how engagement in public and other transport planning could be more effective if rooted in theories of democratization. The following section introduces concepts of knowledge as they emerge in social sciences and planning and considers how they interact with the perceived benefits and procedures currently applied to participation in public transport planning. The next section considers these issues in the context of Chile, a country rapidly transitioning toward car-centered urban planning amidst significant conflicts over social inclusion, justice and equity as they play out in transport and other arenas. Based on these experiences, the following section considers implications for deepening participation from a democratizing perspective, while the chapter concludes with final reflections.

Planning, participation and democratization for sustainability Studies of participation suggest considerable attention to procedural issues, rather than public influence on actual outcomes, whether measured in terms of final plans, projects and implementation or by citizen uptake of innovations and behavioral change. Giering (2011) defines participation as “the process through which transportation agencies inform and engage people in the transportation decision-making process” (my emphasis, p. 1). These are important definitions, since many real-world procedures only inform, but fall far short of involving, members of the public in actual decision-making. Vigar (2006), Giering (2011) and others interested in participation cite multiple reasons for investing resources in well-designed, well-executed and influential participation (Table  9.1), noting three main modalities: political, technical-rational and collaborative (Vigar, 2006). Most participatory efforts focus primarily on providing information and gathering public input, points 2 to 4 in the table, in order to improve agency credibility and implementation, points 5 to 7. While more political processes tend to be hindered by the “push and pull of interests and perceptions of the powerful”, technical-rational processes often fail to reach out to the full spectrum of stakeholders. In contrast, collaborative approaches tend to focus on deliberative processes, in which a mutual learning process occurs, with considerable potential for personal as well as systemic transformation (Vigar, 2006, pp. 268–269). Diversity, mutual need (interdependency) and “authentic dialogue” are crucial ingredients for more collaborative processes to prosper (Booher & Innes, 2002). Most studies of participation focus on procedural rather than outcome issues. Giering, for example, surveyed over 100 transit agencies in Canada and the United States regarding their approaches to citizen engagement, identifying diverse motivations, mainly associated with the legitimacy of agencies themselves (Table 9.2). Vigar’s study, meanwhile, examined participation in a regional transport planning process in the United Kingdom, finding that participation contributed to awareness of and interest in a regional transport strategy-making effort. Results often fell short of both citizens’ and 108

Community and transport decision-making Table 9.1 Reasons for participation in transport planning 1 2 3 4 5 6 7

Democratic purposes Sharing and providing knowledge of others’ experience and local conditions, ‘situated knowledge’ Debating these various ‘knowledges’, developing awareness of associated policy complexity and facilitating learning associated with the problem at hand Generating shared ownership of strategies and programs, thus potentially reducing implementation deficits “Better” decisions that are sustainable and supportable and reflect community values Agency credibility Faster implementation

Source: Points 1–4 based on p. 267, Vigar, 2006; 4–7 based on Giering, 2011 Table 9.2 Elements important to public engagement in transport planning 1 2 3 4 5 6

The more public involvement, the more likely an agency is to judge the outcomes as successful. Determining the “right” questions to ask the public is important. Resources for public involvement are important but do not have to be strictly financial. The value that an agency places on public involvement is critical to its success. Openness and transparency matter and in many cases are most important to the public. Understanding, partnering with and empowering communities can significantly benefit public involvement efforts and the agency.

Source: Giering, 2011, p. 3

policymakers’ aspirations, however, leading to the conclusion that participation requires “an appropriate collaborative process if a policy mechanism is to endure” (Vigar, 2006, p. 264). As Tables 9.1 and 9.2 reveal, strategies in the 20th century typically focused on bringing public opinion into major plans, policies and projects and building public support for these initiatives. Today, however, researchers and policymakers increasingly call for transport to address a plethora of issues: improving health, eliminating barrier effects and encouraging active transport; reducing pollution and boosting energy efficiency, thereby reducing excessive dependency on fossil fuels and restoration of important eco-system services, such as urban vegetation to overcome the heat island and other negative effects. To the earlier concerns that engagement may not achieve significant outcomes, new understanding of mechanisms influencing individual and collective behavior increases the challenges. A recent review of climate change engagement concludes that “information and understanding are not the most important drivers of behavior change, and in many instances – for example, in much of our habitual behavior  – are hardly relevant at all” (Whitmarsh et  al., 2011a, pp. xvi–xvii). Indeed, previous efforts to achieve behavioral change, based on relatively simple models of values, choices and behaviors, indicate these are not linear but rather comprise “complex behavioral ecologies” (Whitmarsh et al., 2011b, p. 3). The so-called “information deficit model, which assumes that the public are ‘empty vessels’ waiting to be filled with information which will propel them into rational action, has implicitly underpinned much public policy, but is widely criticized as inappropriate and ineffective” (Whitmarsh et al., 2011b, pp. 3–4). Habits, which make previous behavior unconscious and therefore automatic, are particularly impervious to social marketing campaigns, largely because very few of the recipients pay attention, undermining any hope that people will process and interiorize the information they 109

Lake Sagaris

contain. Even those who do pay attention are unlikely to change their behavior, given the “strong contextual drivers and automatic processes underpinning habits” (Verplanken, 2011, p.  23). Catching people at specific points in the life cycle with precise interventions may be more effective, as occurs when people who have recently moved receive a free ticket to encourage them to use public transport, along with information about services (Verplanken, 2011, p. 25). In contrast to mass marketing and similar campaigns, attempts to improve engagement with sustainability and climate change, based on four types of environmentally significant individual behavior, have shown more promise. This perspective considers four categories of behavior, the first two occurring in the public and last two in the private sphere: activism, such as participating in a demonstration; non-activist behavior, such as signing a petition or joining an advocacy group; personal commitment, such as recycling or using energy efficient goods at home and professional commitment, such as using sustainable materials or recommending investment solely in environmentally responsible firms (Höppner & Whitmarsh, 2011, pp. 48–49). Moreover, public behavior influences private behavior and vice versa, making it important to foster collective values that “enlarge the practice of active citizenship, to increase people’s efficacy, and to build trust and understanding between citizen and government” (Höppner & Whitmarsh, 2011, p. 49). In this sense, citizen engagement may be “invited”, as when the government calls for citizens to participate in a meeting, or autonomously created, by citizens and civil society organizations.1 Most discussion of engagement in public transport tends to focus on “invited” participation. A growing body of evidence, however, points to the importance of “autonomous” participation by civil society organizations (CSOs) as central to encouraging personal and professional behavioral change, together with collective efforts that carry out grassroots campaigns and other advocacy actions (Whitmarsh et  al., 2011a). The effectiveness of cycling advocacy contrasts with lack of public enthusiasm for bus rapid transit (BRT) and other forms of public transport, despite the enormous resources invested in positioning BRT as an efficient strategy for substantially improving public transport in a sustainability framework (Aldred, 2013; Buehler & Pucher, 2021; Sagaris, 2014, 2015) (see also Chapter 13). Deeper participation speaks more directly to the “democratic purposes” mentioned in Table 9.1. Barber (1998) notes the importance of collective processes for “making society civil and democracy strong”. He defines civil society as the space in which “you” and “I” become “we”, arguing that “a free nation depends for its liberties . . . on a vibrant and pluralistic civil society” (Barber, 1998, p. 45). Other studies trace the relationship between democratic neighborhood associations and national cultures favorable to democratic values (Berry et al., 1993; Thomson, 2001), while specific work on antihighway movements in the United States and Canada reveal their relevance to shifting visions, policies and behavior toward greater sustainability (Hovey, 1998; Ladd, 2008; Mohl, 2012; Sirianni, 2009). More recent work from Europe underlines the relationship between neighborhood development and social innovation (Moulaert et al., 2010). These tendencies are consistent with collaborative planning theory (Healey, 1997), which argues that everyday life and local environments define economic dynamics, identities, networks and lifestyles. Of particular relevance to sustainability transitions is Healey’s concept of “strategic conviction”, which conceives of strategy-making as a process of “deliberative paradigm change” that seeks to alter cultural conceptions, systems of understanding and systems of meaning. It is about more than just producing collective decisions. It is about shifting and re-shaping 110

Community and transport decision-making

convictions. Such re-framing efforts aim to influence the allocation of resources within a governance sphere. (Healey, 1997, p. 245) The main conclusions from reframing become “embedded in the thoughtworlds” of key players, entering into the “flow of routine rather than through deliberate mobilization and struggle” (Healey, 1997, p. 245). How we travel on a daily basis, whether to daycare, work, to shop or to a medical appointment, reflects these habitual patterns that shift only when significant changes occur. Altering habits requires “a considerable effort of political will” (Healey, 1997, p. 247), which in turn requires collective spaces for diverse deliberations. The following section explores these intersections between citizen engagement in transport planning, democratization and goals of individual/collective transformation favoring more sustainable behavior and policy in the context of two Chilean cities. It also considers how to evaluate process and outcomes, using a framework developed by Innes and Booher, summarized by Vigar (2006).

Recent experiences in participation in Santiago and Temuco, Chile Chile is an interesting country to study these issues: by and large, although the economy has grown steadily in recent decades, so has inequality and with it the overwhelming presence of cars on city streets, public squares and as the central priority in transportation planning, exercised mainly by three highly centralized ministries: the transportation ministry, which is responsible for public transport; the housing ministry, which interacts with local municipal governments that have led planning favorable to walking and cycling; and the ministry of public works, which handles enormous resources, much of which have subsidized major highway concessions and other similar infrastructure. While participation is relatively well established in environmental impact assessments, urban projects generally and transport projects specifically have been largely exempted from evaluations and public participation requirements. Notwithstanding, citizen demands have pressured the authorities enough to ensure some efforts to innovate in engagement methods and procedures, particularly in Metropolitan Santiago, the national capital, concentrating 40% of the country’s population and its wealth, measured by GDP, and Temuco-Padre Las Casas, in one of the country’s poorest regions, La Araucanía, 9 hours drive south of Santiago (for a more extensive discussion of these two processes, see Sagaris, 2018).

The New Alameda Providencia bus rapid transit project, Santiago (2015–2017) In Santiago, a civil society organization, Casa de la Paz, led significant innovations in participation in a BRT-based project, the New Alameda-Providencia (NAP), along the city’s most important road, the Alameda de las Delicias. The regional government, which officially has no authority over transport planning, co-ordinated this emblematic project in an effort to win people’s hearts by remaking transportation and public spaces. A central goal was to overcome the prevailing tendency of applying cheap, ugly solutions in low-income neighborhoods versus expensive, green and aesthetically appealing solutions in wealthy areas, thereby transforming a transport system widely perceived as reinforcing inequalities and exclusion. NAP began with a major international competition and jury, which selected a winning proposal based primarily on initial design sketches rather than operating criteria. Participation 111

Lake Sagaris

occurred almost simultaneously with detailed design and engineering development, slowing progress but lending credibility to promises that citizens’ opinions would genuinely influence final results. The participatory process consisted of informational and participatory activities that included thematic (cycling, public space, bus stop design) and territorial (specific tranches of the line and their respective neighborhoods and municipal zones) activities; field visits; participatory mapping and the elaboration of critiques, debates and proposals. The “reverse traffic pyramid” (BicycleInnovationLab, 2011) was heartily endorsed, as it identifies walking and, in this case, universal access as top priority for resources, followed by cycling, public transport, cargo and, in last place, cars. In addition to these modal priorities, citizens expressed strong support for preserving existing green space and improving and planting trees and other forms of vegetation. Transport engineers and related planners tended to treat the project as a bus corridor, while architects and citizens’ organizations argued for making the route more of a boulevard connecting beautiful and significant public spaces. One of the most significant achievements occurred when citizens pressured, neighbors mobilized and the government responded in favor of maintaining, rather than expropriating, a group of houses in the low-income Estación Central neighborhood. Citizens’ proposals were taken up within a series of template designs for specific typologies along the route, a controversial procedure that never truly gave pedestrians and green space the priority that citizens kept demanding. Experimentation with an innovative “Citizens Observatory”, intended to generate more permanent involvement in the project, floundered amidst contradictory mandates and, ultimately, the decision that representatives would be volunteers and individuals rather than representatives elected by and accountable to specific thematic and territorial communities. Despite these efforts, the project struggled to remain afloat as the regional government attempted to mobilize citizen participation while conducting behind-the-scenes negotiations with municipal governments. NAP could not overcome its identity as a project of the Bachelet government (2014–2018) and struggled to remain afloat when first municipal (2016) and then national elections (2017) swept pro-Bachelet forces from power. In 2018, the NAP was unceremoniously canceled by the Piñera government (2018 to present), as being “too expensive”, although it was actually one of the cheapest transport projects on a roster that included US$2.5 billion for each new Metro line and major highway expansion in the city.

“Permanent” participation for the regional transport plan in Temuco-Padre Las Casas In Temuco-Padre Las Casas (2016–2018), the author’s Laboratory for Social Change codesigned a participatory process with the regional transport authorities. They wanted to incorporate the views, needs and aspirations of local communities into their recently developed regional transport plan and generate a permanent system for citizen participation. The process took participation to local communities throughout the region, most of which had had no prior opportunity to exchange opinions directly with decision-makers, whether the political appointee in charge of the regional transport planning secretariat or technical staff involved in day-to-day implementation. Participatory sessions started with a plenary, which provided information on the process and the plan, then focused on small group deliberations that took place as part of a participatory mapping exercise. A substantial number of people with diverse disabilities and caregivers brought issues associated with lack of universal access and basic service guarantees into focus, 112

Community and transport decision-making

leading to support for solutions such as bicycle taxi circuits to connect bus stops with the regional hospital or primary health care facilities that were difficult for many to access. As with NAP, participants took up the reverse traffic pyramid with a passion, calling for more and better inclusion of walking and cycling as integral elements within the regional transport plan. They also reported considerable discrimination, particularly against school children who pay lower fares, and other negative treatment by bus operators, who are privately owned firms, with few incentives to respond to level of service and other standards. The process culminated in an international seminar attended by 200 people and a report with objectives, recommendations and aspirations as defined during the participatory process. A relatively simple institutional innovation was recommended to make participation ongoing: a citizens’ roundtable, a governmental roundtable and a combined roundtable, each of which would meet regularly to identify priorities and monitor progress. The final report also recommended at least one appropriately trained staff person to co-ordinate roundtables and reach out to local communities and connect them to regional transport authorities on an ongoing basis. As with NAP, however, national elections led to changes in the regional transport secretariat, and although most of the public transport staff remained, recommendations for the participatory system were not implemented, although staff did integrate many of the observations into regular work.

Evaluating participation in the two cases: procedure and outcomes At the practical level, there is much confusion about the types of public participation, particularly the way power is distributed among the different actors. In both cities, participation was “invited”, leaving most of the power in governmental hands. Notwithstanding, authorities made a significant attempt to distribute power more evenly, reaching out to specific communities that are often ignored in traditional approaches to participation. Organized groups were encouraged to participate, and the attendance sheets used to monitor participation revealed a strong presence from neighborhood associations, cycling, low-income commerce and other groups and of people of diverse ages, both women and men, and people from different parts of the city. Nonetheless, both the Citizens Observatory (Santiago’s NAP) and the Roundtables (Temuco-PLC Regional Plan) suffered from lack of a deeper organic link with specific organizations and leaders and with decision-makers themselves. In contrast, pro-cycling and walking advocacy that involved strong organizations with committed leadership achieved a shift toward more cycle-inclusive urban planning, bringing diverse perspectives and new knowledge, along with providing greater continuity and credibility to the process (Sagaris, 2015). Mobilizing diverse forms of credibility was central, given the general lack of participation, which had generated enormous distrust between participants and government officials. In the opening plenary for NAP, for example, citizen participants almost left en masse, after political leaders made their speeches then left the meeting for a press conference, without listening to citizen perspectives. Similarly, in Temuco-PLC, participants were so frustrated during opening presentations that the formal presentations were suspended mid-meeting and replaced by a frank, unstructured exchange between neighbors and the head of the regional transport secretariat, relieving tensions enough to hold a lively and productive collective mapping session. Table 9.3 positions these two cases on a spectrum developed by the International Association for Participation (IAP), which defines five levels of participation. These start from merely providing information (one-way communication), moving through consultation (often achieved 113

114

Consult

Inform

Involve

NAP 2015–2017; Plan Regional Temuco-PLC Collaborate

Observatorio NAP 2015–2017; Roundtables Tco-PLC Empower

Cycling Master Plan (Santiago, 2007–2010)

DTPM, office within Regional government GORE Regional government GORE Regional government national transport ministry GORE, municipalities Small group interested in Broad public, users and Neighborhood associations, Neighborhood associations, proPrevious cell and leading public transport decisions, non-users pro-cycling, small business cycling, small business /PYME/ citizen urban transport mainly technical and /PYME/associations, associations, architects, engineers groups: Living academics architects, engineers City, Bicicultura, CicloLRecreovia To place final decisionTo partner with the public in To provide the public with To obtain public feedback To work directly with making in the hands of each aspect of the decision, on analysis, alter natives the public throughout balanced and objective the public. including the development and/or decisions. the process to ensure infor mation to assist of alter natives and the that public concer ns them in understanding identification of the prefer red and aspirations are the problem, alter natives, solution. consistently understood opportunities and/or and considered. solutions. We will implement what We will look to you for We will work with you We will keep you informed. We will keep you you decide. advice and innovation in to ensure that your infor med, listen to and for mulating solutions and concer ns and aspirations acknowledge concerns incor porate your advice and are directly reflected and aspirations and recommendations into the in the alter natives provide feedback decisions to the maximum developed and provide on how public extent possible. feedback on how public input influenced the input influenced the decision. decision.

National transport ministry

Transantiago 2008 to present

Transantiago 2000–2006

Source: Own elaboration, based on spectrum from IAP2 website (column headings), rows Goal and Promise, https://cdn.ymaws.com/www.iap2.org/resource/resmgr/ pillars/Spectrum_8.5x11_Print.pdf, accessed 24 February 2020

Explanation: Using IAP2 classifications for types of participation, we find Transantiago remains minimal, versus NAP, which did involve, but failed to influence decisions, contrasted with the Cycling Master Plan, initiated by citizens, which had a strong impact on public policies.

Promise to the public (IAP)

Goal (IAP)

Citizen

Gover nment

Examples

Table 9.3 IAP2 Spectrum of Public Participation applied to the case studies

Lake Sagaris

Community and transport decision-making

using surveys or advisory boards), involvement, collaboration and empowerment. This spectrum specifies the goals for each level of participation, along with the “promise to the public”. Using this spectrum to evaluate the NAP and Temuco-PLC experiences, it can be seen that both went beyond the traditional informational and consultative levels for the spectrum, attempting to involve participants, but neither fulfilled the promise that “we will implement what you decide”, a failure that, at least partly, reflected institutional arrangements rather than the specifics of each participatory process. Innes and Booher, reported in Vigar (2006), offer a framework enabling greater insight (Table 9.4). For the process criteria, both Chilean experiences did relatively well. For outcomes, however, results were weak, with the Temuco-PLC experience producing some high-quality agreements and creative ideas, while both produced significant but incomplete learning, social and political capital and results that stakeholders could accept. Two main factors were behind these weak outcomes: lack of continuity and lack of genuine shifts in the distribution of power, leadership and co-responsibility. Lack of continuity contributes to a lack of depth, a characteristic the author of this chapter has not seen associated with Table 9.4 Qualitative indicators for evaluating process and outcome applied to the case studies  

Process criteria: The Process

NAP

T-PLC

1

Includes representatives of all relevant groups and significantly different interests Is driven by a purpose and task that are real, practical and shared by group Is self-organizing, allowing participants to decide on ground rules, objectives, tasks, working groups and discussion topics Engages participants, keeping them at the table, interested, and learning through in-depth discussion, drama, humor and informal interaction Encourages challenges to the status quo and fosters creative thinking Incorporates high-quality and diverse information and ensures agreement on its meaning Seeks consensus only after discussion has fully explored the issues and interests and significant effort made to find creative responses to differences Outcome criteria Produces a high-quality agreement Ends stalemate and ensures ongoing input to avoid future conflicts Compares favorably with other planning methods in terms of costsbenefits Produces creative ideas Results in learning and change in and beyond the group Creates social and political capital Produces information that stakeholders understand and accept Sets in motion a cascade of changes in attitudes, behaviors and actions, spin-off partnerships and new practices or institutions Results in institutions and practices that are flexible and networked, allowing the community to respond more creatively to change and conflict

Yes

Yes

Partially No

Partially Partially

Partially

Yes

No Yes

Partially Yes

Partially

Yes

  No No N/A

  Yes No N/A

No Partially No Partially No

Yes Partially Partially Partially No

No

No

2 3 4 5 6 7

  1 2 3 4 5 6 7 8 9

Scale of evaluation ranges from Yes-Partially-No. Source: Own elaboration, using criteria developed by Innes & Booher, organized in Table 1, Vigar, 2006, p. 270

115

Lake Sagaris

discussions of public engagement to date. In some cases, depth is probably not an important consideration for public consultation, for example, when deciding through public consultation where to place a new traffic light or pedestrian crossing. When policies seek to catalyze sustainability transitions, however, to achieve multiple policy goals and the behavioral changes necessary to sustain them, depth is essential. Depth requires continuity beyond the usual life-span of a government. It also requires coherence among the policies, programs, plans and projects developed: it makes no sense for one ministry to strive for better sidewalks and cycle-ways, associated with public transport, while another is obliterating them in favor of highway concessions or other infrastructure. Finally, it requires consistency, with central values that emerge and evolve through the participatory process itself. While power has been central to discussions of the quality of procedural participation since Arnstein’s famous ladder (1969), it has been less explored with regard to outcomes. There is, moreover, a significant interaction between continuity and innovation in power arrangements: oneoff sessions or processes tend to rely on formal procedures favorable to existing power dynamics, hampering the emergence and consolidation of new arrangements. The next section discusses these observations and their implications for participation, public transport and sustainability transitions in more detail, teasing out some key lessons that may be useful to those grappling with similar issues in other contexts.

Discussion: lessons for participation for sustainable, just public transport The motivations and elements summarized in Tables 9.1 and 9.2 reflect an uneven set of values. Some, such as “agency credibility” or “faster implementation”, are extremely expedient, while others, particularly “democratic purposes”, are vital but vague. Classification in terms of Flyvbjerg’s categories of knowledge (Table 9.5) helps to organize and understand them in new and relevant ways (Flyvbjerg, 2001). Most approaches to practicing and evaluating participation focus on procedure, falling mainly within the Techne, or technical, category. This is where participation can turn into a mere checklist of tools and methods, which – even when performed reasonably well – may fail to achieve the multiple outcomes required from transport and engagement in its planning. Much of the content presented falls under Episteme: seeking to prevail, as “universal”, “scientific” and “invariable”. As discussed, however, people do not change based on these two kinds of information alone. In this context, Flyvbjerg’s third category, Phronesis, or situated knowledge, becomes central, with its focus on values, context and action. Emotion, memory, identity could

Table 9.5 Flyvbjerg’s categories of knowledge  

Type

Defined as. . .

1

Techne

2

Episteme

3

Phronesis

Craft/art. Pragmatic, variable, context-dependent. Oriented toward production. Based on practical instrumental rationality governed by a conscious goal. Scientific knowledge. Universal, invariable, context-independent. Based on general analytical rationality. Ethics. Deliberation about values with reference to praxis. Pragmatic, variable, contextdependent. Oriented toward action. Based on practical value-rationality.

Source: Definitions: Flyvbjerg (2001, p. 57)

116

Community and transport decision-making

all be specified within phronesis, as they are necessary to change values, behaviors and policies (Brannigan, 2011). From this perspective, then, it is necessary to go beyond the tools and methods that tend to characterize discussions of participation in public transport planning. Process becomes important, particularly the duration of specific processes but also the creation of ongoing procedures, or institutions, able to maintain participation beyond the initial instance, plan or project. Tools and methods (Table 9.6) must be distinguished from the processes which give them order, meaning and intention: while “tools” are the instruments that may be used, and “methods” tell how they might be used, “process” design (or co-design) organizes them into a meaningful set of encounters and actions. Adequate process design requires sensitivity to diverse kinds of participants, to space, identity and scale and how they shape conflicting and overlapping values, ideas and hopes. Moreover, it requires two additional steps that are often neglected or underrepresented in real-world engagement. The results of participation constitute data that must be processed in a meaningful, succinct and useful way: what these terms mean to those in charge of the process may be quite different from how they are interpreted by diverse participants themselves. Most of this data may be qualitative, consisting of feelings, opinions, sensibilities, values, doubts. Varied interpretations require innovating in language to go beyond Techne and Episteme and embrace Phronesis and beyond, through iterations to validate and re-validate, as the process builds consensuses – and identifies points of conflict or difference that cannot be reconciled. Many participatory procedures increasingly consider how results are “returned” to participants for revalidation. This may take the form of printed information or media coverage (one-way communication), which leaves no room for influential feedback and therefore offers no opportunity to identify and avoid misunderstanding or potential conflict. Sometimes, as occurred with NAP and Temuco-PLC, people are recalled to receive reports of how their input has been processed: even then, procedures may leave little room to adjust, debate or even reject or change some aspects of decisions. Missing from these components of participation is the role of institutions, understood as the rules and instances, the collective Who empowered to guarantee the quality of both procedures and outcomes and their continuity over time. Institutions are central to process quality, to equity, to accountability, but they are also largely invisible, assumed rather than explicit in many engagement processes. NAP, for example, involved a regional government with no formal empowerment to co-ordinate and make decisions on transport policy and plans. It nonetheless created a propitious space for innovation, co-ordinating national and local governments, decision-makers and citizens themselves. This reflected the will of a particular political player, Bachelet’s Nueva Mayoría coalition national government, which also held power in two of the four municipalities through which the project passed. The strength of this structure was that it facilitated innovation; its weakness, however, was that it was unable to consolidate the project enough to guarantee its realization when power shifted and the opposition took over. Similarly, for the regional transport plan in Temuco-Padre Las Casas, even though institutionalizing citizen participation was an explicit objective, recommendations were not followed, most directly due to the change in the national government, which appoints the regional transport secretary. Even if the same coalition had remained in power, however, it is unlikely that participation would have become standard in transport decisions. The 1980 constitution, which was the object of almost daily demonstrations throughout Chile 18 October 2019 to 8 March 2020, is particularly rigid, discouraging innovation and making it impossible for decisive inputs to originate outside the powerful centralized state. Public engagement with climate change and the challenges of achieving just sustainabilities thus emerges as potentially central to institutional reform. Using an action research approach to 117

118 Who

Why

How

What

will guarantee fair, inclusive and respectful: (i) treatment, (ii) rules for engagement, (iii) influence on results and (iv) continuity of participation and results over time.

will we do this: in what order, for what pur pose, with what expectation of results?

will we use the tools?

will we do?

Key Question

Phronesis

Episteme

Techne

Techne

Category

Source: Own elaboration, based on experiences with participation in Chile and field visits to Toronto, London, Seville, Utrecht and elsewhere, using Flyvbjerg’s categor ies of knowledge (Flyvbjerg, 2001). © Laboratorio de Cambio Social, used by permission

Levels of gover nment, by-laws or laws mandating participation, quangos, advisor y commissions, binding plans

INSTITUTIONS

Presentation, expressing preference, participatory mapping

METHODS

Ser ies of field visits, participatory mapping, recording and retur n of results (aspirations, problems, proposals)

Projector, adhesives, colored pens

TOOLS

PROCESS

Example

Components

Table 9.6 Expanding our framework for understanding components of participation

Lake Sagaris

Community and transport decision-making

examine these links, Leith (2011) emphasizes that engagement consists of “building a polity and contributing to the construction of identities, discourses and institutions” (Leith, 2011, p. 116). In light of these reflections, it is possible to enrich Flyvbjerg’s concept of phronesis to explicitly include emotional, identity, memory and other cognitive contents. If this kind of knowledge is seen as an intersection between the life- and thought-worlds, described by Healey, it is possible to move beyond the “reasons” for public engagement, which keep the focus (and the power) within the agency and/or the governmental side of relationships, to consider purposes, more in line with the multiple objectives that transport systems must fulfill in the 21st century. Indeed, a hierarchy of purposes can be considered (Table 9.7) whose aim is to improve public transport’s role as a central part of daily life, people’s individual and social imaginaries, aspirations and pleasures, by meeting the diverse needs of diverse people. From this perspective, several significant shifts occur: • •

•

away from a generic “public” or “consumers” to focus on diverse citizens with rights, responsibilities and obligations; away from individuals toward involvement of (preferably) democratically constituted movements, groups and organizations, able to accumulate knowledge, credibility and capacity to contribute to decision-making and resulting actions, including changes in behavior and habits; away from techniques and procedures (although they remain important), toward a greater focus on mobilizing diverse kinds of knowledge for inclusion in policies, programs and projects.

Table 9.7 Shifting from reasons to purposes for engagement in public transport System-related To create a culture of affection, pride and ownership, fine-tuning stops, stations, hubs, lines and networks to local and regional needs, by Adjusting these closely to the needs, aspirations and preferences of relevant communities, in terms of mobility, access, health and other social goals Responding to specific contextual and territorial needs to access crucial services and concentrate jobs Society-related To contribute to a democratic culture generally by: Sharing and building situated/experiential knowledge of diverse life experiences and local conditions into planning processes and decisions Deliberating in ways that integrate diverse ‘knowledges’, both experiential and technical Contributing to social learning and participation Generating shared ownership of strategies and programs Both To offset inequalities and exclusions due to social differences, with context- and communitysensitive public transit fully integrated fares with support modes (bikeshare, cycletaxis, etc.) To achieve a close match between system, station and other components with the needs of specific user groups and local/regional communities served by specific stops, hubs and lines To make better decisions that reflect both system and community values in a transparent and accountable manner that improves credibility Improve public transport’s integration with other low-cost, clean, healthy modes to facilitate use by people with diverse (dis)abilities Source: Own elaboration starting from “reasons”, Table 36.1, considering phronesis and other aspects of participation and challenges to public transport in the 21st century

119

Lake Sagaris

As Healey puts it, the “knowledge developed in group work is not ‘out there’ waiting to be discovered, but is actively being produced through social interaction and social learning. . . . [People] develop theories about how things are in the course of, and in relation to, action”, with policy-making and the required behavioral shifts becoming “a story about what should be done” (Healey, 1997, pp.  246–259). It is in the writing of this collective story that citizen engagement becomes central and essential to its success.

Final reflections It is interesting to look at the Chilean case when considering the roles and risks inherent in citizen engagement. With some exceptions (Sagaris, 2015), Chilean transport planning has steadfastly avoided genuine public participation in major transport decisions, even during the democratization that followed the military regime (1973–1990).2 The sociopolitical and economic costs have been extraordinarily high. In 2007, despite demands from citizens for participation, implementation of the new public transport system, Transantiago, virtually collapsed the city of Santiago, bringing down a minister and threatening the stability of Bachelet’s first government. Over a decade of improvements have yet to win back public support, with nostalgia for the old yellow buses prominent in recent social protests. Protests were triggered by increases in the Metro (underground rail) fares, when an act of youthful civil disobedience – leaping over turnstiles in a refusal to pay – set off a perfect storm of social movements that go far beyond issues of public transport alone. Many of those in power argued “they hadn’t seen this coming”, while others, Nobel prize winner for economics Joseph Stiglitz among them (Lissardy, 2019), were surprised it had taken so long, given the enormous gap between economic growth and inequality. Others note how the emergence of “critical citizens” and an expanding middle class have stoked the pressures on an institutional framework that consistently favors business and party interests within policy, spending and other decisions (Castiglioni & Rovira Kaltwasser, 2016). For over a decade, new social movements have emerged and sought to influence change, finding virtually no effective channels for citizen participation and institutional innovation. These experiences with citizen participation in transport planning illustrate many of these systemic and institutional factors. Future research on how these issues intertwine and motivate collaboration, conflict or a mixture, catalyzing institutional change or reinforcing barriers and blocks, is necessary to better understand these dynamics in themselves and how they could potentially be mobilized to cross traditional party lines and other divisions and favor greater sustainability and equity in both planning and urban living systems. In the short term, the risks of avoiding significant participation may seem minimal, with benefits distant and diffuse. As with other countries in Latin America, however, Chile’s experience suggests that the debt inherent in democratic deficits inevitably falls due, sometimes in dramatic and unpredictable ways. A recent handbook (Elstub & Escobar, 2019) and initiatives by powerful organizations such as the OECD suggest that citizen engagement, democracy and deliberation are deeply linked and highly relevant to driving systemic interactions in favor of greater sustainability and social justice, and social policies in general. Citizen participation in public transport will not resolve all the outstanding issues, but it can help to build citizen, government and governance skills, the political and social capital necessary to overcome these challenges and obtain the best results possible for the most people. With COVID-19 raising serious questions about overcrowded buses and Metro systems, suitable and deep participation that genuinely addresses users’ concerns and makes them feel 120

Community and transport decision-making

part of planning and operations has become increasingly strategic. As the evidence summarized in this chapter indicates, mere publicity and marketing campaigns will not reach or influence different users groups or draw them into discussions about improvements, which, among other benefits, helps to make them feel co-responsible for the success of the system.

Acknowledgments This research was supported by the BRT+ Centre of Excellence in Bus Rapid Transit, the Centro de Desarrollo Urbano Sustentable (CEDEUS), Pontificia Universidad Católica de Chile, with funding from Conicyt, FONDAP No. 15110020. I  am especially grateful to local and regional partners in Santiago, Ximena Abogabir, Casa de la Paz; Temuco-Padre Las Casas, Rodrigo Vallette and the action research team involved: Ximena Vásquez, Mauricio Saavedra, Marisol Recondo, Daniel Lanfranco, Maya Flores, Gonzalo Cancino, Juan Carlos Muñoz (Pontificia Universidad Católica de Chile); Nicolás Aguilar Farías, Damian Chandía, Universidad de la Frontera.

Notes 1 Citizen organizations may take the form of community-based organizations (CBOs), such as neighborhood associations, which typically involve people with very diverse political and other views; nongovernmental organizations (NGOs), which may be small to large groups of likeminded individuals or advocacy groups, such as pro-cycling or walking organizations. In this chapter, civil society organizations is used as the generic term for these diverse approaches, per global practice. 2 One attempt to overcome this occurred during major regional and nation hearings to propose a new participatory structure for citizen engagement and decision-making in public transport decisions. Again, recommendations were shelved due to institutional and other limitations and the lack of a strong political will to reform and overcome them.

References Aldred, R. (2013). Who are Londoners on bikes and what do they want? Negotiating identity and issue definition in a “pop-up” cycle campaign. Journal of Transport Geography, 30, 194–201. Arnstein, S. R. (1969). A ladder of citizen participation. Journal of the American Institute of Planners, 35(4), 216–224. doi:10.1080/01944366908977225 Barber, B. R. (1998). A place for us: How to make society civil and democracy strong (1st ed.). Hill and Wang. Berry, J. M., Portney, K. E., & Thomson, K. (1993). The rebirth of urban democracy. Brookings Institution. Bickerstaff, K., Tolley, R., & Walker, G. (2002). Transport planning and participation: The rhetoric and realities of public involvement. Journal of Transport Geography, 10, 61–73. Bickerstaff, K., & Walker, G. (2005). Shared visions, unholy alliances: Power, governance and deliberative processes in local transport planning. Urban Studies, 42(12), 2123–2144. BicycleInnovationLab. (2011). The reverse traffic pyramid. https://www.researchgate.net/figure/ The-reverse-traffic-pyramid-Source-Bicycle-Innovation-Lab_fig1_328748845 Booher, D. E., & Innes, J. E. (2002). Network power in collaborative planning. Journal of Planning Education and Research, 21(3), 221–236. doi:10.1177/0739456x0202100301 Brannigan, F. (2011). Dismantling the consumption-happiness myth: A  neuropsychological perspective on the mechanisms that lock us into unsustainable consumption. In L. Whitmarsh, S. J. O’Neill, & I. Lorenzoni (Eds.), Engaging the public with climate change, behaviour change and communication (pp. 84–99). Earthscan, Taylor & Francis. Castiglioni, R., & Rovira Kaltwasser, C. (2016). Introduction: Challenges to political representation in contemporary Chile. Journal of Politics in Latin America, 8(3), 3–24. Cooke, B., & Kothari, U. (2001). Participation: The new tyranny? Zed Books. Elstub, S., & Escobar, O. (2019). Handbook of democratic innovation and governance. Edward Elgar Publishing.

121

Lake Sagaris Flyvbjerg, B. (2001). Making social science matter why social inquiry fails and how it can succeed again. Cambridge University Press. Giering, S. (2011). Public participation strategies for transit. http://onlinepubs.trb.org/onlinepubs/tcrp/tcrp_ syn_89.pdf Healey, P. (1997). Collaborative planning: Shaping places in fragmented societies (2nd ed.). Houndmills, Palgrave Macmillan. Höppner, C., & Whitmarsh, L. (2011). Public engagement in climate action: Policy and public expectations. In L. Whitmarsh, S. J. O’Neill,  & I. Lorenzoni (Eds.), Engaging the public with climate change, behaviour change and communication (pp. 47–65). Earthscan, Taylor & Francis. Hovey, B. (1998). Building the city, structuring change: Portland’s implicit Utopian project. Utopian Studies, 9, 68–79. Innes, J., & Booher, D. (1999). Consensus building and complex adaptive systems: A framework for evaluating collaborative planning. Journal of the American Planning Association, 65(4), 412–423. Ladd, B. (2008). Autophobia: Love and hate in the automotive age. University of Chicago Press. Leith, P. (2011). Public engagement with climate adaptation: An imperative for (and driver of) institutional reform? In L. Whitmarsh, S. J. O’Neill, & I. Lorenzoni (Eds.), Engaging the public with climate change, behaviour change and communication (pp. 100–119). Earthscan, Taylor & Francis. Lissardy, G. (2019). Entrevista con Joseph Stiglitz, nobel de Economía: “La sorpresa fue que el malestar en América Latina tardara tanto en manifestarse”. BBC. Mohl, R. A. (2012). The expressway teardown movement in American cities. Journal of Planning History, 11(1), 89–103. doi:10.1177/1538513211426028 Moulaert, F., Martinelli, F., Swyngedouw, E., & González, S. (2010). Can neighbourhoods save the city? Community development and social innovation. Routledge. Beuhler, R. & Pucher, J. (2021). The future of city cycling (revised ed.). The MIT Press. Sagaris, L. (2014). Citizen participation for sustainable transport: The case of “living city” in Santiago, Chile (1997–2012). Journal of Transport Geography, 41, 74–83. Sagaris, L. (2015). Lessons from 40 years of planning for cycle-inclusion: Reflections from Santiago, Chile. Natural Resources Forum, 39(1), 64–81. doi:10.1111/1477-8947.12062 Sagaris, L. (2018). Citizen participation for sustainable transport: Lessons for change from Santiago and Temuco, Chile. Research in Transportation Economics, 1–8. Sirianni, C. (2009). Investing in democracy: Engaging citizens in collaborative governance. Brookings Institution Press. Thomson, K. (2001). From neighborhood to nation: The democratic foundations of civil society. University Press of New England. Verplanken, B. (2011). Old habits and new routes to sustainable behaviour. In L. Whitmarsh, S. J. O’Neill, & I. Lorenzoni (Eds.), Engaging the public with climate change, behaviour change and communication (pp. 17–30). Earthscan, Taylor & Francis. Vigar, G. (2006). Deliberation, participation and learning in the development of regional strategies: Transport policy making in North East England. Planning Theory & Practice, 7(3), 267–287. www.informa world.com/10.1080/14649350600841446 Whitmarsh, L., O’Neill, S. J.,  & Lorenzoni, I. (2011a). Engaging the public with climate change, behaviour change and communication. Earthscan, Taylor & Francis. Whitmarsh, L., O’Neill, S. J.,  & Lorenzoni, I. (2011b). Introduction: Opportunities for and barriers to engaging individuals with climate change. In L. Whitmarsh, S. J. O’Neill, & I. Lorenzoni (Eds.), Engaging the public with climate change, behaviour change and communication (pp. 1–14). Earthscan, Taylor & Francis.

122

10 LAND VALUE GAINS AND VALUE CAPTURE The potential for financing public transport infrastructure Corinne Mulley, Barbara T.H. Yen and Min Zhang Introduction Cities are the economic powerhouses of today, where an efficient cost-effective public transport system plays an essential role. It facilitates the economically productive agglomerations of firms and improves accessibility to goods, services and activities, which are central to transport activity. Better public transport and transport infrastructure generate benefits for users and help manage urban congestion and climate change, contributing to the goal of greater urban sustainability. Notwithstanding, modern transport system infrastructures are expensive and challenging for governments to finance, competing as they do with the other demands on government. As a result, one of the most challenging problems for transport planning is how to finance public transport investment from land value gains (known as value capture). Value capture is where the benefits from land value gains are monetised to pay for transport infrastructure developments. The implementation of value capture provides the opportunity to open up new and equitable financing because the burden is spread between beneficiaries. The following section provides a brief overview of the theory as to why developments in transport infrastructure can lead to land value gains. The chapter then turns to the size of those gains, because these underpin the potential for success for value capture mechanisms; this is followed by a discussion of value capture strategies and an evaluation of them against a number of normative criteria. As value capture is predicated on capturing the uplift following the implementation of new infrastructure, it is imperative that an ex-ante estimate of such uplift be available for policy-makers. The penultimate section presents new empirical work to predict value uplift showing how value capture policies can be evidenced based on the likely value uplift using a case study from Brisbane, Australia. A final section concludes the chapter and identifies areas for further research.

Land value uplift Land rent theory, developed by Alonso (1964) and Muth (1969), is the theoretical framework linking accessibility to goods and services and land values. These theories hold that land rent (and therefore the underlying land value) reflect accessibility gradients, with higher values 123

Mulley, Yen and Zhang

of rent reflecting higher accessibility. There is a well-established literature demonstrating that new or extended transport infrastructure provides improvements in accessibility and therefore increases in land value – termed land value uplift – with uplift benefits being distributed in relation to the proximity of the location to the infrastructure and both residential and commercial properties. Determining the uplift turns out to be more difficult than the relatively simple theory suggests. The theory identifies land value uplift as coming from changes in accessibility, which in turn changes the underlying land value (or what is called the unimproved land value). However, the underlying land value for residential properties is what is termed the unimproved land value, whereas the residential land values observed by the market are the unimproved land value which has been ‘improved’ by the building of property. This means there are two ways of measuring value uplift: using the market price of residential land (but then the method must control for the way in which the property increases the land value) or using unimproved land values, which are often established for taxation purposes. In some countries, market price is replaced by advertised (or asking) price when market prices are not available, but in practice, market and advertised prices are highly correlated. There are difficulties with either approach since, although market valuations need to be adjusted, the use of unimproved land values assumes that these valuations are correct. Once data issues are resolved, there are a number of possible methods to estimate the extent of uplift. The earliest studies replicated experimental methods and used comparative methods to look at house values before and after the infrastructure was put in place. The method was simple and required average price changes of smallish areas to be compared: these did not take into account the complexities that drive house prices such as the sociodemographic nature of the area, the local amenities or indeed that bigger properties usually achieve higher prices. More sophisticated attempts at measuring uplift then began to use hedonic modelling, which controls for the complexity of house price determination by valuing the different components which make up a house price – the number of bedrooms, bathrooms, and parking; neighbourhood effects and the accessibility of the property to different destinations. Hedonic modelling, in its most basic form, uses multiple regression methodology, which requires strict assumptions which are often broken when modelling house prices, because, for example, there are spatial connections between how desirable a house is and where it is located. With advances in computing power, some of these spatial problems with hedonic models have been overcome by using spatial modelling which controls for the spatial autocorrelation that gives rise to problems. These models give better estimates of the value uplift, but the boundaries of the areas of interest have to be chosen by the modeller. In practice, these tend to be politically driven, for example, local authority boundaries or census areas which have no real relationship to the underlying housing market. A technique called geographical weighted regression (GWR) avoids this by separately considering each data point, but this method is less good at resolving other spatial problems and is data hungry. Finally, difference-in-difference (DID) methods have had an upsurge of interest recently and employ a quasi-experimental design using time series data analysing control and treatment groups to measure an appropriate counterfactual – in this case new public transport infrastructure – to estimate a causal effect. DID can be thought of as a superior form of the earlier comparison methods, usually using a regression method where the measurement of effect is by an interaction term between time and the treatment group dummy variable. The published evidence on uplift appears to be context dependent. The values obtained for uplift are mixed, and some of this difference must come from the difference in method used. RICS (2002), Smith and Gihring (2006, 2009), and Medda (2012) have reviewed over 124

Land value gains and value capture

100 international studies on the impact of public transport on property values, focusing mainly on the impact of heavy rail, metro and light rail. However, a short list of influencing factors includes whether the property is commercial or residential, the mode under consideration, the nature of the neighbourhood and the quality of the infrastructure and its service. Also important is the timing of the uplift and whether generating a network effect, whereby improvements to the network provide access to more destinations, gives value uplift benefits. These are considered in turn in the following sections.

Residential versus commercial In principle, land value will increase with increased accessibility whether it is residential or commercial land. Commercial uplift is in some ways easier to capture, as it might be expected that the commercial entrepreneur would anticipate the uplift and be able to internalise it in their plans: in these cases, the planning approval processes tend to extract the uplift through planning gain, whereby the developer includes elements of a project beneficial to the community in exchange for planning permission. Higgins and Kanaroglou (2016) identify that just over 20% of all land value uplift studies are on commercial property values, including Ko and Cao (2013), Deng and Nelson (2010) and a meta-analysis by Debrezion et al. (2007). As the majority of value uplift studies are concerned with residential properties, the evidence presented in the remainder of this chapter relates primarily to residential property value uplift.

Mode The reviews by Smith and Gihring (2006, 2009) identify the value uplift literature for public transport rail-based projects, and the evidence is strengthened by a more recent review paper that considers only studies from the United States and effects on single-family houses (Higgins & Kanaroglou, 2016). Summarising the results suggests that for heavy rail transit (HRT), uplift appears to fall from the station to 800 m out, suggesting the value uplift is falling over the typical walking distance. Light rapid transit (LRT) has lower overall value uplift and appears more variable, although this is partly because the analysis includes LRT systems on dedicated rights of way and LRT systems in mixed traffic. The contribution of new bus-based infrastructure to land value uplift is an area that is less apparent in the literature, despite the progressive implementation of bus rapid transit (BRT) (see also Chapter  13). A  meta-analysis of BRT found BRT has mixed outcomes in terms of land value uplift and typically lower value uplift than rail-based infrastructure (Zhang & Yen, 2020). Rodriguez and Targa (2004) and Munoz-Raskin (2009) both studied the impact of BRT in a developing country context, focusing on the Transmilenio in Bogota, Colombia, and found that, as for rail, the housing market placed value premiums between 6.8% and 9% on residential properties in the immediate walking proximity of feeder lines to the BRT service. As BRT penetration increases, more studies have become available (Deng & Nelson, 2010, for Beijing, China, with 2.3% annual growth, and Cervero and Kang, 2011, for Seoul, Korea, with between 5% and 10%). Also, whilst LRT and HRT studies have typically been in developed countries, BRT or bus-based infrastructure evidence has typically come from the less-developed countries – particularly South America. This makes comparison less easy. But this highlights one of the important underpinnings of value uplift – that of capitalisation of accessibility. It appears that once built, rail-based infrastructure is regarded as fixed and the improvements in accessibility regarded as permanent. In contrast, despite gaining in popularity due to having greater cost effectiveness 125

Mulley, Yen and Zhang

(vis-à-vis LRT), BRT, as a high-capacity urban public transport system, typically with its own right of way (as for rail-based modes), is not seen as permanent as rail-based systems and so does not realise as great a value uplift. In the developed world, BRT has made a significant contribution to public transport mode share in Brisbane, the only Australian capital city where BRT features prominently. There have been a number of studies in Australia; for example, for the Liverpool Paramatta Transitway (LPT) in North-West Sydney, a repeat sales method identified that houses between 100 and 400 m and units between 800 and 1200 m of the LPT benefit but with negative impact for properties within 100 m of BRT, with the uplift occurring after opening (Mulley & Tsai, 2016). A negative impact was also noted in Mulley (2014), with property prices also declining on average by 9.2% when located within 100 m of the LPT. In Brisbane, a GWR approach showed an average uplift of 0.13% for each 100 m closer to the station, provided residential properties are not too close. Access to train stations was on average negative (Mulley et al., 2016). This negative effect is not uncommon in Brisbane and may be due to poor levels of service for trains, the effect of crime and/or noise from freight lines and the effects of flooding. However, as GWR offers the opportunity to investigate the spatial distribution of value uplift, Mulley et al. (2016) found that the value uplift for distance to BRT stations is relatively stronger at stations further away from the central business district (CBD), with non-significant effects for some BRT stations closer to the CBD. These results align well with other findings and could be expected because more travel time can be saved through use of the BRT by residents living further away from the CBD. Other developed countries’ experience for BRT is more mixed. In Quebec, Canada: only properties located far enough away to avoid noise but close enough to use the BRT experienced value uplift of between 3% and 7% (Dubé et al., 2018). In Pittsburgh, US: uplift of around 16% was identified, but other positive factors were also present (Perk & Catala, 2009), but in Los Angeles, US, no evidence of uplift was found (Cervero & Duncan, 2002). Uplift for the ferry mode is virtually absent in the literature. A study in Brisbane, another GWR study, showed an average of 4% uplift for properties closer to the ferry and 2% for properties closer to the CBD (a 1-km decrease in distance from ferry terminal/CBD leads to a 4%/2% increase in price) (Tsai et al., 2017). For spatial distribution, the uplift was restricted to mature ferry terminals with immature wharfs, where building being new or currently underway not having positive uplift, although this might come as the area matures.

The nature of the neighbourhood and the role of quality in creating value uplift ‘Location, location, location’ is a common saying in relation to the value put on property. This flows through to identifying uplift as different neighbourhood areas and different quality systems appear to attract different levels of uplift. Areas in decline do not achieve the highest value uplift, with, for example, a meta-study of rail finding up to a 32% premium but more modest uplift in the poorer area of Buffalo, New York (Hess & Almeida, 2007). Value uplift only seems to occur when the improvement has a significant impact on travel times (Ryan, 1999). Higher-quality services in general lead to more uplift. Meta-analysis by Debrezion et al. (2007) and Zhang and Yen (2020) shows a positive correlation between value uplift and service quality indices such as frequency, connectivity and travel times. In terms of spatial distribution, value uplift occurs around stations where access was by walk or public transport, whereas park-and-ride stations were associated with disamenity, and walkable areas in a transit-oriented development (TOD) study found value uplift only when TOD replaced 126

Land value gains and value capture

Park and Ride (Kahn, 2007; Mathur & Ferrell, 2013). Finally, higher uplift is associated with higher densities and greater amenities (Atkinson-Palombo, 2010).

Timing of value uplift Cross-sectional studies cannot address when uplift occurs – whether this is upon announcement of the project, when the project build starts or when service commences. The pool of studies looking at this issue of timing is relatively small. However, if uplift is to be captured, it is important to know when the value uplift happens. Ignoring timing issues might well bias the estimation of value uplift, as uplift is not guaranteed to rise linearly. This means that generally, longitudinal studies should be preferred. The evidence that exists is mixed, with little consistency. For Chicago’s rapid transit, the market appeared to anticipate uplift, but uplift was stronger after the line opened (McMillen & McDonald, 2004). With heavy rail in Puerto Rico, there was, in contrast, no impact on land values until operational (Loomis et al., 2012); this was also true in Sydney, Australia, where the Liverpool Paramatta Transitway uplift of 11% occurred after opening (Mulley & Tsai, 2016). In Phoenix, US, the LRT achieved value uplift at each stage of development, more than nine years before opening (Golub et al., 2012), whereas the LR in Portland, US, had uplift for vacant plots after announcement (an increase or up to 70%), but the uplift fell back to 20% after year 2 (Dueker & Bianco, 1999). In the Gold Coast, Queensland, Australia, a longitudinal analysis shows that property prices around the new LRT started to increase right after announcement, with the highest increment to value uplift occurring after solid financial commitments had been made by government (Yen et al., 2018), and then property prices slowed during construction and operation periods. The model also shows uplift effect differences for different sections of the catchment, with the highest increment being for properties with greatest accessibility to LRT stations. Interestingly, this analysis shows that properties sold in 2016 would have a higher price if located closer to the HRT station which was planned for an interchange in the second stage of the Gold Coast LRT, thus suggesting that this stage two might experience value uplift impacts on property prices immediately construction began.

The network effect In practice, much new infrastructure is an addition to a network rather than the creation of a totally new network. The additional value to properties from network extensions is rarely studied and quantified. A recent study in Australia looks at this issue in the context of the BRT Busways in Brisbane, Queensland. The BRT Busways in Brisbane are the result of various network extensions with the first part of the network being the Busway to the southeast (SE) of the city. Mulley et al. (2017) look at the value of the network effect to the residents living close to this SE Busway as a result of the northern and eastern Busway extensions being made. The study suggests the value of the uplift is approximately 3% to the properties close to the SE Busway although the actual timing of this uplift is less certain.

Implications of value uplift for value capture The empirical analysis shows clear spatial variation, suggesting a uniform value capture tax is unlikely to achieve horizontal equity. Whether vertical equity is achieved is uncertain, although as income and property values are positively correlated, there will be confounding effects that may prejudice vertical equity. 127

Mulley, Yen and Zhang

The empirical studies demonstrate that method and context influence the estimated value uplift. The studies also show that new infrastructure forming a backbone to the network, such as the Busways in Brisbane, Queensland, Australia, do better than when the same mode is more tangential to the network, such as the Liverpool Paramatta Transitway in Sydney, NSW, Australia. Importantly, and underresearched, are the value uplift from the network effects from public transport system extensions: this suggests that value capture policies should take a wider rather than narrow spatial viewpoint. However, empirical studies are valuing uplift ex post: for value capture policies to be effective, some estimate of value uplift needs to be made ex ante. The penultimate part of this chapter proposes such an estimation process.

Value capture Governments’ financing systems to date emphasise user fees, which are primarily fares in the case of public transport infrastructure. However, the need for greater investment has resulted in searches for new methods to aid or replace user fees. Value capture relies on recovering the uplift received by property owners (residential and commercial) from enhanced accessibility brought about by new infrastructure. Knowing the uplift is clearly a prerequisite to its capture. Interest in value capture is increasing, see, for example, Aveline-Dubach & Blandeau, 2019; Mathur, 2019; Salon et al., 2019. Value capture policies, as with all taxes, have different characteristics. In this section, six value capture strategies are briefly evaluated in terms of their economic efficiency (is the cost to taxpayers related to the benefits they receive?), equity (how does it affect different income groups and different geographies) and ability to pay a sustainable and predictable revenue and feasibility (political and administrative transparency, perceptions of taxpayers). In terms of applications in a public transport context, value capture to date has taken a number of forms, of which the following are the more frequently used: 1

2

3

4

Land value tax has been used in many jurisdictions to raise finance. This could be enhanced for value capture purposes and would be best based on unimproved land values, assuming that unimproved values are identified as broadly related to underlying market values. Land taxes are relatively economically efficient and are sustainable, although they are likely to be regressive and politically unpopular. Tax increment financing (TIF) levies a tax on the future increment in property value to finance development-related costs. It has rarely been used for transport projects, and there is not much information about its efficiency. It has good feasibility and is regarded as equitable, although there may be some spatial equity issues if overlapping government areas do not benefit from a single TIF area. Special assessments impose charges on property owners based on their geographic proximity to the new infrastructure. It is an economically efficient measure which is generally good on equity (unless the charge is linked to property value, when it tends to be regressive). It is low on sustainability since it has a narrow tax base but often politically acceptable to businesses, although homeowners find it less palatable. Transit-focused development fees are one-time charges on new developments to contribute to costs of growth-related public services such as a public transport system. Other forms of development fees have been common in planning, for example, negotiated payments or planning gain, which are an in-kind contribution to local roads, new parks or other public goods as a condition of planning approval but do not specifically address financing of new

128

Land value gains and value capture

5

6

public transport. It is an efficient way of raising money, as the highest charge is to those most likely to benefit (if charged to developers, it is in effect passed on to the new owners), and equitable, as payments are roughly proportional to benefits. It is sustainable for single development and, as it is low visibility when implemented, politically feasible and regarded as ‘paying their way’. Joint development occurs when a transport facility and adjacent private real estate development are developed together, with the private sector partner making a financial contribution to offset the costs of the transport facility. It is usually good on efficiency grounds, as contributions align with expected benefits and the voluntary nature of payment encourages equity. Revenues are not necessarily sustainable as related to the state of the real estate market but politically palatable to implement. Transit-focused property transaction fees are collected when properties are sold. The collection rate fluctuates with the real estate market. This can be efficient but could be regressive. However, it is simple to administer and does well in terms of keeping pace with inflation.

Other value capture strategies have been used in transport but not specifically public transport financing. These include transportation utility fees, where transport networks, like other utilities, are primarily financed from user fees and air rights agreements and establish development rights above or below a transport facility in exchange for a financial contribution and/or future stream of revenue from land/property or income taxes. No obvious public transport investment has been financed from air rights so far, although the complicated arrangement in Hong Kong between the public transport operator and the government come close to this. The evaluation of these strategies are summarised in Table 10.1. This shows that there is not an overall winner in terms of a strategy that scores well on all of the characteristics. However, as Table  10.1 illustrates, these strategies have been applied, and, as with many policy deliberations, there is no ideal option, and it has become a question of choosing the most suitable when making the choice for capturing the value uplift consequential on new public transport infrastructure.

Predicting value uplift for value capture The review of the value uplift literature demonstrates how it has varied considerably depending on mode, neighbourhood features and other factors that constitute the context of the public transport investment being considered. A sensible value capture policy must be based on likely value uplift since the policy of capture is in place before the public transport investment is made. This means that an ex-ante forecast of likely uplift is necessary, taking into account as many of the contextual factors as possible. This section proposes a three-stage modelling procedure to predict land value uplift effects in a case study, as shown in Figure 10.1. Underpinning this approach is that if the catchment area of the new transport infrastructure matches the characteristics of the existing infrastructure in terms of mode share and neighbourhood characteristics, the new infrastructure development will bring the same house price impact to its catchment area. The case study location is South East Brisbane (SEB), Brisbane, Australia.

Step 1: Transport system development match at system level Since the new infrastructure does not exist, a transport planning model evaluates the future transport impact in terms of mode share, which is the yardstick used to evaluate the development of the

129

130

Good potential Good

Good

Good

Transit-focused real estate fee

Good

Good potential

Transit-focused development fees Joint development

Possibly regressive

Neutral or slightly progressive

Probably neutral

Slightly Depend on structure and regressive exemption

Possibly regressive

Slightly regressive

Ability to pay

Special assessments Good

Little evidence for Good transportation use

Tax increment financing

Good

Cost/benefit to payers

Price signals, economic growth

Good

Equity

Efficiency

Land value tax

Schemes

Table 10.1 Summary of value capture mechanisms

Nar row base, adjustable for growth, one-off opportunity

High

Nar row; High adjustable for growth; cyclical Nar row base, High limited revenue; cyclical

Low costs

Complex

Low costs

Difficult to establish

Complex, costly

Fairly simple

Political Administrative

Feasibility

Low Broad base, modest g rowth, fairly stable High Nar row, for limited projects; keeps pace with inflation, incomes Nar row base, Low limited revenue, one-off opportunity

Adequacy, growth potential, stability

Sustainability

Hong Kong Mass Transit Railway (Verougstraete & Zeng, 2014); Crossrail, London, UK; Grand Par is Express, Par is, France; New York Avenue Metro Station (2001), US; and Subway 7 Line Extension, New York, US Subway 7 Line Extension, New York, US

Crossrail project, Elizabeth line east-east railway, London, UK; New York Avenue Metro Station (2001), Washington, US; and Dulles Metrorail Silver Line Expansion, Washington, US Crossrail, London, UK, and Grand Par is Express, Par is, France

MetroRail Red Line, Texas, US; Crossrail, London, UK; and Subway 7 Line Extension, New York, US

Urban infrastructure in São Paulo, Brazil, and Crossrail, London, UK

Case studies

Mulley, Yen and Zhang

Land value gains and value capture

Figure 10.1 Model procedures for future infrastructure land value uplift

target transport system and identify when the target transport system (SEB extension in this case) has the same anticipated mode share as the existing transport system (current SEB in this case). The standard four-stage model follows the sequence of trip generation, trip distribution, mode choice and traffic assignment to predict mode share (Rosenbaum & Koenig, 1997). Trip generation modelling determines the number of trips entering and leaving each zone, known as attraction and production, respectively. Trip distribution then predicts the movement of passenger trips between each origin and destination zones pair (Solecka & Żak, 2014). Next, mode choice allocates trips to available transport modes (PTV, 2016). Finally, traffic assignment is determined by trip assignment modelling, which assigns origin and destination (OD) pairs to available routes. These mode OD matrices are loaded onto the transport network, giving mode share as the output.

Step 2: Transport system development match in neighbourhood level Step 2 has two parts; the first part estimates neighbourhood characteristics for future transport investment. For example, if the four-stage model reports the SEB extension needs 10 years to reach current SEB levels, the neighbourhood characteristics would be predicted for 10 years later for the catchment areas of SEB extension. The second part is neighbourhood matching after estimating the neighbourhood characteristics. Both parts are described in more detail subsequently. 131

Mulley, Yen and Zhang

Census data is used to estimate neighbourhood characteristics in the first part of this step. In this example, the 2011 census data is used to estimate the neighbourhood characteristics of each statistical area Level 1 (SA1) by the date (denoted by 20XX) when the new transport infrastructure will be operating at the same level as existing infrastructure. The SA1 is the new base unit of output geography for the 2011 Census in the Australian Bureau of Statistics (ABS). It is the smallest level at which census data is reported, with an average population of 400 persons. Data for projected population at that date comes from Queensland government statistics (Queensland Government Statistician’s Office, n.d.) for this case study, whilst other neighbourhood characteristics are projections based on linear regression modelling using Australian Bureau of Statistics census data of 1996, 2001, 2006, 2011 and 2016 (Australian Bureau of Statistics, n.d.). This finally gives a predicted set of neighbourhood characteristics for 20XX so that neighbourhood matching can be undertaken (i.e., percentage of older people [>65 years old], young people [1,012,000 passenger trips (from October 2017) with an average of >37,000 passenger trips/month. The corresponding figures for the Rural and Regional services are >148,000 passenger trips delivered (from November 2018) with an average of >7,000 passenger trips/month. It is acknowledged, though, that such services are expensive to provide, with an average reported cost per trip of $180 (at June  2018) and farebox recovery of 3% compared to 24% for fixed route services.2

Approaches to the design and evaluation of flexible transport services As already noted, FTS can be found in a variety of operating contexts, and there is a substantial body of literature on the evaluation of such services (see, for example, Ambrosino et al., 2004; Brake et  al., 2007; Davison et  al., 2014; Mulley  & Nelson, 2016; Nelson  & Wright, 2012; TCRP, 2008). A  recent international evaluation of FTS schemes is provided by Pettersson (2019). Brake et al. (2006), in their DRT Good Practice Guide designed to assist in the provision of the (then) newly emerging telematics-based FTS, identify three key factors to be aware of. The first requirement is to develop an economic framework. This is an aspect of planning for FTS that has been frequently overlooked (see Mulley and Clifton (2016) for further discussion). A robust 228

Flexible transport services

economic framework is useful, as it provides a clear method for thinking about the nature of costs and a rationale for linking revenues to costs. FTS are frequently described as being not ‘viable’, which is usually taken to mean that they cost too much (see the example of Kutsuplus in Helsinki below). It is important, however, that the service provider have a clear picture of the costs and revenues associated with a particular operation as well as an appreciation of avoidable costs associated with different time horizons. Appropriate service design is a critical stage ideally developed with the key stakeholders involved in a FTS scheme. These usually comprise the travelling public, the transport service operator, the transport authority and the funder. Experience shows that there needs to be a willingness for key actors to be flexible; the completion of comprehensive user requirements (Finn et al., 2004), which is a fundamental yet often overlooked stage, and an awareness of the constraints involved in a FTS (space, time, type of vehicle, payment method). Clear guidance on a range of possible service typologies is given in Engels and Ambrosino (2004) and discussed further in Currie and Fournier (2019). The third key factor for successful FTS provision is awareness raising. The fact that the more flexible a service becomes, the less visible it is to the end user poses an additional challenge from the perspective of service operation. Most customers like the familiarity of bus type and route branding and the way that a service is available (even if they do not use it regularly). It is important that conventional marketing techniques be supplemented by aggressive marketing of the FTS product, for example, through community awareness-raising activities. Strong branding and a willingness to spend money on marketing have been shown to be important contributors to success.

Evaluation methods There have been numerous FTS evaluation exercises. The EC-funded SAMPO (1996–97) and SAMPLUS (1998–2000) projects developed an evaluation framework composed of three main components (Mageean & Nelson, 2003): • •

Assessment objectives and priorities should be identified for the user groups at each site/ location; An estimate of the impact and effectiveness of the service on the user groups should be completed at each site. To achieve this evaluation, indicators for three assessment categories were defined: • • •

•

economic viability (considering operational efficiency and financial performance); service provision (behavioural evaluation and distributional costs and benefits), and technical performance.

Assessment of future markets by analysing the benefits to service providers and operators following the implementation of FTS, for example, expected increase in patronage and cost effectiveness of operations.

The importance of good service design and robust evaluation is briefly illustrated in the following through consideration of three FTS scheme cases. CallConnect is an example of a well-designed successful DRT service which was established in 2001 and has operated on a continuing basis since then. Interconnect (of which CallConnect is an integral component) is a quality network of connecting local bus services designed to improve transport links to destinations throughout rural Lincolnshire in 229

John D. Nelson and Steve Wright

England. It comprises 10 main fixed-route services and 6 DRT feeder services, which are branded as CallConnect. The feeder services help secure the viability of the fixed-route services. Connections are guaranteed to give passengers confidence, and a taxi is offered as a last resort. A second component of the flexible service is 14 area-based DRT services. Since inception, the service has resulted in the level of unmet need in the communities served being reduced by 90%. Today, the subsidy required for DRT is approximately the same as that required for the former much less effective fixed-route services. This has been achieved by the lower costs of operating eight-seater buses, making this type of operation much more affordable for the level of service provided than conventional services, though the low volume of passengers generated from a deeply rural area means that the subsidy per passenger is always going to be high. The Kutsuplus DRT service, Helsinki, Finland, is an example of a high-profile failure. The service started in October 2012 with ten microbuses, with an additional five vehicles added in November 2013. It was discontinued at the end of 2015. It used a combination of technologies (automated vehicle location, trip combination optimisation, vehicle routing, and travel time estimates). Users received trip offer(s) by requesting a trip via browser-based interface or via SMS. The service was stop-to-stop, with about 1,000 bus stops in the area and an additional 50 virtual stops. The operating time was 06:00–24:00. The average fare was around €5.5/trip (compared to an average public transport trip of €3). The service was subsidised by the member municipalities of the Helsinki Regional Transport Authority (HSL) region (€16/trip). There were several “service classes” across the pilot period. Trip pricing was based on the fixed starting fee and the km price calculated as the direct distance between origin and destination. The service classes differed in flexibility (economy, fast, “happy hour”), and group discounts were offered. The ArrivaClick Microtransit service has operated in South Liverpool, UK, since August 2018. There are no fixed routes, with journeys determined by where passengers want to go within an area running from Liverpool city centre to John Lennon Airport in the south of the city. The bus company Arriva worked with the regional transport authority, Merseytravel, to roll out the app-based on-demand public transport service, initially with six 15-seat buses, increasing to 25 vehicles by summer 2019. Passengers can ‘order’ and track a vehicle from the app, which provides them with a guaranteed fare (£2 or less) and allows them to choose their pick-up point and reserve a seat. Computer algorithms match passengers travelling in the same direction, dynamically routing vehicles in real-time to find the optimal route for their trip. The response time is between 5 and 15 minutes. Early indications show that up to 40% of passengers are switching from cars to use the service, and younger passengers are more likely to adopt this form of public transport. Weckström et al. (2018) identify a number of lessons that can be learned from the Kutsuplus experience in terms of things that could have been done better, and these are strikingly consistent with the factors identified by Brake et al. (2006). End-user analysis (i.e. understanding the requirements of the target market) is essential, and significant effort is required if it is to be done well. Also, the lack of targeted marketing was identified as one of the main failures. Another shortcoming was in terms of service usability and integration with a key recommendation that DRT should use both web, SMS and app-based booking systems, as well as enabling pre-ordering more than 45 minutes in advance. From a user perspective, Kutsuplus contained a range of positive features, and these include a service that was complementary to public transport in areas of low public transport accessibility, low cost for the end user, faster travel compared to public transport, comfortable vehicles and an overall innovative feel to the service. 230

Flexible transport services

The case of rural flexible transport services This section considers the particular case of FTS provision in rural areas. As reflected by previous studies (e.g. Ambrosino et al., 2004), FTS has long been considered a key part of the solution for rural mobility; the door-to-door nature of many services can be critical for certain users such as elderly and disabled people. The demand-responsive element means that it can be more economically viable to operate in areas where conventional transport services are unsustainable and, as noted earlier, FTS can also successfully act as a feeder to other services. The EU SMARTA project (SMARTA 2020) has explored the role of frameworks in the provision and regulation of rural mobility, including how the various frameworks contribute to rural mobility provision (including DRT) and how frameworks can also create barriers to rural mobility. The SMARTA project identified and analysed a number of demand responsive transport ‘good practice’ cases (Table 17.1) with a focus on identifying factors contributing to their success. A common success factor is collaborative working (e.g. Flexitec, Belgium; DRT Tejo, Portugal; Regiotaxi, The Netherlands; flexible mobility services, Bulgaria and DRT Castilla y Leon, Spain). Successful operations in these cases were commonly dependent upon close working between municipalities, service providers and public transport operators. In the case of DRT Castilla y Leon, Spain, conventional public transport services were redesigned to optimise connections with DRT services. Flexible mobility services, Bulgaria, demonstrated strong public/ private co-operation between the local municipality and SMEs operating services. For DRT Tejo, Portugal, co-operation was exemplified by a common administration and booking centre for a range of different mobility services to optimise operational costs. Flexitec, Belgium, similarly, pointed to the benefits of combining resources to deliver services in a rational and creative way, in this instance through partnership working between social services, municipalities, public transport operators and DRT service providers. For three cases, technology was identified as a key success factor (Village Bus, Sweden; Prontobus, Italy and on-demand pooling services, Spain). For the Village Bus initiative, web-based booking and management of operations was a key enabler and benefitted both end users and operators. The on-demand pooling services in Spain identified app-enabled booking for the Table 17.1 Factors contributing to successful rural DRT Mobility service

Factor that contributed to success

Example good practice case3

Demand responsive transport

Close working between municipalities, service providers and public transport operators

Flexitec – Belgium, DRT Tejo – Portugal, Regiotaxi – The Netherlands, flexible mobility services – Bulgaria, DRT Castilla y Leon – Spain Village Bus – Sweden, Prontobus – Italy, on-demand pooling services – Spain Suffolk Links – UK, WRDRT – Australia, Regiotaxi – The Netherlands Bummelbus in Luxemburg, Village Bus in Sweden, DRT Castilla y Leon – Spain

Technology

Addressing social exclusion

Close attention to local needs

231

John D. Nelson and Steve Wright

user and the efficiency of the software algorithm for planning services as being important factors in ensuring customer satisfaction and operational efficiency respectively. The Prontobus initiative was essentially a technology upgrade for an existing DRT operation. The new software allowed for real-time information provision and booking services for users by app or web. The software also enhanced back-office operational efficiency, enhanced service monitoring opportunities and enabled communication with the driver via a tablet. A number of cases identified addressing social exclusion as an important success factor (Suffolk Links, UK; Western Region DRT [WRDRT], Australia; Regiotaxi, The Netherlands); indeed, social inclusion is a motivation behind many FTS schemes. For WRDRT in Australia, services were targeted towards socially disadvantaged groups to enhance their mobility and level of social inclusion. This approach yielded social, health and safety benefits – the latter due to the reduction in the requirement to drive for people who were losing confidence in their driving abilities. Similarly, for Suffolk Links, the focus of operations was on households in areas overlooked by public transport. In both cases, a focus on low-cost fares was deemed important to ensure patronage by low-income groups. Close attention to local needs was a success factor for all DRT good practices. Of particular note in this regard was the Bummelbus in Luxemburg, where services were extended to support participation in after-school activities by children. The Village Bus in Sweden utilised volunteer drivers with clear understanding of local needs. For DRT Castilla y Leon in Spain, effective design and planning of services was deemed a critical success factor and involved close community engagement in identifying gaps in public transport service provision and ensured shared services responded directly to local needs. For the majority of rural services, funding the service remains the biggest barrier to implementation and long-term sustainability. The cost of provision far exceeds the revenue generated. In rural environments, there is little scope to increase revenues significantly, since the passenger numbers are relatively low and fares are often equivalent to conventional bus, as these services are often bus replacements serving essential trips. This differs from the Microtransit emerging in urban/suburban areas, which offer a more responsive service and which can be viewed as closer in nature to a taxi. In rural areas, total fare revenue frequently does not cover driver wages, and subsidy requirements are often required for 60% of operating costs. Integrating statutory health and education transport demands into the design of the flexible service can bring in extra funding for the service from these public-sector budgets (Mounce et al., 2018). Where this is not possible, it becomes necessary to reduce operating costs through use of smaller vehicles (Wright, 2013) and lowercost providers (Mulley et al., 2012). In particular, there have been an increasing number of rural FTS services delivered by community transport providers, often using volunteer drivers and volunteer booking staff. In evaluating good practice, it is important to capture the ‘hidden’ social benefits, for example, for health or wellbeing. Rural mobility solutions are often more about meeting social needs rather than just providing transport. Indeed, it is the activity that the transport supports which is important, such as accessing a service. The cost to society of not providing such services is many times higher than the cost of transport. A missed hospital appointment due to a lack of transport is valued at around £126 (at 2015 prices). If people are unable to leave their homes, services especially for elderly and disabled may be required to be brought to their home: For example, in rural Scotland, a doctor home visit is £100 more than a surgery appointment; a nurse home visit is £52 more than a surgery consultation; Meals on Wheels costs £20 per meal, and home help costs are £15 per hour (Nelson et al., 2017). Much of these costs could be avoided when people have transport to access services provided in their communities. 232

Flexible transport services

A rural mobility solution which is successful in one context cannot be assumed to be replicable in another (different) context with the same level of performance (Buchanan & Partners, 2003). The objective of a transferability analysis is to use the results of a “system” or solution implemented at one location to assess the validity of that system if implemented at an alternative site. An understanding of market environments (e.g. deregulated or not) is important and should be taken as a precondition in terms of knowledge required by those assessing the potential for transferability. Current market demands and objectives of local authorities and operators should also be established at the location in question. Finally, it is expected that the effectiveness of the current mobility offer, and the extent to which market demands will be met by the proposed new intervention within available funding streams, will be considered. Hence, it is better to think about ‘best fit’ rather than a literal transfer of solutions and to focus on adapting solutions to different local contexts.

Conclusions This chapter identified the considerable experience with flexible transport services gained over several decades. Nevertheless, there is still uncertainty around the most effective way to deploy FTS. This chapter has explored the development and impact of FTS across a variety of applications and concludes in this section with a summary of key findings and some suggestions for further research. An important finding for service design is that the most appropriate service type for an area depends on many factors – but mainly the demand or potential demand. This means that different business models are required in rural and urban/suburban areas. Importantly, the social value of FTS is high in rural areas (e.g. lifeline services), yet this is a factor often overlooked by many assessment and evaluation studies. There is evidence of commercial opportunities for FTS being pursued in urban/suburban areas where it can be positioned as an alternative to car use. For both the urban/suburban and the rural model of FTS delivery, there are several important considerations. It is advisable to build upon a core of existing services (i.e. to use FTS to complement, not supplement). It is noticeable that Microtransit start-ups have often been in direct competition with public transport rather than complementing it, offering a higher-quality service for slightly higher fares. This reinforces the need for partnership with local transport authorities to promote integration with public transport services or replace poorly performing public transport services and to compete with the private car rather than duplication that competes with public transport services. To achieve this, it is important to design a flexible service that responds to identified community needs (this requires the need to earn community support and involve community leaders and to communicate to all stakeholders). Another requirement is to choose the most appropriate vehicles (taking account of size and need for wheelchair access, environmental considerations and cost). Finally, it is essential to ensure that the technologies adopted are suitable for the local context and user requirements and do not create barriers (travellers do not appreciate having to download and book using a different application for adjacent trials). Despite the lessons learned from the great wealth of experience with FTS, there are a number of open questions which would benefit from further research. Prominent amongst these is the need to design a network which truly integrates FTS and fixed-route services (some of which in future may be operated by autonomous vehicles). In this respect, achieving fare integration between FTS and fixed-route services remains important, as does the need to raise awareness about what FTS is and how it works. More research into the commercial viability of Microtransit services is required, including research into the best areas to implement such 233

John D. Nelson and Steve Wright

services. Mobility as a Service (MaaS) systems might help overcome the fare integration issue, but more research is needed on how to best include non-timetabled (flexible) public transport services within MaaS. This is more likely for Microtransit solutions, where an ‘always available’ service exists. In the short term, these future requirements of FTS design and operation should be seen in the light of the impact of the COVID-19 pandemic, which is likely to make travellers more wary of travelling in smaller confined spaces, although the ability to book ahead may be considered an advantage. It will be necessary to consider the effect that COVID-19 will have on market demands and operator supply and whether this will lead to more or less interest/need for FTS solutions. For example, it may be that passengers spread in small groups across smaller vehicles (FTS) are better than larger numbers of passengers socially distanced on fewer large vehicles (conventional buses) – at greater cost of provision but with greater flexibility of operation.

Notes 1 The proportion of 18–75-year-olds owning or having access to a smartphone in the United Kingdom increased from 52% in 2012 to 87% in 2018. Nearly 9 in 10 smartphone users (87%) use their phones for travel purposes, with navigation and route planning being the most popular uses. Source: Department for Transport. (2018). Transport and technology public attitudes tracker. Wave 3. https://assets. publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/803347/trans port-and-transport-technology-public-attitudes-tracker-wave-3-report.pdf 2 These figures should be treated with caution, since they represent a snapshot within the first year of operation but are nevertheless indicative. 3 Available at: https://ruralsharedmobility.eu/good-practices/

References Ambrosino, G., Nelson, J. D.,  & Romanazzo, M. (Eds.). (2004). Demand responsive transport services: Towards the flexible mobility agency. ENEA. www.enea.it/en/publications/abstract/Demand-ResponsiveTransport-Services-towards-the-Flexible-Mobility-Agency Brake, J. F., Mulley, C., & Nelson, J. D. (2006). Good practice guide for demand responsive transport services using telematics. Department for Transport. Newcastle upon Tyne, University of Newcastle upon Tyne. https:// eprint.ncl.ac.uk/file_store/production/12981/B9C57C8D-4AD9-4ED6-80E7-C05258B34516.pdf Brake, J. F., Mulley, C., Nelson, J. D., & Wright, S. (2007). Key lessons learned from recent experience with Flexible Transport Services. Transport Policy, 14(6), 458–466. doi:10.1016/j.tranpol.2007.09.001 Buchanan and Partners. (2003). Transferability of best practice in transport policy delivery: Final report. Prepared for Scottish Executive by Buchanan and Partners. Currie, G.,  & Fournier, N. (2019). Why most DRT/Micro-Transits fail  – What the survivors tell us about progress. 16th International Conference on Competition and Ownership in Land Passenger Transport (Thredbo 16). Singapore. Davison, L., Enoch, M., Ryley, T., Quddus, M., & Wang, C. (2014). A survey of demand responsive transport in Great Britain. Transport Policy, 31, 47–54. doi:10.1016/j.tranpol.2013.11.004 Department for Transport. (2018). Transport and technology public attitudes tracker. Waves 3 summary report. https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/ file/803347/transport-and-transport-technology-public-attitudes-tracker-wave-3-report.pdf Engels, D., & Ambrosino, G. (2004). Service typologies and scenarios. In G. Ambrosino, J. D. Nelson, & M. Romanazzo (Eds.), Demand responsive transport services: Towards the flexible mobility agency (pp. 55–73). ENEA. Enoch, M., Potter, S., Parkhurst, G., & Smith, M. (2004). INTERMODE: Innovations in demand responsive transport. Department for Transport and Greater Manchester Passenger Transport Executive. Final report. Department for Transport, London. www.semanticscholar.org/paper/INTERMODE%3AInnovations-in-demand-responsive-Enoch-Potter/ce5b135208d2438ed04cc7939e071a4994fc8aae Finn, B., Ferrari, A., & Sassoli, P. (2004). Goals, requirements and needs of users. In G. Ambrosino, J. D. Nelson, & M. Romanazzo (Eds.), Demand responsive transport services: Towards the flexible mobility agency (pp. 33–54). ENEA.

234

Flexible transport services Lewis, K. (2019). The delicate balance of microtransit: Service vs. efficiency, or transformational opportunity vs. just the latest fad? Passenger Transport, 77(15). https://www.apta.com/wp-content/uploads/ PT_081919_Microtransit-in-depth_reprint-002.pdf Mageean, J. F., & Nelson, J. D. (2003). The evaluation of demand responsive transport services in Europe. Journal of Transport Geography, 11(4), 255–270. doi:10.1016/S0966-6923(03)00026-7 Mounce, R., Wright, S., Emele, C. D., Zeng, C., & Nelson, J. D. (2018). A tool to aid redesign of flexible transport services to increase efficiency in rural transport service provision. Journal of Intelligent Transportation Systems, 22(2), 175–185. doi:10.1080/15472450.2017.1410062 Mulley, C., & Clifton, G. (2016). Decision making in flexible transport. The importance and application of the “golden rule”. In C. Mulley & J. D. Nelson (Eds.), Paratransit: Shaping the flexible transport future (pp. 153–166). Emerald Publishing Limited. doi:10.1108/S2044-994120160000008008 Mulley, C., & Daniels, R. (2012). Quantifying the role of a flexible transport service in reducing the accessibility gap in low density areas: A case-study in North-West Sydney. Research in Transportation Business and Management, 3, 12–23. www.sciencedirect.com/science/article/pii/S2210539512000077 Mulley, C., & Nelson, J. D. (2009). Flexible transport services: A new market opportunity for public transport. Research in Transportation Economics, 25, 39–45. doi:10.1016/j.retrec.2009.08.008 Mulley, C., & Nelson, J. D. (Eds.). (2016). Paratransit: Shaping the flexible transport future. Emerald Publishing Limited. www.emerald.com/insight/publication/doi/10.1108/S2044-994120168 Mulley, C., Nelson, J. D., Teal, R., Wright, S. D., & Daniels, R. (2012). Barriers to implementing flexible transport services: An international comparison of the experiences in Australia, Europe and USA. Research in Transportation Business and Management, 3, 3–11. doi:10.1016/j.rtbm.2012.04.001 Nelson, J. D., & Phonphitakchai, T. (2012). An evaluation of the user characteristics of an open access DRT service. Research in Transportation Economics, 34, 54–65. doi:10.1016/j.retrec.2011.12.008 Nelson, J. D., & Wright, S. D. (2012). Flexible transport services. Special Issue of Research in Transportation Business and Management, 3. www.sciencedirect.com/journal/research-in-transportation-business-andmanagement/vol/3 Nelson, J. D.,  & Wright, S. D. (2016). Flexible transport management. In M. Bliemer, C. Mulley,  & C. Moutou (Eds.), Handbook on transport and urban planning in the developed world (pp. 452–470). Edward Elgar Publishing. Nelson, J. D., Wright, S. D., Masson, B., Ambrosino, G., & Naniopoulos, A. (2010). Recent developments in flexible transport services. Research in Transportation Economics, 29, 243–248. doi:10.1016/j. retrec.2010.07.030 Nelson, J. D., Wright, S. D., Thomas, R.,  & Canning, S. (2017). The social and economic benefits of community transport in Scotland.  Case Studies on Transport Policy, 5(2), 286–298. doi:10.1016/j. cstp.2017.01.001 Pettersson, F. (2019). An international review of experiences from on-demand public transport services (K2 Working Paper 2019:5). www.k2centrum.se/sites/default/files/fields/field_uppladdad_rapport/on-demand_pt.pdf SMARTA. (2020). https://ruralsharedmobility.eu/ TCRP, National Academies of Sciences, Engineering, and Medicine. (2008). Guidebook for measuring, assessing, and improving performance of demand-response transportation. The National Academies Press. doi:10.17226/23112 Transport for NSW. (2018). Future transport 2056 strategy. NSW Government. https://future.transport.nsw. gov.au/sites/default/files/media/documents/2018/Future_Transport_2056_Strategy.pdf Transport for NSW. (2020). https://transportnsw.info/travel-info/ways-to-get-around/on-demand Volinski, J., & Raton, B. (2019). Microtransit or general public demand – Response transit services: State of the practice. A synthesis of transit practice. TCRP Synthesis 141, The National Academies Press. doi:10.17226/25414 Weckström, C., Ullah, W., Mladenović, M. N., Nelson, J. D., Givoni, M., & Bussman, S. (2018). User perspectives on emerging mobility services: Ex post analysis of Kutsuplus pilot. Research in Transportation Business and Management, 27, 84–97. doi:10.1016/j.rtbm.2018.06.003 Westerlund, Y. (2016). Development and status for large-scale demand responsive transport. In C. Mulley & J. D. Nelson (Eds.), Paratransit: Shaping the flexible transport future (pp. 58–73). Emerald Publishing Limited. doi:10.1108/S2044-994120160000008004 Wright, S. (2013). Designing flexible transport services: Guidelines for choosing the vehicle type. Transportation Planning and Technology, 36(1), 76–92. doi:10.1080/03081060.2012.745757 Wright, S., Emele, C. D., Fukumoto, M., Velaga, N. R., & Nelson, J. D. (2014). The design, management and operation of flexible transport systems: Comparison of experience between UK, Japan and India. Research in Transportation Economics, 48, 330–338. doi:10.1016/j.retrec.2014.09.060

235

18 INFORMAL PARATRANSIT IN THE GLOBAL SOUTH Roger Behrens, Saksith Chalermpong and Daniel Oviedo

Introduction The term ‘paratransit’ is used in this chapter to describe flexible modes of passenger transportation that do not follow fixed schedules or adhere to specified service spans. In the Global North, paratransit services are most commonly associated with demand-responsive transport systems provided for persons with disabilities (see also Chapter 17). In the Global South, however, paratransit services are provided for the general population. The extent to which these services operate with public authority sanction, declare income, and are tax compliant determines whether they are ‘informal’. The scale of paratransit operations is also significantly larger in the Global South. With notable Latin American exceptions, in many, if not most, cities, paratransit services carry the majority share of the passenger market. Informal paratransit is therefore the norm rather than the exception, and the attention it receives in the academic literature relative to mass fixed-route public transport is disproportionately small. While providing essential access for large portions of city populations, the quality of paratransit services in the Global South is often poor (Cervero, 2000; Cervero  & Golub, 2007). While the nature and severity of service quality problems are no doubt contextually variant, there are some commonalities: drivers can compete aggressively for passengers in the road space, overloading vehicles and disobeying traffic laws; more lucrative routes can be overtraded, while service on less lucrative routes or times of the day is not supplied; vehicles can remain in service too long, becoming increasingly unsafe and polluting; and in-lane vehicle boarding and alighting can reduce already limited road capacities. The scope of this chapter encompasses two forms of paratransit. The first are public transport services, typically provided by minibuses with around 15 seats but ranging from 4-seater sedans to midi-buses with 35 seats. The second are for-hire services, typically provided by motorcycles but including a range of non-motorised and motorised two- to four-wheeler vehicle types. This rich diversity of services takes many different forms and is called many different names. ‘In the market’ competition sometimes results in elaborate decoration to attract passengers, as can be found amongst ‘jeepneys’ in Manila, ‘jingle buses’ in Karachi, and ‘tap taps’ in Port-au-Prince. Colloquial names are often derived from the initial fare (e.g. Mexican ‘peseros’ and Tanzanian ‘daladalas’). The chapter’s scope excludes ‘formal’ forms of paratransit (e.g. ‘dial-a-ride’ buses for 236

Informal paratransit in the Global South

the disabled and metered taxis) insofar as they operate with government sanction and are fully compliant with tax and employment regulations. The intent of the chapter is to synthesise what is known of informal paratransit services in cities of the Global South, to consider what causes common service quality problems, and to discuss prevailing policy issues and emerging trends. The chapter is divided into several sections. The next three discuss business organisation, regulatory regimes, and operating practices, respectively, considering variations across the major regions of the Global South (Africa, Asia, and Latin America). This is followed by reflections on similarities and differences across these regions in discussions of current policy challenges and emerging trends before final conclusions are drawn.

Business organisation Current knowledge of informal paratransit business organisation  – how owners and drivers organise themselves collectively, who owns vehicles, and how drivers are remunerated – is discussed in the contexts of Africa, Asia, and Latin America in the following subsections.

Africa Paratransit organisation and business models in African cities take a variety of forms. Operators are organised collectively at varying geographical scales and with differing levels of public authority sanction and support. In the case of public transport services, in some countries, there are national or regional associations established to represent collective owner interests (e.g. the Matatu Owners Association in Kenya), some of which are supported and sanctioned by government agencies (e.g. the South African National Taxi Council) (Klopp & Mitullah, 2016; McCormick et al., 2016). A more ubiquitous form of collective organisation are route associations formed by operators to protect and self-regulate their markets. In South Africa, affiliation to a registered owner association is a mandatory requirement for an operating license, as is membership in either a co-operative or a transport management company for a public service vehicle license in Kenya (Behrens et al., 2017a). For-hire services also have some national operator associations to advance collective interests (e.g. the Uganda National Boda Boda Association). More prevalent are area or route associations (Bishop & Amos, 2015; Bishop et al., 2018; Ehebrecht et al., 2018), although the literature suggests that the prevalence of operators not affiliated to a route association is higher in motorcycle-taxis than in minibuses (Kisaalita & Sentongo-Kibalama, 2007; Mutiso & Behrens, 2011). Vehicle ownership patterns among public transport services range from owner-drivers to small fleet owners and in a few instances to collectively owned midi-buses (e.g. larger savings and credit co-operatives [SACCOs] in Nairobi) (Behrens et al., 2017a). In instances where vehicle owners are government or police officials, regulatory capture can occur (Klopp & Mitullah, 2016). Because of the smaller economies of scale and lower market entry costs associated with motorcycles, for-hire services typically have a higher proportion of owner-drivers, although this varies from country to country (Ehebrecht et al., 2018; Schalekamp & Saddier, 2019). Driver employment conditions are seldom formalised and can be exploitative (Diaz Olvera et al., 2016). Driver remuneration amongst public transport services is commonly based on a ‘target system’ in which drivers essentially keep cash fare revenue less fuel (and, if applicable, vehicle crew and minor maintenance) expenses and a daily or weekly vehicle rental payment to the owner (Behrens et al., 2016; Schalekamp & Saddier, 2019). Alternative, less common, remuneration models take the form of a commission based upon an agreed-upon portion of 237

Behrens, Chalermpong and Oviedo

weekly farebox revenue and salaries (typically with a ridership bonus incentive, as in the case of City Hoppa in Nairobi). The greater proportion of owner-drivers amongst for-hire operators means that such remuneration models are less pervasive, but in instances where motorcycle-taxi drivers are employed, similar target or commission systems prevail (Diaz Olvera et al., 2016).

Asia The business organisation of paratransit operators in Asia is determined in part by vehicle type. Smaller vehicles, such as human-powered pedicabs and motorcycle-taxis, tend to be owned by drivers. One exception is cycle rickshaws in India, where an operator may own a large fleet of vehicles (Kumar et al., 2016). For larger vehicles, including passenger vans and minibuses, greater capital requirements often mean that vehicle owners and drivers are different individuals, although the owner-driver model of large paratransit vehicles is also common (Chalermpong et al., 2016; Kumar et al., 2016). Operators who are fleet owners are sometimes registered as limited companies or partnerships, but some are private individuals. Some operators employ drivers to operate the vehicles, while others rent vehicles to individuals (Wicaksono et al., 2015; Kawaguchi et al., 2013). There are various forms of collective organisation. Route associations or driver unions are common for route-based services, such as passenger vans in Thailand and Indonesia, ‘jeepneys’ in the Philippines, and shared auto rickshaws in India (Chalermpong et al., 2016; Kawaguchi et al., 2013; Kumar et al., 2016; Rañosa et al., 2017) and location or area-based driver’s associations for for-hire services, such as motorcycle-taxis in Thailand and Philippines and tricycles in Cambodia (Ratanawaraha & Chalermpong, 2015; Phun et al., 2015). With the exception of motorcycle-taxis in Thailand, these associations have no legal status. One of the main reasons for the formation of associations is to gain an exclusive right to operate in a certain area. An individual with connections to the authority may establish an association and accept members in an area or route who wish protection (Chalermpong et al., 2016). Public officials, who are paid off by the ‘owner’ of the association, loosely enforce laws on association members and prevent non-members from operating in the area. Members may pay daily, weekly, or monthly dues to the association for operational expenditures, including parking space rental, utilities, insurance, and management fees (Ratanawaraha & Chalermpong, 2015). Paratransit operators who employ drivers normally pay a wage on a per-trip or daily basis (Cervero, 2000; Chalermpong et al., 2016). Some have a daily target of fare revenue, with a bonus payment to drivers who exceed the target. Drivers receive no other employment benefits but may receive a bonus for reliability or a good driving record (Ratanawaraha & Chalermpong, 2019). Operators may rent their association membership rights for short-term operations and may also sell the membership with association approval.

Latin America In Latin America and the Caribbean (LAC), forms of informal paratransit have diversified from minibuses and shared five-seater taxis or slightly larger vehicles (e.g. approximately ten-seater ‘camionetas’ in Caracas or ‘gualas’ in Cali) to more individualised services such as motorised and human-powered two- and three-wheelers (Heinrichs  & Bernet, 2014; Jauregui-Fung et al., 2019). While the first type of services has been an almost ubiquitous feature of LAC cities, the latter have emerged more recently as a response to limited connectivity and gaps in the market (Cervero, 2000; Oviedo & Titheridge, 2016; Sengers& Raven, 2014). Like the organisation of paratransit in Africa and Asia, the model of separate vehicle owner 238

Informal paratransit in the Global South

and driver(s) is common, with many owners and even more drivers making it an industry difficult to co-ordinate and negotiate with. Three models are identified across the breadth of informal paratransit supply in the LAC context: owner-driver (where the vehicle owner operates the vehicle), owner-employer and driveremployee (where the owner hires the driver), or owner-employer and driver-renter (where the owner rents the vehicle to the driver). Such arrangements are, however, more common in the case of low-capacity vehicles providing unrouted services such as shared taxis and motorcycleand bicycle-taxis (Gamble  & Puga, 2017; Heinrichs et  al., 2017). In the more consolidated routed services, larger fleet owners are registered as private firms, or associations of owners and drivers form to strengthen their capacity to negotiate with local authorities. Similarly, in some cases such as the jitney service operators in Peru, Mexico, and Colombia, associations have been created to represent the interests of specific groups and improve their standing to tender for formalised services (Hidalgo & Huizenga, 2013; Venter et al., 2019; Willoughby, 2013). Employment conditions for many workers are precarious, resulting from decades of operational arrangements with perverse incentives and uncontrolled competition. In a handful of cases of large fleet owners and registered companies and associations, formal employment with social security is provided. However, in most cases in the owner-employer and driver-employee/ driver-renter category, vehicles are rented to drivers for a fixed daily fee with all revenues after rent and operating costs determining driver income. This incentivises drivers to work long hours and compete for passengers in what has been called locally guerra del centavo (‘the penny war’) and leads to rapid deterioration of vehicles, decreased quality of service, and marginal profits (Ardila &Rodríguez, 2000; Ardila, 2007; Estache and Gómez-Lobo, 2005). In Tláhuac, a recent study has found that motorcycle-taxi drivers have no employment rights or access to social safety nets, with adverse impacts on quality of life and work (Berrones-Sanz, 2018).

Regulation Current knowledge of informal paratransit regulation – what regulatory regimes are imposed by public authorities and how operator associations regulate themselves – is discussed in the contexts of Africa, Asia, and Latin America in the following subsections.

Africa In most African cities, paratransit operations are subjected to two regulatory systems. The first takes the form of ‘self-regulation’ imposed by operator associations. The primary purposes of self-regulation are usually market entry control, although this is often related more to kinship and community than to balancing supply with demand and protecting routes from competitors. On occasions, route protection can become violent (e.g. the ‘taxi wars’ in South Africa following market entry deregulation in the late 1980s) (Dugard, 2001). In some instances, recognising that there is overtrading, associations seek to ration member access to the market through controlling vehicle order in rank queues (e.g. at Mitchells Plain in Cape Town) (Behrens et al., 2017b). In some cities, operator associations collude in the setting of fares, while in others, drivers engage in opportunistic dynamic pricing. In better-organised and better-resourced associations, driver recruitment and management and vehicle tracking functions can be transferred from individual vehicle owners to association officials (e.g. in some larger SACCOs operating out of Nairobi) (Behrens et al., 2017a). The transport management companies in Nairobi (e.g. Kenya Bus Services and City Hoppa) also enter into franchise agreements with vehicle owners concerning branding and operating rules (McCormick et al., 2013). 239

Behrens, Chalermpong and Oviedo

The second regulatory system takes the form of the rules imposed by the public authority. The complexity of regulatory regimes varies widely, according to the powers, capacity, and resources of the authority. In least-resourced contexts, open markets prevail, and regulation is reduced to whatever traffic law enforcement may exist. A common regulatory regime is the issuing of quantity operating licenses via licensing boards, but in the absence of adequate market entry enforcement, this defaults to a form of quality licensing relating to vehicle specification and driver requirements. Some authorities set fares (e.g. ‘chapas’ fares in Maputo and ‘daladala’ fares in Dar es Salaam). Some innovative regulation regimes take the form of franchises (e.g. in Dakar as a condition for vehicle renewal funding), service contracting (e.g. in Mitchells Plain, Cape Town, as part of a network rationalisation), and performance incentives (e.g. a driver financial incentive scheme in Durban) (Kumar  & Diou, 2010; Saddier et  al., 2019; Schalekamp & Klopp, 2018).

Asia In many Asian countries, including India, Indonesia, the Philippines, and Thailand, fixed-route paratransit services are regulated by local authorities (Institute for Transportation and Development Policy, 2019; Kumar et al., 2016; Ratanawaraha & Chalermpong, 2018; Regidor et al., 2009). To obtain an operating license or a franchise, similar to bus regulation regimes, the operator must meet various conditions, including vehicle standards and maintenance record keeping, driver’s training and licensing, and parking permits. The authorities nominally plan and regulate routes, stops, service span, and number of vehicles, as well as fare. The main objective of the regulation in some cases is to prevent traffic congestion caused by the operation of paratransit vehicles (Arnado et al., 2017). For-hire paratransit operations, such as motor tricycles and pedicabs, are also regulated in some countries, mostly to limit their numbers and to control congestion (Wicaksono et al., 2015). However, motorcycle-taxis are unregulated in most countries, except for Thailand, where an operating license and a commercial driving license are required (Cerio, 2017; Ratanawaraha & Chalermpong, 2015). In lower-income countries where a regulatory system has not yet been adopted, such as Laos and Cambodia, regulation of paratransit operations is limited to vehicle registration and driving licenses. In practice, however, even in countries with a formal regulatory system, such as Thailand, India, and Indonesia, regulations are loosely enforced by officials due to, among other reasons, the difficulty of enforcement, lack of personnel, corruption, and regulatory capture (Institute for Transportation and Development Policy, 2019; Kumar et  al., 2016). For this reason, in countries with high levels of corruption, unauthorised paratransit operations can be rife. In addition, vehicles or drivers licensed for one service but operating another are also common. To be able to operate in a certain area, these unauthorised operators must still belong to an operator’s association in that area and must generally follow the rules set by that association. In countries with weak regulation, misuse of vehicles is rampant, including modification and overloading of passenger vans and extending motorcycle-taxi passenger seating through a wooden plank (Cerio, 2017; Kumar et al., 2016). Self-regulation is usually practiced at the operator association level. For example, in India, drivers operating on the same routes make collective fare-setting decisions, subject to some government regulation (Kumar et al., 2016). In Thailand, since most associations are dominated by the association’s owner (i.e. the well-connected person who established the association), rule-setting decisions, including to accept new members, to set fare, and dispatching rules, are generally up to the owner, but members are often consulted (Ratanawaraha & Chalermpong, 2015, 2019). In some associations, these rules are made by a board of member representatives. In 240

Informal paratransit in the Global South

yet other associations, the decisions are made by all members through a vote. Failure to follow the rules, in most cases, can result in expulsion from the association.

Latin America In LAC, public authorities have developed regulations and controls and promoted initiatives to integrate semi-formal buses and minibuses into a fully integrated system with clear targets of access, affordability, and service quality (Rodriguez et al., 2017; Willoughby, 2013). Such transitions have required an overhaul of operations, as well as a reorganisation of incumbent owners and drivers into co-operatives or unions to guarantee their participation after strong tensions and demonstrations (Rodriguez et al., 2017). An example of such a reform is the vehicle circulation restriction for motorised rickshaws in Barranquilla, where three-wheelers are available only in two colours (orange and blue) and this distinction is used as a basis to restrict the operation of half the fleet during different days of the week. Self-regulation of operations is also commonplace in both routed and unrouted informal and semi-formal services. In local paratransit in informal settlements on the peripheries of cities such as Bogota, community leaders define the alignments of the main routes serving their neighbourhoods, usually seeking to maximise coverage but also sometimes seeking to exclude neighbourhoods because of frictions between local leaders (Oviedo & Titheridge, 2016). In Quito, evidence suggests that self-regulation leads to increased service quality attention to features like fares, waiting time, vehicle condition, and safety (Gamble & Puga, 2017).

Operation Current knowledge of informal paratransit operating practices  – how route networks are arranged, where passengers board and alight, when and how frequently service is provided, and how fares are set and collected – is discussed in the contexts of Africa, Asia, and Latin America in the following subsections.

Africa Few studies have been undertaken on paratransit network organisation and operation practices. Amongst public transport services, the extent of paratransit service networks has been mapped in several cities (e.g. AccraMobile in Accra, Experimental Transit in Cairo, Mapa Dos Chapas in Maputo, and Digital Matatu in Nairobi), revealing extensive geographical coverage (Klopp & Cavoli, 2017; Saddier et al., 2016). Fewer studies (in Accra, Cape Town, and Lubumbashi) have collected data on route typologies and operating practices such as headways, layovers, boarding and alighting, spans, and fare collection (Saddier & Johnson, 2018; Du Preez et al., 2019; Behrens et al., 2017b; Kerzhner & Martens, 2018). These studies reveal a high level of demand responsiveness. Consequently, even when prohibited by operating license restrictions, route deviations occur in response to passenger requests, and service headways during peak periods tend to be small. While on-route stopping is common, boarding and alighting tends to be concentrated at ranks, as drivers are reluctant to risk an unviable trip by departing without a full passenger load. In the off-peak, passenger waiting times can therefore lengthen considerably as vehicles fill more slowly and layover times are longer. Service spans are usually determined by periods when trip-making is profitable; hence, spans can differ from those of scheduled mass public transport. While there have been numerous attempts to introduce cashless systems, the dominant means of fare collection remains cash (Tinka & Behrens, 2019). 241

Behrens, Chalermpong and Oviedo

Amongst for-hire services, service networks have also been mapped in some cities (e.g. Dar es Salaam, Kampala, and Yaoundé), but there is less literature on detailed operating practices (Goletz & Ehebrecht, 2019; Evans et al., 2018; Kemajou et al., 2019). Trip distances tend to be shorter than minibus services, seldom further than 6 km, providing door-to-door service as well as links to higher-capacity modes (Ehebrecht et al., 2018; Goletz & Ehebrecht, 2019). Both passengers and goods are transported. Drivers operate out of numerous small stations distributed across the city, and given a prevalence for overtrading, inactivity rates can be high (Kemajou et al., 2019; Schalekamp & Saddier, 2019). While fare collection remains predominantly cashbased, cashless fare collection (CFC) systems (e.g. Gozem in Lomé, SafeBoda in Kampala, and YegoMoto in Kigali) have achieved greater success than minibus paratransit (Tinka  & Behrens, 2029). The greater proportion of owner-drivers means that the battle for farebox control between owners and drivers, which has disrupted CFC initiatives in minibus paratransit, is not as prevalent.

Asia Fixed-route paratransit services in Asia are generally operated out of major activity centres, such as markets, shopping malls, transportation hubs, and commercial and employment centres in central business districts, as well as community centres in suburban areas. Routes are usually concentrated along arterials that serve commuter travel demand. Route lengths may range from less than 5 to over 20 km in the case of jeepneys in Philippines (Regidor et al., 2009). In Thailand, the lengths of passenger van routes range from 15 to 50 km. In some countries, such as Cambodia and Indonesia, pick-up and drop-off can occur anywhere along the route (Cities Development Initiatives for Asia, 2011), while in others, such as Thailand and Philippines, boarding and alighting should occur, at least on paper, only at designated locations. In countries where operating licenses are required for fixed-route services, operators must follow the designated routes or risk being fined by the authority. Paratransit services are generally not operated according to a predefined schedule but dispatched on a fill-and-go basis, although some associations may specify a maximum headway to limit passengers’ waiting time. Driver unions in some fixed paratransit routes in India collectively determine minimum service frequency and penalise non-compliance (Kumar et al., 2016). Fare collection is usually done on board by cash payment directly to the driver or sometimes to a conductor. In Thailand, the fare collection of many passenger van services is done at the boarding station by a cashier of the route association, also in cash. Like fixed-route paratransit operators, for-hire paratransit operators usually wait for passengers at major activity centres. In addition, some operators cater to tourists and make agreements with hotels to provide services for guests (Cities Development Initiatives for Asia, 2011). Motorcycle-taxis in Thailand are operated out of a ‘win’, a location-based association to which the operator belongs. Members of a motorcycle-taxi win wear an orange-coloured vest that shows the ‘win’ location (Ratanawaraha  & Chalermpong, 2015). Each driver must enter a queue and is dispatched to serve customers in that order. According to a self-imposed rule, motorcycle-taxi drivers are not allowed to pick up a passenger outside their ‘win’ territory and must return to their own station and re-enter the queue to pick up the next customer. Violent conflicts between members of different ‘wins’ often break out when this operating rule is violated. Motorcycle-taxi drivers in other countries, such as Vietnam and the Philippines, are far less self regulated and can pick up and drop off passengers anywhere. These operators must usually pay protection fees to government officials to avoid law enforcement. Despite having a

242

Informal paratransit in the Global South

license to operate, motorcycle-taxi drivers in Thailand also have to pay off officials, through the ‘win’ association, to avoid prosecution.

Latin America Studies of informal and semi-formal paratransit in LAC have mapped routed services at different scales and in different contexts, highlighting the adaptability and demand-responsive nature of their operation. In cities such as Lima and Quito, paratransit routes respond to needs to access areas unserved by formal transport supply, as well as appearing in corridors where they engage in direct competition with formal services in the absence of sufficient enforcement (Gamble & Puga, 2017; Jauregui-Fung et al., 2019). Analyses of the evolution of routed semiformal and informal paratransit in the region suggest that there is little restriction on operating times, frequencies, and headways and that whenever pricing is regulated, groups of operators tend to apply pressure on local authorities to adjust fares to cover increasing operating costs (Jauregui-Fung et al., 2019; Rodríguez Baquero et al., 2003; Yañez-Pagans et al., 2018). There is also a high incidence of support economies arising around informal paratransit that serve as support structures for their operation. Many minibus routes have ‘calibrators’ located at different points of the journey, who keep track of the time vehicles pass a given point (usually an intersection) and inform the next driver of the estimated headway to adjust speed and frequency (Parra et al., 2005). Research exploring informal paratransit without fixed routes suggests that these services play different roles, depending on the level of operator consolidation, available infrastructure, topography, and sociospatial distribution of the population and income (Heinrichs et al., 2017; Oviedo & Dávila, 2016; Suárez et al., 2016). Moreover, there is a high degree of complementarity and interoperability between shared taxis, jitneys, motorcycle-taxis and bicycle-taxis, and higher-capacity modes such as bus rapid transit (BRT) systems (Heinrichs et  al., 2017). In mid-sized Colombian cities, the services motorcycle-taxis provide range from last-mile trips to the majority of city-wide trips, such as in Monteria (Goldwyn & Vergel-Tovar, 2018). Motorcycle-taxis in Mexico tend to operate without insurance and often do not provide helmets to passengers, resulting in high vulnerability and an inability to deal with post-crash liability (Berrones-Sanz, 2018). In Colombia and Cuba, bicycle-taxis are forced to negotiate for road space and compete for short-distance passengers. In Colombia, they set their own fares, while in Cuba, there is a complex negotiation with local authorities, despite being considered informal (Heinrichs et al., 2017; Warren & Ortegon-Sanchez, 2016).

Policy challenges In the context of the common service quality problems outlined earlier, across all regions of the Global South, there have been public sector plans to either formalise incumbent operators or replace them with formal, scheduled public transport undertakings, often BRT. Institutional reorganisation to facilitate public transport reform has also been common. In Thailand, for example, due to high crash injury and fatality rates, the government announced the phasing out of passenger vans that are licensed as fixed-route transit vehicles (Mahitthirook & Nanuam, 2017). In the Philippines, the Public Utility Vehicle Modernisation Program introduced in 2017 required jeepney operators to replace vehicles in poor condition and sought to devolve route planning and franchising responsibilities from the central to local governments (Land Transportation Franchising and Regulatory Board, 2017).

243

Behrens, Chalermpong and Oviedo

Experiences from the first cities in Africa to attempt to replace paratransit services (e.g. Rea Vaya in Johannesburg, MyCiTi in Cape Town, and UDART in Dar es Salaam) suggest that comprehensive replacement is neither feasible nor perhaps even desirable, particularly given the demand responsiveness of paratransit entrepreneurs compared to formal, scheduled modes in the context of forecast rates of unprecedented African urbanisation. Thus, as an alternative to the replacement of informal paratransit, a key policy challenge is now improving paratransit integration into complementary ‘hybrid’ public transport networks, composed of trunk trains, large buses, and paratransit feeders. In instances where smaller paratransit vehicles provide linehaul services better suited to scheduled larger vehicle operations, the policy process to introduce willing changes in service routing is likely to be fraught with difficulty both in terms of operator willingness as well as entrenched and outdated regulations. Irrespective of whether city authorities attempt to integrate paratransit with new scheduled mass public transport services, a further policy challenge is how best to improve the quality of paratransit services. At the root of service quality problems lies an exploitative remuneration model in which drivers are incentivised to drive recklessly as they attempt to fit as many service trips in the peak as possible and a cash-based business in which vehicle depreciation is not costed as an operating expense and cost inefficiencies are difficult to identify. In the case of motorcycle-taxis in particular, improving road safety through improved compliance with traffic laws presents a further considerable challenge. It is unclear at present whether policy interventions aimed at mandatory change, as in the case of contracts with associations imposed on members, or voluntary change, as in the case of financial incentives for desirable operating behaviour, are likely to be more successful. Current policies of these types require careful monitoring and evaluation. While there is broad support for paratransit formalisation and replacement in Asian cities, implementation requires strong political commitment, legal power, expertise, and resources that may not be available. For example, although provisions for financial assistance such as soft loans are usually included in the reform proposal, small paratransit operators generally oppose fleet modernisation, citing an unacceptable financial burden (David, 2019). A lack of political will to carry out the reform also threatens implementation, as is evident in the case of Thailand’s uncertainty in its policy to ban passenger vans (Hongtong, 2019). To further complicate the situation, institutional barriers are also common, since the reforms often involve many agencies in different ministries, in which silo thinking generally prevails (Wu & Pojani, 2016). Policy interventions also often must occur within the context of debilitating corruption and compromised enforcement capacity (Gwilliam, 2011; Klopp  & Mitullah, 2016; Rasmussen, 2012). In some instances, it makes greater economic sense for operators to remain noncompliant, because bribes equivalent to fines are elicited by traffic police even when operators are compliant. Further, any policy attempt to realign paratransit service routes to a more complementary ‘hybrid’ public transport network or to change operating practices to improve service quality requires a sound understanding of the base business models and operating practices. Without a grounded understanding of base conditions, windows of opportunity, path dependencies, and vested interests capable of derailing policy interventions are unlikely to be recognised. The structural workings and business models of paratransit services are often insufficiently understood to inform the design of public transport reform projects. With respect to formalisation and replacement, Latin American cities have experienced more success. Experiences like those of Transantiago in Santiago and Transmilenio in Bogota have facilitated the transfer of ideas around how to deal with various degrees of resistance and the operational and social challenges in achieving full regularisation (Hidalgo & King, 2014). Some cities in LAC have introduced a more active role for authorities in determining the structure of 244

Informal paratransit in the Global South

the transport network, levels of service, and frequencies, forcing a separation between revenue collection and operating activities (Estache & Gómez-Lobo, 2005). As city authorities improve their ability to address structural deficits in infrastructure and their capacity for planning and delivery of public transport services, new policy positions have been adopted in relation to informal public transport. In Lima, recent efforts at introducing new forms of public transport governance and regulation have operationalised new standards for pricing and employment and improved enforcement (Jauregui-Fung et al., 2019). However, initial efforts on the ground, in the form of pilot corridors, have been met with resistance by both passengers and incumbent operators. One of the main challenges that new policies and regulations face is potential resistance from incumbent informal and semi-formal operators to be absorbed or replaced by integrated public transport systems and to lose control of farebox revenue collected in cash. An additional policy challenge in the LAC context is the role of informal and semi-formal operators in the adoption of new agendas for sustainable development and crash fatality reduction (Duduta et al., 2015; Hidalgo & Huizenga, 2013; Jauregui-Fung et al., 2019). Policies such as Vision Zero, which have been adopted in cities like Bogota, involve stringent controls over speed and safety in motorised transport, as well as calling for a combination of policy, design, enforcement, and regulation that are likely to affect paratransit. The successful adoption of such agendas will require high levels of recognition, participation, and negotiation between informal operators and public authorities in order to jointly develop strategies that contribute to the overarching objectives while reducing unintended consequences such as loss of livelihoods.

Emerging trends Across the Global South, the dominant emerging trend amongst paratransit services is perhaps the mushrooming of potentially disruptive digital platforms related to their operation and use. The widespread availability of mobile phones in Asian cities has altered the organisation and operation of paratransit services. Even before smartphones became ubiquitous, paratransit operators in some cities made use of cellular phones for vehicle dispatching and passenger hailing. The recent diffusion of smartphones has enabled ride-hailing apps to enter the market. Ride-hailing operators range from large regional players, such as Grab and GO-JEK, to small local operators, such as PassApp (Cambodia) and Angkas (Philippines) (Silalahi et  al., 2017; Phun et al., 2018). These operators provide not only conventional sedan taxi ride-hailing services but ride-hailing services for motorcycle-taxis (GO-JEK) and other paratransit services (PassApp) as well. In a 2019 survey of African cities, Boutueil and Quillerier (2020) identified in the region of 135 digital platforms. Briter (2018) similarly identified over 150 mobility companies on the continent. These digital platforms include, amongst others, journey planning (e.g. Ma3Route in Nairobi), ride-hailing (e.g. Swvl in Cairo), and cashless fare collection (e.g. Gona in Lagos). As in Asia, this growth has been fuelled by the diffusion of smartphones, accessible geo-location technologies, and portable connectivity (Schalekamp & Saddier, 2019). Given the robust competition between emerging start-ups for market share, inevitably there will be a culling of digital platforms as some fail. As noted earlier, in the case of CFC initiatives, for instance, most initiatives amongst minibus services have failed (e.g. Bebapay in Nairobi, Faircard in Pietermaritzburg), while initiatives amongst motorcycle-taxis have proven more likely to endure (e.g. SafeBoda in Kampala, ZemExpress in Cotonou) (Tinka & Behrens, 2019). As in other parts of the world, technology disruptions have already occurred in for-hire services, with ride-hailing services gaining market share at the expense of metered taxis in many cities. It remains to be seen 245

Behrens, Chalermpong and Oviedo

whether these digital platforms will find as fertile ground amongst informal paratransit operators. It also remains to be seen whether these digital platforms might facilitate the introduction of ‘Mobility-as-a-Service’ systems that include informal operators and enable better ‘first-last mile’ paratransit service integration with mass trunk public transport (see also Chapter 22). In LAC cities, a fertile ecosystem for start-ups has given rise to either homegrown initiatives or adaptations of platforms from elsewhere to local conditions. Local and global entrepreneurs have introduced new forms of unregulated services such as microtransit and ride-hailing. Latin America is the fastest-growing and most profitable region for the ride-hailing company Uber, with the highest number of simultaneous trips, more than 25 million monthly active riders across 15 countries (Moed, 2018). Brazil is Uber’s second-largest market in the world, with 500,000 drivers (Darlington & Londoño, 2017). Uber has developed new products for specific market needs in the region, such as Uber Lite (a less data-demanding version of the Uber app) for areas where internet connection is more costly or slower. The company also launched UberMoto for motorcycle-taxis in the Dominican Republic. Aside from Mexico and Chile, microtransit services have not gained significant market share in LAC. In Mexico City, there are a few companies, the most popular of which are Urbvan and Jetty. Urbvan started as a pilot in 2016 with only 5 vehicles, growing to 230 in 2020. Jetty was founded in 2016 and has recently expanded operations to Puebla. In Santiago, Urbvan started operating in 2018. The operation of these market entrants has been hampered by a lack of public sector regulation and violent acts on drivers by incumbent operators (Flores Dewey, 2019). Digital platforms impact for-hire paratransit organisation and operation. Operators no longer need to gather at designated hubs to wait for passengers but connect directly via the app. Because of the apps, motorcycle-taxi drivers can be more productive and earn more income and can supplement income during the off peak by providing food and parcel delivery services (Phun et al., 2018). Relying on ride-hailing apps for access to passengers and other businesses, paratransit operators pay subscriptions to ride-hailing service providers rather than protection fees to the route- or area-based association, thereby removing the need to join the association. Some ride-hailing service providers in Asian cities also provide welfare and fringe benefits to drivers, such as liability insurance and loans for purchasing new vehicles. Regulatory authorities have typically been slow to catch up with these technology disruptions. Several countries have enacted regulations for ride-hailing operators or are in the process of doing so, but most only target ride-hailing sedan cars competing with conventional metered taxis (Mutiarin et al., 2019; Li et al., 2018). In Asia, only Indonesia has introduced regulations for ride-hailing motorcycle-taxis, but effectiveness remains to be evaluated (Ford & Honan, 2017). In LAC, apart from Mexico and Brazil, ride-hailing has yet to be regulated on a national scale.

Conclusion This chapter set out to synthesise current knowledge of informal paratransit services in cities of the Global South and to discuss prevailing policy issues and emerging trends. While illustrating that the sector is highly heterogeneous, a review of business, regulatory, and operating practices showed that paratransit services are usually operated by small businesses, organised into associations that exert varying degrees of self-regulation. Service operations are seldom free of state regulation, but the extent of the regulation and enforcement can vary considerably. Across the three regions analysed, operating environments often have considerable infrastructure deficits. Services are, in many cases, a response to gaps left by formal public transport undertakings. Driver employment conditions can be exploitative, and it is not uncommon for drivers to bear the revenue risk. Prevailing business models, however, make operators demand responsive, often providing the only service available to socially vulnerable groups. 246

Informal paratransit in the Global South

Common policy challenges include integration with other public transport modes and service quality and safety improvement. These policy challenges are compounded by poorly resourced regulatory authorities, often subjected to pervasive corruption. An emerging trend across the Global South takes the form of potentially disruptive technologies appearing in the informal paratransit sector, most commonly in the form of ride-hailing apps. These platforms may have a significant impact on operating practices. Few cities have regulatory frameworks in anticipation of this change. Experience suggests that attempts to change the business models and operating practices of incumbent operators through the overhaul of regulatory and enforcement frameworks will often be met with resistance. Careful analysis of local dynamics and policy options is therefore required in this sector.

Addendum As a final note, this chapter was written before the global outbreak of the COVID-19 pandemic and the lockdown regulations that ensued. At the time of writing this addendum, there is widespread conjecture on the lasting impact that the period of imposed social distancing may have on behaviours across a variety of sectors. This is particularly true for the passenger transport sector, where the impacts of lockdown regulations have been extensive. The purpose of this addendum is to discuss the possible long-term impacts of the COVID-19 pandemic on informal paratransit. While lockdown regulations have varied across countries, a common feature has been either the suspension of public transport services or the imposition of vehicle occupancy limitations. As noted in this chapter, a feature of informal paratransit services is their high vehicle occupancy and productivity, driven by small profit margins and driver remuneration models directly linked to ridership. When suspensions or occupancy limitations are imposed on formal mass transit services, there are a variety of contractual or fare policy mechanisms in place to compensate operators for the resultant loss of earnings. When suspensions or occupancy limitations are imposed on informal paratransit services, the brunt of the earnings loss is felt by the vehicle drivers and owners. Hence, lockdown compliance in some countries of the Global South has been weak, because to comply means running at a loss. Moreover, in many low-income neighbourhoods where residents are unable to work remotely or to substitute already precarious livelihoods, people have continued commuting despite lockdown restrictions, with informal paratransit as their main lifeline to maintain income. The COVID-19 pandemic may have long-term impacts on technology disruption and state subsidisation. With regard to technology, the need to reduce contact between passengers and vehicle crews, as well as to manage occupancy through reserved seating, has aligned the objectives of some pilot technology projects, such as Jetty in Mexico City, closely to pandemic responses. This might accelerate their disruptive effect in the long term. With regard to subsidisation, there has been a longstanding call from informal paratransit owner associations for operating subsidies, usually motivated by an equity argument to level the playing field with other public transport modes. These calls have generally been resisted by regulatory authorities on the grounds that they cannot subsidise tax and labour non-compliant services, that subsidising informality would be counterproductive to policy objectives of industry formalisation, and that expanding public transport subsidisation is not fiscally feasible. However, the operating losses in a vulnerable economic sector resulting directly from lockdown regulations imposed by the state arguably represent a stronger case for some form of immediate state subsidisation or compensation. It is possible that regulatory authorities may be forced to soften their earlier reluctance to subsidising paratransit during lockdown, and this may lead to innovative forms of subsidy that endure beyond COVID-19. Some commentators have, for instance, suggested that 247

Behrens, Chalermpong and Oviedo

subsidies might take the form of subsidised fuel costs, to the benefit of drivers, or subsidised vehicle purchase repayments, to the relief of owners. An important policy challenge will be to ensure that any such state support benefits quality of service to the passenger rather than subsidising operating inefficiencies.

References Ardila, A. (2007). How public transportation’s past is haunting its future in Bogotá, Colombia. Transportation Research Record, 2038, 9–15. doi:10.3141/2038-02 Ardila, A., & Rodríguez, D. (2000). How chaos does not destroy ridership: Operations of an exclusive busway carrying more than 35,000 passengers per hour per direction. Transportation Research Record, 1726, 1–7. doi:10.3141/1726-01 Arnado, J. D. D., Gogo, H. B., & Abocejo, F. T. (2017). Public utility jeepneys (PUJs) in Cebu City, Central Philippines: Regulations and implications to traffic congestion. International Conference on Law, Business, Education and Corporate Social Responsibility (LBECSR-17), Manila. Behrens, R., McCormick, D.,  & Mfinanga, D. (2016). An introduction to paratransit in sub-Saharan African cities. In R. Behrens, D. McCormick,  & D. Mfinanga (Eds.), Paratransit in African cities: Operations, regulation and reform (pp.  1–41). Routledge. www.routledge.com/Paratransit-in-AfricanCities-Operations-Regulation-and-Reform-1st-Edition/Behrens-McCormick-Mfinanga/p/ book/9780415870337 Behrens, R., McCormick, D., Orero, R., & Ommeh, M. (2017a). Improving paratransit service: Lessons from inter-city matatu cooperatives in Kenya. Transport Policy, 53, 70–88. doi:10.1016/j.tranpol.2016.09.003 Behrens, R., Zuidgeest, M., Hawver, H.,  & Birungi, C. (2017b). Case study investigation of unscheduled feeder and scheduled trunk service relationships in Cape Town. Paper presentation. 36th Southern African Transport Conference. http://hdl.handle.net/2263/62736 Berrones-Sanz, L. D. (2018). The working conditions of motorcycle taxi drivers in Tláhuac, Mexico City. Journal of Transport and Health, 8, 73–80. doi:10.1016/j.jth.2017.04.008 Bishop, T., & Amos, P. (2015). Opportunities to improve road safety through ‘boda-boda’ associations in Tanzania. Final Report. AfCAP project TAN2015G. Africa Community Access Partnership (AfCAP). Bishop, T., Barber, C., Adu, J., Afukaar, F., Muhia, G., Mwaipopo, H., Ekirapa-Kiracho, E., Rettie, N., Krasnolucka-Hickman, A., Divall, D., & Porter, G. (2018). Enhancing understanding on safe motorcycle and three-wheeler use for rural transport, RAF2114A. Africa Community Access Partnership (AfCAP). Boutueil, V., & Quillerier, T. (2020). The rise of paratransit digital platforms in large African metropolises: Competitive strategies between international and local platforms. Paper presentation. Transportation Research Board Annual Meeting. Briter. (2018). Digitising mobility in Africa. Retrieved January  1, 2020, from https://briterbridges.com/ innovation-maps. Cerio, C. T. (2017). The motorcycle taxi phenomenon in the Philippines: Is the demise of jeepney possible? International Journal of Tourism and Hospitality Reviews, 4(1), 44–52. doi:10.18510/ijthr.2017.421 Cervero, R. (2000). Informal transport in the developing world. UN-HABITAT. Cervero, R., & Golub, A. (2007). Informal transport: A global perspective. Transport Policy, 14(6), 445– 457. doi:10.1016/j.tranpol.2007.04.011 Chalermpong, S., Ratanawaraha, A.,  & Sucharitkul, S. (2016). Market and institutional characteristics of passenger van services in Bangkok, Thailand. Transportation Research Record, 2581(1), 88–94. doi:10.3141/2581-11 Cities Development Initiatives for Asia. (2011). Informal public transportation network in three Indonesian cities. Retrieved December  4, 2019, from https://cdia.asia/publication/informal-public-transporta tion-networks-in-three-indonesian-cities/ Darlington, S., & Londoño, E. (2017). Brazil becomes Uber’s latest regulatory battleground. Retrieved November 11, 2019, from www.nytimes.com/2017/11/05/world/americas/uber-brazil-regulation.html David, L. M. (2019). A  day in the life of a jeepney driver: Why modernization cuts deep. ABS-CBN News. Retrieved December  11, 2019 https://news.abs-cbn.com/spotlight/multimedia/slideshow/ 09/30/19/a-day-in-the-life-of-a-jeepney-driver-why-modernization-cuts-deep Diaz Olvera, L., Guézéré, A., Plat, D., & Pochet, P. (2016). Earning a living, but at what price? Being a motorcycle taxi driver in a sub-Saharan African city. Journal of Transport Geography, 55, 165–174. doi:10.1016/j.jtrangeo.2015.11.010

248

Informal paratransit in the Global South Duduta, N., Adriazola, C., Hidalgo, D., Lindau, L. A., & Jaffe, R. (2015). Traffic safety in surface public transport systems: A synthesis of research. Public Transport, 7(2), 121–137. doi:10.1007/s12469-014-0087-y Dugard, J. (2001). From low-intensity war to mafia war: Taxi violence in South Africa, 1987–2000 (Vol. 4). Centre for the Study of Violence and Reconciliation. Du Preez, D., Zuidgeest, M., & Behrens, R. (2019). A quantitative clustering analysis of paratransit route typology and operating attributes in Cape Town. Journal of Transport Geography, 80. doi:10.1016/j. jtrangeo.2019.102493 Ehebrecht, D., Heinrichs, D., & Lenz, B. (2018). Motorcycle-taxis in sub-Saharan Africa: Current knowledge, implications for the debate on ‘informal’ transport and research needs. Journal of Transport Geography, 69, 242–256. doi:10.1016/j.jtrangeo.2018.05.006 Estache, A., & Gómez-Lobo, A. (2005). Limits to competition in urban bus services in developing countries. Transport Reviews, 25(2), 139–158. doi:10.1080/0144164042000289654 Evans, J., O’Brien, J., & Ch Ng, B. (2018). Towards a geography of informal transport: Mobility, infrastructure and urban sustainability from the back of a motorbike. Transactions of the Institute of British Geographers, 43, 674–688. doi:10.1111/tran.12239 Flores Dewey, O. (2019). App-based collective transport service in Mexico City: A start-up case study. International Transport Forum Discussion Papers (No. 2019/01). OECD Publishing. Ford, M., & Honan, V. (2017). The Go-jek effect. ISEAS-Yusof Ishak Institute. Gamble, J., & Puga, E. (2017). Is informal transit land-oriented? Investigating the links between informal transit and land-use planning in Quito, Ecuador. Lincoln Institute of Land Policy (Working Paper WP19JG1). Goldwyn, E., & Vergel-Tovar, E. (2018). Mapping the unmapped transit network of Bogotá, Colombia. Paper presentation. Transportation Research Board 97th Annual Meeting. https://trid.trb.org/view/1495944 Goletz, M., & Ehebrecht, D. (2019). How can GPS/GNSS tracking data be used to improve our understanding of informal transport? A discussion based on a feasibility study from Dar es Salaam. Journal of Transport Geography. doi:10.1016/j.jtrangeo.2018.08.015 Gwilliam, K. (2011). Africa’s transport infrastructure: Mainstreaming maintenance and management. World Bank. Heinrichs, D., & Bernet, J. S. (2014). Public transport and accessibility in informal settlements: Aerial cable cars in Medellín, Colombia. Transportation Research Procedia, 4, 55–67. doi:10.1016/j.trpro.2014.11.005 Heinrichs, D., Goletz, M.,  & Lenz, B. (2017). Negotiating territory: Strategies of informal transport operators to access public space in urban Africa and Latin America. Transportation Research Procedia, 25, 4507–4517. doi:10.1016/j.trpro.2017.05.346 Hidalgo, D.,  & Huizenga, C. (2013). Implementation of sustainable urban transport in Latin America. Research in Transportation Economics, 40(1), 66–77. doi:10.1016/j.retrec.2012.06.034 Hidalgo, D., & King, R. (2014). Public transport integration in Bogota and Cali, Colombia: Facing transition from semi-deregulated services to full regulation citywide. Research in Transportation Economics, 48, 166–175. doi:10.1016/j.retrec.2014.09.039 Hongtong, T. (2019). Old van replacements no longer required. Retrieved December 5, 2019, from www. bangkokpost.com/thailand/general/1729679/old-van-replacements-no-longer-required Institute for Transportation and Development Policy. (2019). Public transport reform guideline for Indonesian cities. Retrieved December  5, 2019, from www.itdp.org/publication/ public-transport-reform-guideline-indonesian-cities/ Jauregui-Fung, F., Kenworthy, J., Almaaroufi, S., Pulido-Castro, N., Pereira, S.,  & Golda-Pongratz, K. (2019). Anatomy of an informal transit city: Mobility analysis of the metropolitan area of Lima. Urban Science, 3(3), 67. doi:10.3390/urbansci3030067 Kawaguchi, H., Kuromizu, K., & Yagi, S. (2013). A minibus supply control measure in Indonesia. 13th WCTR, Rio de Janeiro. Kemajou, A., Jaligot, R., Bosch, M., & Chenal, J. (2019). Assessing motorcycle taxi activity in Cameroon using GPS devices. Journal of Transport Geography, 79, 102472. doi:10.1016/j.jtrangeo.2019.102472 Kerzhner, T., & Martens, K. (2018). The role of labour informality in paratransit services: Case study of Lubumbashi, Democratic Republic of the Congo. Paper presentation. Transportation Research Board 97th Annual Meeting. https://trid.trb.org/view/1495748 Kisaalita, W., & Sentongo-Kibalama, J. (2007). Delivery of urban transport in developing countries: The case for the motorcycle taxi service (boda-boda) operators of Kampala. Development Southern Africa, 24(2), 345–357. doi:10.1080/03768350701327319 Klopp, J., & Cavoli, C. (2017). Mapping minibuses in Maputo and Nairobi: Engaging paratransit in transportation planning in African cities. Transport Reviews, 39(5), 657–676. doi:10.1080/01441647.2019. 1598513

249

Behrens, Chalermpong and Oviedo Klopp, J., & Mitullah, W. (2016). Politics, policy and paratransit: A view from Nairobi. In R. Behrens, D. McCormick, & D. Mfinanga (Eds.), Paratransit in African cities: Operations, regulation and reform (pp. 79–99), Routledge. www.routledge.com/Paratransit-in-African-Cities-Operations-Regulation-and-Reform-1st-Edition/ Behrens-McCormick-Mfinanga/p/book/9780415870337 Kumar, A., & Diou, C. (2010). Bus renewal scheme in Dakar: Before and after (No. 66939, pp. 1–95). World Bank. Kumar, M., Singh, S., Ghate, A. T., Pal, S., & Wilson, S. A. (2016). Informal public transport modes in India: A case study of five city regions. IATSS Research, 39(2), 102–109. doi:10.1016/j.iatssr.2016.01.001 Land Transportation Franchising and Regulatory Board. (2017). PUV modernization. Retrieved December 5, 2019, from http://ltfrb.gov.ph/index.php/puv-modernization-2/ Li, Y., Taeihagh, A., & De Jong, M. (2018). The governance of risks in ridesharing: A revelatory case from Singapore. Energies, 11(5), 1277. doi:10.3390/en11051277 Mahitthirook, A.,  & Nanuam, W. (2017). Microbuses to replace passenger vans this year. Bangkok Post. Retrieved December  5, 2019, from www.bangkokpost.com/thailand/general/1175508/ microbuses-to-replace-passenger-vans McCormick,  D., Mitullah,  W., Chitere,  P., Orero,  R.,  & Ommeh,  M. (2013). Paratransit business strategies: A  bird’s-eye view of matatus in Nairobi. Journal of Public Transportation, 16(2), 135–152. doi:10.5038/2375-0901.16.2.7 McCormick,  D., Schalekamp,  H.,  & Mfinanga,  D. (2016). The nature of paratransit operations. In R. Behrens, D. McCormick,  & D. Mfinanga (Eds.), Paratransit in African cities: Operations, regulation and reform (pp.  59–78). Routledge. www.routledge.com/Paratransit-in-African-Cities-OperationsRegulation-and-Reform-1st-Edition/Behrens-McCormick-Mfinanga/p/book/9780415870337 Moed, J. (2018). Uber’s wild ride to make Latin America its fastest growing region. Retrieved February  24, 2020, from www.forbes.com/sites/jonathanmoed/2018/12/20/is-uber-operating-illegally-in-its-fastestgrowing-region/#eb9f77d19251 Mutiarin, D., Nurmandi, A., Jovita, H., Fajar, M., & Lien, Y. N. (2019). How do government regulations and policies respond to the growing online-enabled transportation service (OETS) in Indonesia, the Philippines, and Taiwan? Digital Policy, Regulation and Governance, 21(4), 419–437. doi:10.1108/ DPRG-01-2019-0001 Mutiso, W., & Behrens, R. (2011). Boda boda bicycle taxis and their role in urban transport systems: Case studies of Nakuru and Kisumu, Kenya. 30th Southern African Transport Conference: Africa on the Move, Pretoria. http://hdl.handle.net/2263/17308 Oviedo, D., & Dávila, J. D. (2016). Transport, urban development and the peripheral poor in Colombia: Placing splintering urbanism in the context of transport networks. Journal of Transport Geography, 51, 180–192. doi:10.1016/j.jtrangeo.2016.01.003 Oviedo, D., & Titheridge, H. (2016). Mobilities of the periphery: Informality, access and social exclusion in the urban fringe in Colombia. Journal of Transport Geography, 55, 152–164. doi:10.1016/j. jtrangeo.2015.12.004 Parra, F. R., Garcias, C. M., Rogers, R., Asher, F., Castells, M., Lerner, J., & Koolhaas, R. (2005). El transporte público colectivo en Curitiba y Bogotá. Revista de Ingeniería. Universidad de Los Andes. Phun, V. K., Masui, R., & Yai, T. (2015). The future prospects of paratransit service in Phnom Penh. Proceedings of the 7th Regional Symposium on Infrastructure Development, 20(7), 221–226. Phun, V. K., Masui, R., & Yai, T. (2018). Operational characteristics of paratransit services with ridehailing apps in Asian developing cities: The Phnom Penh case. Journal of Transportation Technologies, 8(4), 291. doi:10.4236/jtts.2018.84016 Rañosa, L. L., Fillone, A. M., & De Guzman, M. P. (2017). Jeepney service operation and demand in Baguio City, Philippines. Proceedings of the Eastern Asia Society for Transportation Studies, 11. Rasmussen, J. (2012). Inside the system, outside the law: Operating the matatu sector in Nairobi, Urban Forum, 23, 415–432. doi:10.1007/s12132-012-9171-z Ratanawaraha, A., & Chalermpong, S. (2015). Monopoly rents in motorcycle taxi services in Bangkok. Thailand. Transportation Research Record, 2512(1), 66–72. doi:10.3141/2512-08 Ratanawaraha, A., & Chalermpong, S. (2018). How operators’ legal status affects safety of intercity buses in Thailand. Transportation Research Record, 2672(31), 99–109. doi:10.1177/0361198118758286 Ratanawaraha, A., & Chalermpong, S. (2019). Operational models, drivers’ compensation, and bus service quality in Bangkok. Paper presentation. 98th Annual Meeting of the Transportation Research Board. Regidor, J. R. F., Vergel, K. N., & Napalang, M. S. G. (2009). Environment friendly paratransit: Reengineering the jeepney. Proceedings of the Eastern Asia Society for Transportation Studies, 7, 272–272. doi:10.11175/eastpro.2009.0.272.0

250

Informal paratransit in the Global South Rodriguez, C., Peralta-Quirós, T., Guzman, L. A.,  & Reyes, S. A. C. (2017). Accessibility, affordability, and addressing informal services in bus reform. Transportation Research Record, 2634, 35–42. doi:10.3141/2634-06 Rodríguez Baquero, L. E., & Nuñez Cetina, S. (2003). Empresas públicas de transporte en Bogotá: Siglo XX. Alcaldia de Bogotá. Saddier, S., & Johnson, A. (2018). Understanding the operational characteristics of paratransit services in Accra, Ghana: A  case study. Paper presentation. 37th Southern African Transport Conference. http://hdl. handle.net/2263/69574 Saddier, S., McLachlan, N., & Dass, D. (2019). Measuring the evolution of passenger satisfaction following the introduction of scheduled services: The case of the 7th Avenue Minibus-Taxi Association in Mitchells Plain. Paper presentation. 38th Southern African Transport Conference. www.satc.org.za/assets/3a_06-saddiermclachlan-dass.pdf Saddier,  S., Patterson,  Z., Johnson,  A.,  & Chan,  M. (2016). Mapping the jitney network with smartphones in Accra, Ghana: The AccraMobile Experiment. Transportation Research Record, 2581, 113–122. doi:10.3141/2581-14 Schalekamp, H., & Klopp, J. (2018). Beyond BRT: Innovation in minibus-taxi reform in South African cities. Paper presentation. 37th Southern African Transport Conference. http://hdl.handle.net/2263/69575 Schalekamp, H., & Saddier, S. (2019). Emerging business models and service options in the shared transport sector in African cities. Volvo Research and Educational Foundation. Sengers, F., & Raven, R. (2014). Metering motorbike mobility: Informal transport in transition? Technology Analysis and Strategic Management, 26(4), 453–468. doi:10.1080/09537325.2013.870991 Silalahi, S. L. B., Handayani, P. W.,  & Munajat, Q. (2017). Service quality analysis for online transportation services: Case study of GO-JEK. Procedia Computer Science, 124, 487–495. doi:10.1016/j. procs.2017.12.181 Suárez, M., Murata, M.,  & Delgado Campos, J. (2016). Why do the poor travel less? Urban structure, commuting and economic informality in Mexico City. Urban Studies, 53(12), 2548–2566. doi:10.1177/0042098015596925 Tinka, A., & Behrens, R. (2019). Cashless fare collection in sub-Saharan African paratransit: A review of experiences. Paper presentation. 38th Southern African Transport Conference. http://hdl.handle.net/2263/74251 Venter, C., Mahendra, A., & Hidalgo, D. (2019). From mobility to access for all: Expanding urban transportation choices in the Global South (Working Paper). World Resources Institute. Warren, J. P., & Ortegon-Sanchez, A. (2016). Designing and modelling Havana’s future bus rapid transit. Proceedings of the Institution of Civil Engineers – Urban Design and Planning, 169(2), 104–119. doi:10.1680/ jurdp.15.00015 Wicaksono, A., Lim, I., Muromachi, Y., Vergel, K. N., Choocharukul, K., Tan, V. H., & Yai, T. (2015). Road-based urban public transport and paratransit in six Asian countries: Legal conditions and intermodal issues. Journal of the Eastern Asia Society for Transportation Studies, 11, 227–242. doi:10.11175/ easts.11.227 Willoughby, C. (2013). How much can public private partnership really do for urban transport in developing countries? Research in Transportation Economics, 40(1), 34–55. doi:10.1016/j.retrec.2012.06.038 Wu, I., & Pojani, D. (2016). Obstacles to the creation of successful bus rapid transit systems: The case of Bangkok. Research in Transportation Economics, 60, 44–53. doi:10.1016/j.retrec.2016.05.001 Yañez-Pagans, P., Martinez, D., Mitnik, O. A., Scholl, L., & Vazquez, A. (2018). Urban transport systems in Latin America and the Caribbean: Challenges and lessons learned. Technical Note N° IDB-TN-01518, Inter-American Development Bank.

251

19 TAXIS Matthew W. Daus

History and evolution of taxi industry regulation This chapter reviews the history and evolution of the taxi industry – its current status, future challenges, and opportunities – with an emphasis on the United States as a case study, along with a sampling of comparisons from global jurisdictions, including England, Argentina, and Singapore. The regulatory framework of taxi industries is diverse in these jurisdictions and reflects the particular history to develop an approach to allow for taxi service to meet local needs; as the cultural dynamic and evolution of urbanized metropolitan areas differ from country to country, so do the regulations. In Argentina, the birth of one of the earliest forms of shared-ride taxi services, the colectivo, evolved as taxi drivers were allowed the freedom to operate. A comparison of industries with stricter regulations, such as England and the United States, and industries with less onerous regulations, such as Argentina and Singapore, illuminates the point that there is no single approach to creating regulations for transportation needs; each jurisdiction must identify and address its own respective challenges that arise over time. The word “taxi” is generally associated with a classic New York City yellow Checker cab. Hollywood blockbusters often paint the picture of pedestrians lining the streets of the city with their hands raised, seeking a ride. The global industry, however, is much more complex. Regulations change annually throughout the world as new innovations and issues present themselves, creating a dynamic paradigm that constantly seeks to predict and address the problems of tomorrow. While New York City may be the global hotspot for taxi services, it represents a small portion of a large market. Dating back to 1605, the first for-hire “vehicles” were horse-drawn carriages operated in Paris and London (Gilbey, 1903, p. 29). The design was simple: a large, four-wheeled carriage, pulled by two horses that could transport passengers from point A to point B. In 1834, Joseph Hansom, an English architect, invented the Hansom cab (Lay, 2018). The small, easily maneuverable one-horse carriage offered quicker, cheaper services to the public and became the most popular model of for-hire vehicle in Paris, Berlin, St. Petersburg, and New York. Thus began the evolution of the taxi. After the Hansom cab, taxis evolved at a much faster rate. The first gasoline-powered taximeter cabs were introduced by Gottlieb Daimler, a German engineer, in 1897. At the same time, Samuel’s Electric Carriage and Wagon Company (E.C.W.C.) introduced its first fleet of 12 electric hansom cabs in July 1897 (Farrell, 2018). The E.C.W.C. added 252

Taxis

50 more vehicles before 1898, when the company was reconstituted by its financial backers to form the Electric Vehicles Company (ECV). The ECV built the Electrobat electric car, placing 100 cabs on the road by 1899. Two notable milestones in automotive history took place that year: the first speeding ticket was issued to Jacob German, driving an electric taxicab, in May, and the first automotive crash in New York occurred in September, when Henry Bliss was struck by an electric taxicab. The ECV operated until 1907, when a fire destroyed 300 of their vehicles, which, aided by the Panic of 1907, ultimately led to the company’s demise (Farrell, 2018). Later in 1907, Harry N. Allen, a New York business leader, imported 65 gasolinepowered cabs from France and started the age of combustion engine taxis that still forms the basis of the sector today. Every jurisdiction in the world has a form of taxi service with a regulatory framework that mirrors the history of the commercial and legal needs of for-hire service. In London, for instance, regulation of the taxi industry dates back to 1636 under Charles I, who issued a proclamation restricting the number of hackney coaches to 50 due to congestion concerns. Soon after, in 1710, the licensing system that exists in London today was created. Hackney carriage licensing was largely administered and enforced by the Metropolitan Police, as it passed the famous Metropolitan Public Carriage Act of 1869, a large portion of which is still relevant today. As technology advanced, however, so did the need for specific legislative action in larger metropolitan areas to address new issues. The most recent organization to regulate the taxi service, Transport for London (TfL), drafts legislation for the entire for-hire vehicle industry and gathers trip data to continually improve on current regulations. TfL is currently responsible for all day-to-day operations of London’s entire public transport network, including private for-hire vehicles, taxicabs, and buses. As seen in the United States and many other countries around the globe, TfL sets regulated taxi fares to prevent overcharging and maintain equilibrium in the market. Comparatively, the taxi industry in other markets such as Singapore and Argentina has fewer regulations in place when compared to the United States or London. Taxis were a more recent addition to the Singaporean economy; first introduced in 1910 by C.F. Wearne and Co., taximeters were imported from the United Kingdom and equipped to Rover cars (The Straits Times, 1910). In the 1960s, pirate taxis emerged as an alternative to Singapore’s lack of adequate public transport. As a result, Singaporean officials formed a committee to review and implement regulations to protect and promote the practice of licensed taxi work (The Straits Times, 1966). This trend continued until 1998, when the Land Transport Authority (LTA), a statutory board of the Ministry of Transport in Singapore, adopted a more laissez-faire approach by deregulating taxi fares entirely, which forced the market to correct itself and promoted more competition between companies. Contrary to most Western taxi markets, service providers in Singapore set their own fares, allowing them to adapt to a dynamic marketplace. In Argentina, similar deregulation has occurred; taxicabs currently have minimal regulations in place to promote competition between competitors. In 1928, a group of taxi drivers launched a new mode of for-hire transportation that involved charging fixed rates and established routes for four passengers at a time (Singh, 2018). This practice is known as the colectivo and was the first form of shared-ride transportation service offered in the country. By 1930, Buenos Aires had largely been urbanized; railway lines facilitated public transport to settlements past the city limits, creating a large metropolitan area with a population of over 3 million people (Singh, 2018). The colectivo quickly evolved into a popular form of transportation; eventually, the colectivo transformed into a bus service adopted as the main source of public transport in Buenos Aires (Singh, 2018). Currently, the Buenos Aires Ministry of Transportation regulates all ground transportation modes, including taxis, buses, and trains. To enforce these policies, 253

Matthew W. Daus

agreements have been made with national security forces, including the Argentine National Gendarmerie, Naval Prefecture, and Argentine Federal Police. The Ministry of Transportation has some regulations in place to prevent overcharging and other common issues within the taxi industry, but the restrictions for newer transportation network companies, for instance, are almost non-existent. Currently, a battle between the Buenos Aires Ministry of Transportation and Uber is underway, as the ride-hail giant attempts to expand its services in Argentina. Whether government officials choose to implement strict regulations or deregulate the industry completely, the ultimate goal of taxi regulation is to maintain an equilibrium of supply and demand in the market. While the regulatory framework of each country may differ, the conditions for market equilibrium remain constant for everyone – regulators seek to find the sufficient number of taxis for their respective jurisdictions, as the economic health of the industry is dependent on adequate cash flows for taxi drivers. Stable and sufficient ridership is critical to provide adequate wages for drivers. History has shown that an oversupply of taxis is often accompanied by fare-cutting price wars, low driver wages, traffic congestion, driver strikes, and illegal activities, including riots. During the 1920s and 1930s, financial hardships in the United States, leading to the Great Depression, led to a sharp increase in the number of taxi drivers, and total cabs in New York peaked at 21,000 in 1931. Cheating, hustling, false advertising, stealing, and extortion increased as tensions rose in the taxi industry, which ultimately drove New York City officials to place the taxi industry under police control in 1925 (Jackson, 2010). This temporary solution helped address this rogue activity, but ultimately, the underlying oversupply of taxis was not addressed until 1937 by legislation sponsored by New York City Alderman Lew Haas. The concept of a taxi medallion was first implemented in New York City in 1937 as a means of preventing oversaturation of the taxi market and to keep traffic congestion at a sustainable level. Sponsored by Alderman Haas, the “Haas Act” capped the number of taxis at 13,595, creating the medallion system (Rodriguez & Levin, 2020). Several jurisdictions in the United States currently use taxi medallions as a method of limiting the number of authorized taxis in a city, including New York City, Boston, Miami, Chicago, San Francisco, and Philadelphia. In New York City, taxi medallions are fully transferable if approved by the regulating agency, the New York City Taxi and Limousine Commission (the TLC) (New York City Taxi & Limousine (TLC) Rule §58–43), creating a secondary market for trade in medallions. The Haas Act of 1937 divided taxicabs into two categories of ownership: independent medallions and corporate medallions (or “mini-fleet medallions”). The corporate medallions were owned in lots of at least two medallions, and there were no restrictions on the number of corporate medallions a person or entity could own (TLC Rule §51–03). Independent medallions, which constituted approximately 42% of the NYC taxi fleet (as required by the Haas Act), were single medallions for which a person could own one, or part of one, independent medallion. Until 2016, the independent medallions also had a regulatory requirement that the owner must also drive the independent medallion a minimum amount of 900 hours each year [TLC Rule §51–03(p)]. In 2017, the New York City Council repealed Section 19–504(i) of the New York City Administrative Code, which was the provision that had maintained the 58% to 42% ratio of corporate to independent medallions that was established by the Haas Act (New York City, N.Y., Local Law No. 59 Int. No. 1475-A (2017) [effective Mar. 21, 2017]). Although the classification of independent and corporate medallions still exists, with the removal of the driving requirements and limits on ownership, there are no legal or operational differences between corporate and independent medallions. Currently, in New York City, the TLC has regulatory oversight of medallion taxicabs, as well as livery services, black cars, luxury limousines, and other for-hire vehicles (FHVs) operating 254

Taxis

in the city. The TLC was created in 1971 to regulate and improve the quality of taxi and livery services, to establish transportation policies to govern these services, and to create and enforce professional and uniform standards to ensure public safety. As of December  2019, the TLC licenses and regulates 13,587 medallion taxicabs, 199,481 licensed drivers, and 116,449 for-hire vehicles. The TLC also regulates taximeter shops, medallion taxicab brokers, and taxicab agents (NYC Taxi & Limousine Commission, 2019). In addition to the TLC, New York State has legislative and regulatory power over the taxi and FHV industry. In 2004, the TLC introduced several technological improvements in yellow cabs, known as the Taxicab Passenger Enhancement Program (TPEP). It is composed of four main components: a driver information monitor, a passenger information monitor, a credit/debit card payment system, and a global positioning system (GPS). The GPS, importantly, ushered the yellow cabs into the era of big data. Some of the trip metrics captured include pick-up and drop-off timestamp, pick-up and drop-off location, trip duration, and itemized fares. The cost of implementing the TPEP system was initially offset by a 26% fare increase, the largest fare increase in the history of yellow cabs (NYC Taxi & Limousine Commission, 2004). From approximately 1970 until 2015, medallion prices generally showed a consistent and strong upward trend, fueled by a growing demand for yellow cab service and a limited number of available medallions, as well as a relatively stable regulatory environment in New York City and other similar positive macroeconomic factors. The availability of longer loan terms and the introduction of medallion leasing have attracted individuals to enter the medallion system, either by owning a single medallion or leasing medallions to drivers or by forming corporate entities that hold multiple medallions and function as fleet management taxi companies. As shown in Figure 19.1, yellow cabs mostly operate within the Manhattan Core (south of 96th Street on the east side of Manhattan and 110th Street on the west side of Manhattan), with the remaining trips distributed at the two airports within New York City: John F. Kennedy International Airport and LaGuardia Airport. When taxi medallions were first traded on the open market in 1947, the average price of a medallion in New York City was $2,500 (NYC Taxi & Limousine Commission, 2014). Four decades later, the average price had risen to $100,000 (Carmody, 1985). This was followed by an even more dramatic rise in subsequent decades. After receiving New York State approval to issue new medallions, the City held its first three medallion auctions between 1996 and 1997. At the auction in September 1997, independent medallions sold for a high of $233,210, while corporate medallions sold for a high of $285,555 each (Buettner, 1997). Revenue from these auctions is received by the City. At City-held auctions in April and October 2004, independent medallions reached a high of $360,000, and corporate medallions reached a high of $407,551 per medallion. At auctions in June 2006, independent medallions sold at a high of $425,102, while corporate medallions reached $554,148 per medallion. At the May 2008 auction, independent medallions reached $524,000, while corporate medallions sold for $656,000 per medallion. There were no auctions between 2009 and 2012, but the value of medallions continued to rise in the secondary market, with the average price of an independent medallion and corporate medallion each reaching $936,117 and $1,160,500, respectively, in 2013. But that was not the end of the medallion auctions. After New York State authorized the sale of 2,000 additional wheelchair-accessible vehicle taxi medallions in 2011, the City held three more auctions between November 2013 and March 2014, generating more than $400 million for the City. It was during these last two auctions that sale prices peaked. At the February 2014 auction, the average winning bid for independent accessible medallions was $863,742. At the subsequent auction in March 2014, the average winning bid for corporate wheelchair-accessible medallions was $1,164,379 per 255

Matthew W. Daus

Figure 19.1 Trip density by pick-ups, December 2019 Source: Based on data from NYC Taxi & Limousine Commission, 2019

medallion. The March 2014 auction was the most recent TLC auction before the onset of the current taxi medallion crisis (NYC Taxi & Limousine Commission, n.d.a).

Disruptions on the existing taxi industry Since the outbreak of the COVID-19 pandemic, the taxi industry in the United States has been affected severely, much like the rest of the world. In New York City, ridership plunged by 94% from the first week of March 2020 to the week of May 4. Due in part to the fear of contracting the virus (at least 50 drivers have died from the virus), roughly 83% of drivers have stopped working altogether (Chan, 2020). The situation is no different in other cities. In Chicago, the number of taxi rides in March  2020 dropped by more than 63% compared to March  2019 (McCall, 2020). In Las Vegas, taxi ridership fell by 97% from February  2020 to April  2020 (Akers, 2020). In San Diego, the number of taxi drivers who quit rose faster than the average 256

Taxis

unemployment rate (Lewis, 2020). Needless to say, adequate cleaning methods, personal protective equipment distribution, and virus transmission prevention mechanisms (like partitions) have become priority items for taxi fleet management. To this end, the U.S. Department of Labor’s Occupational Safety and Health Administration (OSHA) has issued an alert listing safety tips to help reduce the risk of exposure to the coronavirus in the rideshare, taxi, and car service industry (OSHA, 2020). In addition to these containment actions, municipal taxi regulators have also stepped up to assist drivers. In New York City, the TLC has created a meal delivery program for $53 per route to vulnerable populations. In Washington D.C., the Department of For-Hire Vehicles (DFHV) has repurposed its microtransit program to provide trips to hospital workers in partnerships with taxi companies and Via. In Chicago, the Business Affairs and Consumer Protection (BACP) has eliminated the passenger portion of all paratransit taxicab fares as a way to encourage ridership (Daus, 2020b). While the long-term impact of the pandemic on the taxi industry remains to be seen, it was already experiencing major disruptions from transportation network companies (TNCs). A transportation network company is a company that provides pre-arranged, on-demand passenger transportation services for compensation by connecting riders to drivers through an online digital application or platform, such as a smartphone app. In March 2009, Uber Technologies, Inc. (Uber) – the first TNC – was founded as UberCab in San Francisco, California. UberCab officially began operations in San Francisco in June 2010. In May 2013, Lyft, Inc. (Lyft), previously operating as Zimride, launched itself as a ride-hailing company in several U.S. cities. Similar app-based services have emerged across the globe, including Gett and Via, Didi in China, Ola in India, Careem in Dubai, and Grab in Singapore. These services have competed with the local taxi industries head-on, affecting the incumbent sector adversely. Due to the growth of Uber and Lyft, yellow cab ridership in New York City declined by 51% between 2013 and 2019. Over the span of four years, the yellow cab sector saw its market share decline from roughly two-thirds in 2015 to less than one-quarter of the market share in 2019 (NYC Taxi & Limousine Commission, n.d.b), as shown in Figures 19.2 and 19.3. In Chicago, only 60% of its 7,000 taxicabs were in operation as of 2019. In Philadelphia, between fall 2014 and summer 2016, taxicab ridership declined by 40%. Boston saw taxi ridership plunge by 22% in the first half of 2015 (Graham, 2018). In New York City, the price of corporate medallions peaked in 2013 and started to drop in 2014. By 2018, the situation was so difficult that several taxi medallion owner-drivers, facing increased financial burdens, committed suicide. Many more taxi owners and drivers were in financial hardship. Lenders that provided financing to medallion owners were also affected due to the growing number of distressed loans in their portfolios. By the end of 2019, an average medallion was worth roughly 20% of its peak value (NYC Taxi  & Limousine Commission, n.d.c). The precipitous fall in medallion value was not confined to New York City. In other cities with similar medallion systems, the fall in price between a peak year and 2019 was even more severe, as shown in Figures 19.4–19.8. While the taxi medallion system is used principally in the United States, the concept of limiting the number of taxi licenses is common around the world. Many cities have similar regulatory systems in place to control the number of taxis that can legally operate in their jurisdiction. Other jurisdictions, however, choose to deregulate the market in a hope that an equilibrium can be achieved through competition. One of the major topics that regulators across the globe continue to discuss is the integration of TNCs into the existing taxi framework. TNCs are now allowed to operate and are regulated, in some form, throughout the United States. The same is not necessarily true for cities outside the United States and countries around the globe, which have mixed views on TNCs and how to regulate them. 257

Matthew W. Daus

Figure 19.2 NYC ridership comparison, 2015–2019 Source: Based on data from NYC Taxi & Limousine Commission, 2019

Figure 19.3 NYC taxi market share by trip volume, 2015–2019 Source: Based on data from NYC Taxi & Limousine Commission, 2019

258

Taxis

Figure 19.4 NYC medallion average price, 1947–2019 Source: Based on data from NYC Taxi & Limousine Commission, 2019

Figure 19.5 Chicago medallion price, 2007–2019 Source: Based on data from City of Chicago Department of Business Affairs and Consumer Protection, 2019

259

Matthew W. Daus

Figure 19.6 Philadelphia medallion price, 2006–2019 Source: Based on data from Philadelphia Parking Authority, 2019

Figure 19.7 Miami medallion price, 1997–2019 Source: Based on data from Miami Department of Transportation and Public Works, 2019

260

Taxis

Figure 19.8 Boston medallion price, 2000–2019 Source: Based on data from City of Boston, 2019

In Canada, provincial and municipal governments are taking a more cautious approach to TNC regulation. For example, TNCs were only allowed to operate in British Columbia in January 2020, making Vancouver the last major city in North America to approve TNCs [Passenger Transportation Act, S.B.C. 2004, c. 39, as amended by Passenger Transportation Amendment Act, 2018, S.B.C. 2018, c. 53 (Bill 55) (effective Sept. 3, 2019)]. In Toronto, TNC drivers will now face the same training and experience requirements as taxi and limousine drivers (Toronto Municipal Code, Ch. 546). In Calgary, the current city administration is recommending a full review of the livery transport bylaw, which governs both taxis and rideshares, with the help of the public to level the playing field (Walter, 2019). European countries tend to take a more cautious and stringent approach to TNC regulation, as regulators view new business models and innovations with caution until the full consequences are determined. This has made it difficult for TNCs to expand as quickly in Europe. In a landmark ruling in December 2017, the European Court of Justice (ECJ) ruled Uber is a transportation provider – and not a technology platform – subject to the same rules as taxis (Court of Justice of the European Union, 2017). In Barcelona, Uber and fellow ride-hailing company Cabify ceased their operations in January 2019 in response to the city tightening its regulations in response to protests from local taxi drivers (Jones, 2019). On November 25, 2019, Transport for London notified Uber that it would not renew the company’s license to operate in London (Robinson, 2019). Uber appealed the TfL’s decision to Westminster Magistrates’ Court on December 13, 2019, and Uber is able to continue to operate in London pending the outcome of the appeal, which could take months or years to resolve (Reuters, 2019). Uber 261

Matthew W. Daus

has been banned in Hungary and Bulgaria, and some services have been banned in cities across France, Italy, Finland, Germany, and the Netherlands (Shead, 2019). In Asia, TNCs have made inroads in China and Southeast Asia, but less so in the rest of the region. In China, Didi Chuxing is the most prominent player, while Grab from Singapore is a key player in the Southeast Asia region, with services spanning eight countries and 170 cities. In 2016, China became the first major economy to issue laws for TNCs on a national level, in addition to coming up with a set of guidelines for taxi reform. China’s TNC regulation focuses primarily on driver and vehicle standards, as well as efforts to restrain illegal activity (Noguellou & Renders, 2018). In Singapore, the authorities have proposed two classifications of operating licenses: one for street-hail services and another for ride-hail services (Tang, 2019). Elsewhere in Japan, South Korea, and Taiwan, taxi operators have been influential in dissuading regulators from allowing TNCs to compete head-on with taxis (Li, 2019). In Australia, TNCs fall under the “booking service providers” category of the Point to Point Transport (Taxis and Hire Vehicles) Act 2016 and can only operate legally if they have a business registration for their car and a hire car driver authorization.

The socioeconomic impacts of the transportation disruption movement Of concern with the rapid expansion of TNCs is the adverse impact the TNCs have had on underserved communities, the environment, people with disabilities, and the labor force of the on-demand sharing economy. If these issues are to be successfully addressed, jurisdictions must reconcile the regulatory frameworks among the various for-hire vehicle sectors, and that means providing a “level playing field”. In many jurisdictions in the United States, TNCs operate outside the traditional regulatory framework that applies to taxis and for-hire vehicles (livery, black car, and limousines). This includes vehicle license caps, licensing procedures and fees, commercial insurance costs, fingerprint background checks, and other requirements mandated by the taxi and FHV industry. The regulations for what are often interchangeable services tend to be uneven and skewed in favor of TNCs, to the detriment of the traditional taxi and for-hire vehicle industries. To add to the confusion, the shared mobility industry is plagued by discrepancies in use and definitions of specific terms (SAE International, 2018). In turn, this often creates ambiguity and confusion for policymakers, regulatory agencies, and the riding public.

Wheelchair accessibility and equity Business models common among app-based on-demand transportation services, such as TNCs, claim to provide a transportation alternative “for all”. When further examined, it becomes clear that these claims fall somewhat short. Throughout the United States, the proliferation of TNCs has slowed progress toward wheelchair-accessible shared mobility. TNC vehicles and drivers rarely have the capacity and ability to accommodate mobility devices such as wheelchairs and scooters. Typically, TNCs are not held to the same accessibility mandates or standards as the traditional for-hire vehicle industry, and TNCs even go as far as to argue that the Title III of the Americans with Disabilities Act (the “ADA”), 42 U.S.C. § 12181 et seq. does not apply to their operations because they are not a transport provider (see e.g., Access Living of Metropolitan Chicago et al. v. Uber Technologies Inc. et al., case number 19–2116, U.S. Court of Appeals for the Seventh Circuit). Under the ADA, taxicab companies are not required to purchase an accessible automobile. If a taxi company purchases a larger vehicle, like a van, it is subject to the same rules 262

Taxis

as any other private entity primarily engaged in the business of transporting people that operates a demand-responsive service [Transportation Services For Individuals With Disabilities (ADA), U.S. Code of Federal Regulations, title 49, § 37.29(b)]. Equity in transportation is both a human and civil rights issue. Access to affordable and reliable transportation enhances opportunities and is necessary in addressing unemployment, poverty, and other equal-opportunity goals, such as access to health care services and good schools. Current federal and state transportation spending programs have been discrepant between communities and populations and have not equally benefited all. Negative effects of these transportation decisions are broadly felt and have long-lasting effects. To ensure equity for the riding public, the same advantages and considerations must be given to all potential passengers. If equity is to be successfully addressed, jurisdictions must first address the uneven regulatory and financial resource advantages that smartphone application on-demand dispatchers have (Daus, 2016).

Traffic congestion and the environment The exponential growth of TNCs adds countless vehicles to the road and impacts cities’ efforts to address the number of personal motor vehicles on the road. Unregulated growth challenges years of transport planning and policy that sought to mitigate congestion and pollution and encourage shared mobility and mobility management. When additional vehicles are added to the road, harmful environmental impacts also proliferate with the spreading of volatile organic compounds, carbon monoxide, sulfur dioxide, fine particulate matter, greenhouse gases, and air toxins. Congestion has impacted local businesses and taxpayers. Additional vehicles on the road have also necessitated additional travel time and public funds spent on road repair, while business, government, and labor force activity have been negatively impacted by traffic jams and gridlock. As urban areas are projected to continue growing, policy makers need to consider how they will allow TNCs to continue to expand while avoiding a cataclysmic collision with environmental and sustainability policies.

Corporate social responsibility TNCs consistently market themselves as socially responsible businesses, yet there have been instances of tax avoidance, which deprives the jurisdictions within which they operate of potentially large sums of tax revenue. Uber, for example, created a web of global subsidiaries, limited partnerships, and holding companies and entered into separate and distinct agreements with these entities to safeguard itself from taxes in foreign jurisdictions and from domestic taxes on foreign income (Daus, 2016). Comparatively, local taxicab and for-hire vehicle trips are typically subject to state and local taxes, increasing the operators’ cost burdens and pressuring them to charge higher fares – if those fares are not set by regulation – than their TNC counterparts. This puts the traditional industry at a further competitive disadvantage.

The gig worker economy The definition of a “sharing economy”, as applied to TNCs, has led to a divergence in policies throughout jurisdictions on how these services should be regulated. TNCs have used this to their advantage and have expanded the number of vehicles and drivers in many cities, arguing that their service is different from the traditional for-hire transportation service. At the same time, this unregulated expansion has a major impact on the labor markets of cities that are being 263

Matthew W. Daus

affected by vehicles. TNC service is best described as an “access economy”, where companies facilitate access to FHV service through their app-based platforms. This has had a severe impact on driver income. TNCs use an independent contractor model to extract maximum profits from the service that drivers provide, which has led to driver unrest and litigation related to employment status. Taxi and TNC drivers are typically independent contractors, not employees. The terms “employee” and “independent contractor” are labels for tax and legal purposes. The main distinction between an independent contractor and an employee is the degree of control and independence that the hiring entity exerts over the worker. Independent contractors set their own hours, use their own tools, and have their own clients. Employees follow a schedule set by their employer and use their employer’s tools and methods to accomplish specific work at the direction of the employer, among other things. There are different aspects and tests for control in various U.S. states, but it all comes down to economics. Employers are responsible for paying for benefits, payroll taxes, unemployment insurance, and other costs for their employees, whereas independent contractors pay their taxes and benefits in full, receive no compensation for accidents or health-related incidents, and are ineligible for unemployment benefits.

The future of the taxi industry In the United States, the taxi industry market has shrunk significantly, and some may say taxis in certain high-density cities have been decimated by TNCs taking over those markets. In some cities, such as Los Angeles, taxicabs remain in business primarily due to airport and wheelchairaccessible transportation business – with street hails and pre-arranged general passenger transport having been taken over almost entirely by TNCs. Unless major business or regulatory changes happen, such as TNCs going bankrupt or states changing their laws, the future of what remains of the taxi industry in most cities will be limited to niche services. The non-emergency medical transportation (NEMT) field is ripe for disruption, with paratransit providers and brokers facing competition from new software technology companies, TNCs, and wheelchair-accessible taxicabs (Koffman, 2016) (see also Chapter 17). This niche area, which TNCs originally avoided, shows great opportunity for taxicab services operating in a pre-arranged and on-demand capacity – given government subsidies and the stability of such services. For many years, the taxi industry fought against wheelchair access, but now those companies that built up such services, and those that were required by local regulation to do so, are in an ideal position for growth. Also, many cities with large urban public transport systems may migrate over their public paratransit system to replace buses or mini-shuttles with point-to-point accessible taxicabs and TNCs to outsource all trips under one system or platform with subsidies using an app that could form the basis for a MaaS platform over time (see also Chapter 3). In terms of the environment, taxicab regulation historically sought to promote a supply and demand balance not just for congestion mitigation purposes but also to ensure a living wage for drivers. In New York City, the large growth of high-volume apps like Uber and Lyft has led the City to enact a cap on the number of for-hire vehicle permits, as well as implementing anticruising regulations to penalize companies for vehicles driving in the central business district of Manhattan without passengers (New York City, N.Y. Local Law 2018/147 [effective Aug. 14, 2018]). It was only a matter of time until transportation regulators and elected officials became aware of the exponential growth that has not only contributed to more congestion but also to wage diminution for drivers. But this wake-up call came about, in large part, as a result of the number of medallion owner-driver suicides that caused the New York City mayor, the New 264

Taxis

York City Council, and the TLC to act. Other regulators are closely watching the congestion pricing scheme and FHV cap implemented in New York City, and it is very possible that other cities and states will follow through on replicating such laws in the near future (Congestion Surcharge, N.Y. Tax Law, Article 29-C). Some type of reform or bailout is expected to compensate taxi medallion owner-drivers for the loss of equity and asset value for their medallions and to protect the taxi or overall for-hire vehicle supply number and wages. This is not unlike the reasons during the Great Depression the closed medallion system was created in the first place. Unfortunately, history has repeated itself. When taxicab unions were active during the mid-20th century up until the advent of leasing in the 1970s, which led to the decertification of unions for collective bargaining purposes, many local regulators protected driver earnings by transferring lease cap limitations and commissions set forth in labor agreements into regulations that prohibited taxi owners from charging more than certain amounts. While these lease caps or gate fees are still law, on the TNC side of the equation, states like California and New York are looking to eliminate the independent contractor status of both taxicab and TNC drivers, through California’s Assembly Bill No. 5 legislation (Daus, 2020a). The elements of control over the industry by owners that may result could change it forever. Also, in New York City, a local law that guarantees minimum payments to drivers for a trip dispatched by a high-volume for-hire service (currently Uber, Lyft, and Via) could end up becoming a trend nationally given that the economic rewards for drivers may meet or even be greater than the wages guaranteed for employees under Assembly Bill 5 (N.Y.C. Admin. Code § 19–549, L.L. 2018/150, effective Aug. 14, 2018). Two extreme approaches on either side will not work for the industry or the drivers, as many drivers do not wish to work set hours, yet they desire minimum pay and benefits. Most likely, there will be legislation tailored to gig workers that achieves both results: better wages and benefits, but not de facto control over every aspect of the worker’s daily activities, including the ability to work for numerous employers. This battle will be decided based on labor politics and whether elected officials wish to return to union control and collective bargaining for taxi and TNC drivers or instead opt for a dependent worker scenario that does not involve unionization or full employee status but preserves elements of independent contractor status along with certain benefits and wage guarantees. The broader politics of conservative capitalism versus democratic socialism or progressivism in the United States may lead to certain politically liberal/progressive states and cities enacting such laws first. In terms of other niche services, taxicabs, if they can capitalize on TNC shortcomings, could be the go-to mode for underserved and underbanked communities, as they still accept cash and their rates may be less than those of TNCs. One difficulty is that, in large urban environments, taxicabs continue to cluster on the central business district core, where more affluent passengers are choosing TNCs over taxis. An orchestrated effort or regulatory intervention may need to take place, like the green outer borough taxi system in New York City, to allow for or steer taxicabs to fill the underserved and unbanked communities. It is possible that TNCs may end up taking over service to transportation deserts through public-private partnerships with public transport agencies, such as the Federal Transit Administration’s Sandbox Demonstration Programs, which allow for seamless app integration between TNCs and public transport trains and buses to finish the first and last mile of a trip (see also Chapter 22). Taxicabs would be perfect for such solutions but will need to fight against the lobbying and well-funded public relations machine of TNCs to enter and develop this area of service. In terms of data and privacy, taxicab companies and TNCs have traditionally resisted attempts by regulators to access data, and regulations guard against invasion of privacy in the United States, Europe, and elsewhere (see also Chapter 24). However, to implement connected 265

Matthew W. Daus

and automated vehicle networks, MaaS, universal apps, and/or government subsidy contracts, taxicabs and TNCs will be compelled to share data either by regulation or through business necessity. New York City was the first city to require such data, but across the United States and beyond, mobility providers are slowly starting to share data, and laws will need to be passed to ensure privacy and guard trade secrets as the existing legal paradigm for government control of such data was not built for such innovations when freedom of information laws were enacted. In terms of governance, the system is disjointed and fragmented, with conflicting regulations for taxicabs; TNCs; and other submodes like limousines, buses, shuttles, liveries, and black cars. Uniformity of licensing standards is likely to happen over a period of time and may end up somewhere in between the stringent regulatory regime of taxicabs and the loose “regulation light” TNC model. The duty of care required of business travel and for government and other work will lift the standard of care naturally outside of government licensing requirements, and taxicab regulations on fares, color regimes, and vehicle choice will continue to be loosened. Taxis may look and operate more like TNCs, and regulations may force TNCs to become safer and more responsible. There are also too many different government agencies with conflicting rules for these transportation modes. While it may make sense to have multimodal agencies like the San Francisco Municipal Transportation Agency, Transport for London, the Singapore Land Transport Authority, or the Dubai Roads and Transport Authority, it is unlikely that these consolidations will take place overnight, yet they will need to occur before autonomous and connected vehicle networks appear on a wide-scale basis. The taxicab industry has not taken advantage of developing automated and connected vehicles and has not electrified or upgraded its fleets, as the industry is currently on life support from the growth of the TNCs and the impact of the COVID-19 pandemic and also has its hands tied due to numerous regulations that preclude it from innovating. It is possible that after A.B. 5 becomes a reality and the market experiences mergers and acquisitions and ultimately shrinks in size, bigger players with more capital may emerge that would control more taxicabs, black cars, and limousines, and then true innovation, which costs money to do, could occur, especially under unified ownership. In New York City, based on sales and auctions of distressed medallions, a single player, Marblegate Asset Management, a Connecticut private equity firm, has amassed ownership of almost one-quarter of the medallion taxicabs or medallion loans. If this company leads and places new electric and autonomous vehicles on the road with enhanced services, then this could be a revolutionary change for the industry. Only time will tell, but there is opportunity for widespread change and the reinvention and comeback of the taxicab. TNCs, however, are likely to stay, as they have become too big to fail, and if they cannot turn a profit, they are more likely to be bought or resort to reorganization bankruptcy than shut down or be dissolved or liquidated. Many of the previous predictions are based upon the unique situation in the United States, and the same does not necessarily hold true for other countries or continents. In Europe, TNCs were controlled and could not grow as much as they have in the United States, and taxicabs are still relatively strong in European and Middle Eastern countries. Canada has many of the same problems and issues with TNCs as the United States, but the problems are not as severe and are more slowly materializing in some provinces. Australia has experienced TNC disruption more in big cities like Sydney and Melbourne than other more suburban or less-populated cities around the continent, but the disruption to taxicabs in those urban populations is similar to the that in United States and Canada. South America, Southeast Asia, and Eurasia have many different TNCs, but the dichotomy and politics of the U.S. taxi experience are not the same around the world. Many of the service and safety issues are truths and problems, but the future in those countries may turn on much different factors than discussed here involving the U.S. taxi model. 266

Taxis

In stating this, there is a future for taxicabs, but what it will look like and how well it will function is dependent on the taxi industry taking advantage of niche opportunities and lessons from TNCs on how to better serve the customer.

Conclusion Where taxis were once the leader in for-hire passenger transportation, they are now struggling to keep up and have lost significant market share in those jurisdictions with the app companies have launched operations within the past decade. This is especially so in markets that have TNCs operating with a reduced regulatory framework when compared to taxis. While taxis have seen their market share plummet with the advent of TNCs, new modes are entering the marketplace. More mobility options like electric bikes, electric scooters, bicycles, electric skateboards, shared bicycles, electric pedal-assisted bicycles, micro transit, demand-responsive transport, and other shared mobility will further compete with taxis for passengers. As the public transport industry rapidly evolves, so must taxis. As new ways of providing service and doing business develop, so too must the taxi business adapt. To navigate the road forward, taxis will need to be nimble and ready – and willing – to adapt to changes in the market. They will also need creative problemsolving to not just survive but thrive again. However, without equally flexible regulations that allow taxis to evolve, the industry will be hampered to make material, necessary changes. Based on what is observed today, the following are suggestions for future research areas: • Emerging technology. With the development of taximeters and subsequent in-vehicle technology and dispatching systems, the taxicab industry has shown that innovation through technology is a part of its history. A keen topic for future research is how – and whether – taxicabs will again utilize technology to compete with the app companies to address passenger demand. If taxis use new and emerging technologies similar to TNCs, what effect does this have on rider demand for taxis? Taxis may have an image issue – that is, even if taxis adopt technologies similar to TNCs, passengers may still prefer TNCs over taxis due to the perception that the TNCs are user friendly and technologically advanced. • The impact of the pandemic. As various regions around the world reopen from the COVID-19-related lockdown and social distancing norms are mandated, it will be of interest to examine the long-term impact on the taxi industry and how the reluctance of passengers to return to public transport or shared rides may lead to new opportunities for taxicabs, particularly those taxicabs with partitions and stringent vehicle disinfecting criteria, as well as new business lines that cross-pollinate passenger trips with food and package delivery, which has happened in many jurisdictions. • The impact of micromobility. Having already lost market share to the app companies within the past decade, there may be further erosion due to electric bikes, electric scooters, bicycles, electric skateboards, shared bicycles, and electric pedal-assisted bicycles. Additional research will be needed to determine whether these modes of transportation are a substitute for taxis that further reduce taxis’ share of the market or whether they are their own market and, if so, whether they complement taxi services.

References Access Living of Metro. Chicago v. Uber Techs., Inc., 958 F.3d 604 (7th Cir. 2020). Akers, M. (2020, May 7). Taxi trips in Las Vegas down 97 percent since February. Las Vegas Review-Journal. www.reviewjournal.com/business/tourism/taxi-trips-in-las-vegas-down-97-percent-since-february-2

267

Matthew W. Daus 023518/?eid=EAAAALriw%2BCPS3MlsiGok6cQkXPoEP%2Feug7dUp8pqMVTlEu5xlCex5NLa1f yRwUgfPbttQ%3D%3D&ecf34Msi=1165&emi8s9Kj=608 Buettner, R. (1997, October 1). Taxi medallion roll to a record price. New York Daily News. www.nydai lynews.com/archives/news/taxicab-medallions-roll-record-price-article-1.772341 Carmody, D. (1985, December 16). Taxi medallion cost pushed to $100,000 by heavy demand. The New York Times. www.nytimes.com/1985/12/16/nyregion/taxi-medallion-cost-pushed-to-100000-byheavy-demand.html Chan, M. (2020, May  15). New York city’s taxi drivers are in peril as they brave the coronavirus and uncertain futures. TIME. https://time.com/5836223/nyc-taxi-drivers-coronavirus/?eid=EAAAAAp ABkHLF5DmYHhW1gVx2FL%2BUoV%2Bzm7TxjBV8fC9QReFis8o8eVrPFJUG6aSgY4leg%3D %3D&ecf34Msi=1171&emi8s9Kj=608. Congestion Surcharge, N.Y. Tax Law, Article 29-C. www.nysenate.gov/legislation/laws/TAX/A29-C Court of Justice of the European Union. (2017, December 20). PRESS RELEASE No 136/17. https:// curia.europa.eu/jcms/upload/docs/application/pdf/2017-12/cp170136en.pdf Daus, M. (2016). The expanding transportation network company “equity gap”. University Transportation Research Center. www.whosdrivingyou.org/wp-content/uploads/2016/08/Equity-Report-FINAL-11232642.pdf Daus, M. (2020a, January  30). The infamous A.B.5  – California vs. New York: A  primer to prepare for the labor fight of the century. Black Car News. www.blackcarnews.com/article/the-infamousa-b-5-california-vs-new-york Daus, M. (2020b, June 1). How transportation regulators are responding to COVID-19. Chauffeur Driven. www.chauffeurdriven.com/news-features/industry-news/2565-how-transportation-regulators-areresponding-to-covid-19.html Farrell, R. (2018, July 23). Arrival of electric taxis in NYC in July 1897. UPS Battery Center Ltd. www. upsbatterycenter.com/blog/arrival-of-electric-taxis-new-york-1897/ Gilbey, W. (1903). Early carriages and roads. Vinton & Co. Ltd. Graham, J. (2018, October  21). Competition hacks taxi business. Boston Herald. www.bostonherald. com/2018/10/21/competition-hacks-taxi-business/. Jackson, K. (2010). The encyclopedia of New York City. Yale University Press. Jones, S. (2019, January 31). Uber and Cabify to suspend operations in Barcelona. The Guardian. www. theguardian.com/world/2019/jan/31/uber-cabify-suspended-operations-barcelona. Koffman, D. (2016). Transportation network companies and paratransit: Issues and opportunities. In C. Mulley & J. D. Nelson (Eds.), Paratransit: Shaping the flexible transport future (pp. 377–390). Emerald Publishing Limited. doi:10.1108/S2044-994120160000008018 Lay, M. (2018). The harnessing of power: How 19th century transportation innovators transformed the way the world operates. Cambridge Scholars Publishing. Lewis, B. (2020, May 18). Pandemic pushes half of San Diego’s taxis off the road. CBS8. www.cbs8.com/ article/travel/pandemic-pushes-half-of-san-diegos-taxis-off-the-road/509-fb7ea473-63be-419bb7a4-391dc2cd8d75?eid=EAAAAJj%2FAuWJNMreegvpDWorUqeW7fOW9vutLYCS0D4afvBp4%2 FezexVNlTL0ahvaiUOZyA%3D%3D&ecf34Msi=1172&emi8s9Kj=608 Li, L., & Cheng, T-F. (2019, May 27). Uber spat with Taiwan could force exit from market. Nikkei Asian Review. https://asia.nikkei.com/Business/Companies/Uber-spat-with-Taiwan-could-force-exit-from-market. McCall, V. (2020, April  30). Chicago cab drivers face ‘impossible situation’ during pandemic. WBEZ. www.wbez.org/stories/chicago-cab-drivers-face-impossible-situation-during-pandemic/bcc7ea0fa14a-4e68-a662-b170cf89e544. New York City, N.Y., Code § 19–549, L.L. 2018/150 (effective 2018, August 14. www.westlaw.com/ Document/I7C2B4F34D88540F3984BE5AC03A19380/View/FullText.html?transitionType=Defaul t&contextData=(sc.Default)&VR=3.0&RS=cblt1.0 New York City, N.Y., Local Law 2018/147. (effective 2018, August 14). https://legistar.council.nyc.gov/ View.ashx?M=F&ID=6467078&GUID=F5AFBAEE-1A39-4540-B4F1-D386701C52B9 Noguellou, R., & Renders, D. (2018). Uber & Taxis. Comparative Law Studies. Bruylant. Retrieved February 8, 2021 from https://dial.uclouvain.be/pr/boreal/object/boreal:201981 NYC Taxi & Limousine Commission. (n.d.a). Current and past auction results. https://www1.nyc.gov/site/ tlc/businesses/medallion-auction.page NYC Taxi & Limousine Commission. (n.d.b). Yellow taxi, Green taxi, and FHV monthly indicators. https:// www1.nyc.gov/site/tlc/about/aggregated-reports.page NYC Taxi & Limousine Commission. (n.d.c). Medallion transfers. https://www1.nyc.gov/site/tlc/busi nesses/medallion-transfers.page

268

Taxis NYC Taxi & Limousine Commission. (2004). 2004 annual report to the New York City Council. https:// www1.nyc.gov/assets/tlc/downloads/pdf/2004_annual_report.pdf NYC Taxi & Limousine Commission. (2014). 2014 taxicab fact book. https://www1.nyc.gov/assets/tlc/ downloads/pdf/2014_tlc_factbook.pdf NYC Taxi & Limousine Commission. (2019). New York City Taxi and Limousine Commission 2019 annual report. https://www1.nyc.gov/assets/tlc/downloads/pdf/annual_report_2019.pdf OSHA (2020). COVID-19 guidance for rideshare, taxi, and car service workers. www.osha.gov/Publications/ OSHA4021.pdf Reuters. (2019, December 13). Uber submits appeal to regain London taxi license. https://www.reuters.com/ article/us-uber-britain/uber-submits-appeal-to-regain-london-taxi-license-idUSKBN1YH1ST Robinson, G. (2019, November 25). Application for the renewal of a private hire vehicle operator’s license. Transport for London. http://content.tfl.gov.uk/tfl-decision-letter-to-ull-25-november-2019.pdf. Rodriguez, Y., & Levin, S. (2020). Report of the Taxi Medallion Task Force. https://council.nyc.gov/ data/wp-content/uploads/sites/73/2020/01/Taxi-Medallion-Task-Force-Report-Final.pdf SAE International. (2018, September  24). Taxonomy and definitions for terms related to shared mobility and enabling technologies. www.sae.org/standards/content/j3163_201809/ Shead, S. (2019, May  10). Uber’s rocky road to growth in Europe: Regulators, rivals and riots. Forbes. www.forbes.com/sites/samshead/2019/05/10/ubers-rocky-road-to-growth-in-europe-regula tors-rivals-and-riots/#4c6837d05c67 Singh, D. Z. (2018). The auto-colectivo: A cultural history of the shared taxi in Buenos Aires (1928–33). The Journal of Transport History, 39, 55–71. doi:10.1177/0022526618762344 Tang, S. K. (2019, July 8). New bill looks to regulate private-hire car operators with licenses and fare policy. Channel News Asia. www.channelnewsasia.com/news/singapore/p2p-passenger-transport-industry-billtabled-parliament-11701974. The Straits Times. (1910, June  30). Taxicabs for Singapore. https://eresources.nlb.gov.sg/newspapers/ Digitised/Article/straitstimes19100630-1.2.72 The Straits Times. (1966, February  25). Policy on taxis to be reviewed.  https://eresources.nlb.gov.sg/ newspapers/Digitised/Article/straitstimes19660225-1.2.28 Toronto Municipal Code, Ch. 546. www.toronto.ca/legdocs/municode/toronto-code-546.pdf Walter, C. (2019, October 29). Rideshare rules in Calgary to be reviewed with public feedback. Daily Hive. https://dailyhive.com/calgary/rideshare-rules-calgary-review

269

PART C

Public transport environment

20 THE HEALTH IMPACTS OF PUBLIC TRANSPORT Melanie Crane and Christopher Standen

Introduction In 1953, a seminal paper was published which revealed an association between physical activity and coronary heart disease (CHD) (Morris et al., 1953). The study assessed the amount of physical activity done by London Transport workers during the course of their work, comparing the physical activity achieved by bus conductors during their work with the amount of activity achieved by the drivers. It was found that bus conductors, whose roles on the buses were more physically active, had lower early mortality rates than the drivers. Not only was there a clear difference in the incidence of CHD, but also in the severity of the disease, which was attributed to different levels of activity between the two roles. Since this landmark study, great strides have been made in learning how health and transport are related and how different modes of travel can influence health outcomes.

What are the health impacts of public transport? Public transport can contribute to health in many positive ways. In this chapter, the most significant ones are discussed, including chronic disease prevention through physical activity and benefits to mental health and wellbeing. These impacts are summarised in Table 20.1. How public transport enables access to health services and the potential walkability of an area – and its associated benefits – is explored. The negative health impacts from air pollution and road trauma associated with public transport are considered, with comparisons to other modes of transport. Finally, the association between public transport use and the risk of contracting infectious diseases is discussed. The role of public transport in reducing social exclusion and associated health risks is covered in Chapter 26.

Physical activity Physical inactivity is estimated to cost the global economy more than $67.5 billion in healthcare expenditure and lost productivity (Ding et al., 2016). Physical activity is important for normal growth and development, cardiorespiratory health, musculoskeletal and brain health (including cognitive function), sleep and quality of life (World Health Organization, 2010). Improving 273

Melanie Crane and Christopher Standen Table 20.1 Summary of main health impacts of public transport provision Potential benefits Physical activity

Air pollution

Road injury and trauma Stress and wellbeing

Potential harms

Promotes active travel (walking and cycling for transport) Provides opportunities for incidental physical activity in egress Reduces the risk of chronic diseases (cardiovascular diseases, diabetes, obesity) Supports walkable environments Reduced exposure to air pollution (if alternative clean fuel or electric vehicles are used) Reduced dependency on private vehicles Lower risk of injury than other modes Improves social connection through accessibility

Improves access to health care and community services

Access to services

Risk of infectious diseases

Pollution from vehicles using fossil fuels and non-exhaust sources (brake and tyre wear) Exposure to air pollution inside bus stops Risk of injury falling within vehicle or in egress Long-distance commuting produces stress and lower quality of life Noise pollution contributing to noise annoyance and sleep disturbance Promotion of unhealthy foods and beverages at transport stations increases the risk of obesity Crowding increases the spread of infectious airborne diseases (e.g., influenza) Long trip durations increase the spread of some contagious infections

physical activity can reduce the risk of illness from chronic diseases (including heart diseases, type II diabetes and some cancers); improve cardio-respiratory and muscular fitness; lower the risk of anxiety, depression and dementia, and increase life expectancy (Kohl 3rd et al., 2012; Lee et al., 2012). New evidence also suggests certain population groups can receive specific benefits by engaging in adequate physical activity: a lower risk of fall-related injuries for older adults; reduced risk of excess weight gain, gestational diabetes and postpartum depression in pregnant women; improved bone health and weight status in young children; improved cognitive function in youth; and reduced risk of all-cause mortality in people with chronic medical conditions (Piercy et al., 2018).

What is a sufficient amount of physical activity? The World Health Organization (WHO) recommends adults should do at least 150 minutes of moderate-intensity physical activity throughout the week, or at least 75 minutes of 274

The health impacts of public transport

Figure 20.1 Physical activity domains

vigorous-intensity physical activity or an equivalent combination of both. Additional benefits are gained by increasing the intensity of physical activity and the time to 300 minutes per week and doing muscle-strengthening activities (such as climbing stairs) on two or more days per week. WHO recommends children and adolescents do at least 60 minutes of moderate- to vigorous-intensity physical activity daily, with muscle- and bone-strengthening activities done at least three times per week.

How does public transport improve physical activity? There are four main domains through which physical activity is obtained: recreation, occupation, housework (including gardening) and transport (Figure 20.1). Global trends have seen a substantial decline in transport-related physical activity in the last 30–50 years (Kohl 3rd et al., 2012). “Active transport” is any form of transportation that incorporates physical activity and is generally inclusive of walking and cycling. Public transport is often also included because it generally incorporates some form of physical activity, such as walking/cycling to or from a public transport stop/station. Guidelines in many countries consider public transport stops/stations within walking distance if they are less than 400 metres or ¼ mile from the trip origin or destination. Over a five-day working week, this distance – if walked – can provide a means of achieving sufficient amounts of physical activity. In a systematic review of public transport studies reporting on physical activity levels published before 2012, Rissel et al. (2012) established that a person who travels to work by public transport over a week will acquire 8–33 minutes of physical activity per day (on average 15 minutes). Over a week, this can equate to 150 minutes of light to moderate physical activity. The authors also note the physical activity benefits of public transport appear to mostly accrue to the least active people in the community. That means public transport provides an effective way of encouraging the least active in the community to participate in some form of physical activity. 275

Melanie Crane and Christopher Standen

It is now recognised that a small amount of incidental physical activity, if undertaken regularly (such as walking even short distances to and from public transport provision), is important for health  – particularly for disrupting sedentary behaviour (too much sitting time), which independently contributes to poor health (Word Health Organization, 2010). Public transport use supports incidental activity, thus reducing the prolonged sitting time that is a feature of many office jobs. Standing while waiting at the bus or train station, standing up on the train or ferry, walking between the bus and train services or climbing stairs to a station platform all provide small opportunities for incidental physical activity. Recent systematic reviews provide further insight into the physical activity benefits of public transport. Collating the evidence from evaluation studies measuring the before-andafter effect of new public transport interventions (light rail and bus rapid transport systems) confirms the associations made by Rissel et al. (2012). The latest evidence suggests that proximity to new public transport does encourage walking and can lead to a significant increase in light to moderate physical activity among new public transport users (Hirsch et al., 2018; Xiao et al., 2019). The provision of public transport also increases energy expenditure, equivalent to about 30 minutes additional activity per week (Xiao et al., 2019). Cardio-metabolic health benefits (reductions in obesity prevalence, type II diabetes and cardiovascular risk) are also found, such that using public transport for commuting reduces body mass index [BMI: weight(kg)/height(m)2], while shorter residential distance from new bus services may also contribute to obesity reduction (Patterson et  al., 2019). These studies also reveal research and practice challenges: the ability to contribute to greater-intensity (moderate to vigorous) activity necessary to improve morbidity outcomes, such as through cycling to/from a station, remains dependent upon the availability of cycling infrastructure integrated with public transport systems (Xiao et al., 2019). Likewise, interventions in other domains need to help support an increase in total physical activity in the population so that population levels of sufficient physical activity increase overall (and do not decrease in another domain as a trade-off) (Hirsch et al., 2018).

WALKABILITY AND PUBLIC TRANSPORT PROVISION

One of the four key objectives of the Global Action Plan on physical activity 2018–2030 is to promote the integration of transport and urban planning policy to deliver walkable city neighbourhood designs and transport systems that will promote walking and cycling (World Health Organization, 2018a). Walkability reflects the ease of walking around a neighbourhood (Frank et al., 2010). The walkability of an area is determined by land use diversity (mix of residential, commercial, institutional, and public and green spaces for recreation), residential density, street connectivity, desirability (safe, attractive and accessible) and pedestrian/bicycle infrastructure separate from road traffic (Giles-Corti et al., 2016; Saelens & Handy, 2008; Sallis et al., 2012). Distance to frequent public transport services also contributes to walkability by connecting and promoting walkable areas. Neighbourhoods that are highly walkable provide physical activity and other health benefits, including psychological benefits for all ages, including older adults (Hajna et al., 2015). Various walkability indices have been developed at neighbourhood and district levels to identify neighbourhood-level built environment characteristics contributing to a neighbourhood’s high or low level walkability. Several commercial walkability scores exist across countries, including the Walk Score that measures distance to amenities and how well an area is serviced by public transport (Hall & Ram, 2018).

276

The health impacts of public transport

Ambient air pollution The World Health Organization estimates that ambient air pollution causes at least 4.2 million deaths every year, as well as the loss of 84 million years of life or productive life (disabilityadjusted life years) (World Health Organization, 2016). Children and unborn babies, older adults and those with existing lung or heart diseases are most susceptible; for example, a recent systematic review by Khreis et al. (2017) revealed a significant association between asthma onset in children and ambient air pollution. Air toxins known to affect human health include particulate matter (PM), oxides of nitrogen, ozone, carbon monoxide and sulphur dioxide – major sources of which are motor vehicles and industry. Sources of PM include vehicle exhaust pipes and brake, clutch and tyre wear. PM causes cardiovascular and respiratory incidents, leading to worse symptoms and health care utilisation, and long-term exposure increases mortality; there is no safe level of PM exposure (Anderson et al., 2012). Proven measures for reducing the health effects of ambient air pollution include improving access to public transport (with its lower emissions per passenger-kilometre), fuel efficiency standards for buses and other vehicles and restricting private vehicle use (Landrigan, 2017). However, many towns and cities worldwide, in low- and high-income countries, continue to welcome and encourage motor vehicle traffic (Laurance & Arrea, 2017), resulting in a significant public health burden (Kheirbek et al., 2016; Mueller et al., 2017).

Impacts on public transport users Air toxin concentrations decrease with distance from their source, meaning higher exposure for people living, working or attending childcare/school/college close to busy roads or travelling along them. When travelling, people sharing a lane with motor vehicle traffic generally have the highest exposure: a systematic review of commuters’ exposure to ambient air pollution by mode of transport found that car and bus occupants have the highest exposure (Cepeda et al., 2017), followed by bicycle users and pedestrians, with rail commuters having the lowest exposure. Actual inhalation dose, however, was found to be highest for bicycle users and pedestrians – which includes people accessing and egressing public transport stations/stops – because physical activity increases ventilation rate. Cepeda et al. (2017) did not include travel by passenger ferry in their review; however, a separate study by Chan et al. (2002) found particulate levels inside Hong Kong’s passenger ferries and non-air-conditioned buses and trams to be higher than those inside other types of public transport. Public transport users are also exposed to ambient air pollution during stop/station access and egress, waiting time and transfer. Air quality inside semi-enclosed bus shelters next to busy roads can be worse than outside the shelter – but it can be improved by orienting shelter openings away from the road, prohibiting smoking, and locating stops near open space (Hess et al., 2010). Levels of ambient air pollution inside semi-enclosed railway stations with diesel locomotives, and inside underground stations, can be higher than those along busy roads (Moreno et al., 2014). Thus, it is difficult for public transport users in urban areas to avoid exposure to ambient air pollution. In highly polluted Asian cities, many public transport commuters wear face masks to protect themselves; however, most use simple surgical masks, which are not effective in filtering most air pollutants, as opposed to N95/P2-rated masks, which can be – if correctly fitted (Wong et al., 2017). According to the Hierarchy of Controls – a system promoted by numerous safety organisations to minimise or eliminate exposure to hazards (Centers for Disease Control

277

Melanie Crane and Christopher Standen

and Prevention, 2018) – personal protective equipment, such as masks, should be the last resort. Addressing the source of the hazard is the most effective control. (In some cultures, e.g., Japanese, it is also common for people with respiratory infections to wear surgical face masks when in public or using public transport, as a courtesy to prevent infecting others (Burgess & Horii, 2012). In this case, a hazard source is being addressed through mask wearing. Infectious disease risks of public transport are discussed in more detail later in this chapter.)

Impacts on the general public Air toxin emissions per passenger-kilometre are generally lower for buses and trains than for private motor vehicles. However, induced demand and the self-limiting nature of road congestion (Duranton & Turner, 2011) mean that increasing the use of public transport is unlikely to result in a significant and enduring reduction in private vehicle traffic in congested urban areas – unless private vehicle traffic is contemporaneously restricted through pricing or reductions in road or parking capacity. For example, converting general traffic lanes to exclusive bus lanes makes public transport more attractive while also reducing private motor vehicle traffic and associated air pollution. Using cleaner fuels or electric propulsion in public transport vehicles can help. For example, buses running on compressed natural gas instead of petrol or diesel were introduced in Delhi, India, in 2001  – after which ambient levels of carbon monoxide, sulphur dioxide and aromatic hydrocarbons decreased, although levels of nitrogen dioxide increased (Ravindra et al., 2006). Transitioning bus fleets (and private vehicle fleets) from internal combustion to electric propulsion will help to reduce ambient air pollution. However, up to 90% of the particulate pollution generated by road traffic comes from non-exhaust sources, namely “brake wear, tyre wear, road surface abrasion and resuspension in the wake of passing traffic” (Thorpe & Harrison, 2008, p. 270).

Road injury and trauma The World Health Organization reports that road traffic crashes kill an estimated 1.35 million people each year (World Health Organization, 2018b); millions more are seriously injured and/ or permanently disabled, and road traffic crashes are the leading cause of death for people aged 5–29 years.

Risks for public transport users Public transport passengers have a lower risk of being killed or injured in a crash than people walking, cycling or driving (Teschke et al., 2013). Most public transport trips do include walking and/or cycling for access, egress and transfer; however, the distances – and therefore exposure – are relatively low, meaning public transport users have a lower overall crash injury risk than users of other surface transport modes. Public transport passengers can also be injured by falling inside a vehicle. Elvik (2019) estimates this risk to be 0.3–0.5 injuries per million passenger-kilometres and 0.8–1.7 injuries per million passenger trips. A further trauma risk is violence or harassment, which can occur during any stage of a public transport journey (in-vehicle, waiting, access, egress or transfer). The objective and perceived risks vary considerably across countries, regions and cultures; however, they tend to be greater for women than for men (Neupane & Chesney-Lind, 2014).

278

The health impacts of public transport

Many public transport authorities are taking steps to address this gender inequality; for example, Transport for London’s ‘Report it to stop it’ campaign (BBC News, 2018) and women-only rail carriages in Iran, Japan, India, the UAE, Egypt and Brazil (Newton, 2016). Gekoski et al. (2015) have also found evidence that additional staffing, policing and security cameras can be effective. They warn, however, that many campaigns “are perceived to implicitly or explicitly blame or shame women for sexual harassment and assault, effectively re-victimising them” (p. 38), citing examples from Iran, Singapore, Canada and the United Kingdom.

Risks for the general public Buses and street-running trams share space with pedestrians, bicycles and other vehicles, while many railways have at-grade crossings for pedestrian and/or vehicle traffic. Public transport vehicles are large and heavy, meaning the consequences of one crashing into another vehicle or a pedestrian can be severe. On the other hand, they are usually operated by highly trained drivers, while trams and guided buses move along predictable paths. A 2005 analysis of pedestrian fatalities in the United States (Paulozzi, 2005) concluded that, per vehicle-kilometre, buses were about ten times more likely to kill a pedestrian than a car. However, only 1.5% of total pedestrian fatalities involved buses, owing to lower overall vehiclekilometres. It should also be noted that average passenger occupancy of buses is higher than that of cars, so the risk per passenger-kilometre for buses is comparable to that of cars. That said, if the widely adopted aim of zero road fatalities is to be achieved, more must be done to reduce the risk of injury posed by all vehicles, including buses and other public transport vehicles, to vulnerable road users. For example, the practice of designing intersections to give a green walk signal to pedestrians concurrently with a green signal for buses (and other vehicles) turning across their path must be seriously reconsidered. For the aforementioned reasons, increasing the use of public transport is unlikely to result in significant and enduring reductions to injuries caused by private vehicles in congested urban areas – unless private vehicle traffic volumes and speeds are also further restricted.

Stress and wellbeing Connections between mental health and wellbeing (quality of life) in relation to transportation are becoming an increasing concern, particularly as cities grow in population density and spatially, with implications for health associated with traffic congestion and travel times (GilesCorti et al., 2016).

Wellbeing while travelling The last decade has seen an emerging interest in valuing subjective wellbeing or quality of life, particularly with regard to the individual’s experience of their trip or journey (Abou-Zeid & Ben-Akiva, 2011; Carse, 2011; De Vos et al., 2013; Ettema et al., 2010; St-Louis et al., 2014) and life satisfaction (Bergstad et al., 2011; De Vos et al., 2013; Delbosc, 2012). Travel satisfaction is often used by transport planners and researchers to measure transport-related wellbeing; yet, this does not adequately capture aspects of quality of life considered foundational to health – such as physical health, mental health and social interactions (Lee & Sener, 2016; Rissel et al., 2016). Life satisfaction may be directly influenced through physical mobility or indirectly through access to lifestyle choices.

279

Melanie Crane and Christopher Standen

In general, public transport has tended to be associated with poorer quality of life; however, the evidence is not so clear cut (Gottholmseder et al., 2009; Novaco & Gonzalez, 2009; StLouis et al., 2014; Wheatley, 2014). Most research attention has focused on mental distress (stress) associated with commuting to work. Commuting stress can have a profound impact on psychological adjustment (Novaco & Gonzalez, 2009). Long commuting times have been linked to depression and higher psychological distress among both car commuters and public transport users (Ding et al., 2014; Feng & Boyle, 2014; Gottholmseder et  al., 2009). People moving to newly constructed greenfield developments on the outskirts of cities (where housing development usually precedes public transport infrastructure provision) have also reported unanticipated stress (Kent et  al., 2019). Stress can also be a factor in short distances when the journey is out of the individual’s control or the individual feels impeded, such as in conditions of increased traffic congestion (Gottholmseder et al., 2009). Wener and Evans (2011) suggest that car commuters experience higher levels of stress than public transport users because of the greater unpredictability of the trip (Wener & Evans, 2011). Gatersleben and Uzzell (2007) compared the experiences of drivers, cyclists, walkers and public transport users amongst university employees in a small study the United Kingdom. The findings from this study reveal that car users experience the most stress in their journey, while public transport users experience boredom as well as stress. Similar findings have been found by Morris and Guerra in the United States (Morris, 2015; Morris & Guerra, 2014). There are a few things to note. First, these studies have mainly focused on individual transport modes and have largely ignored multi-modal trips involving public transport and active travel. A more recent study on commuting stress of hospital workers included uni-modal and multi-modal active travel (public transport, walking and cycling). Compared with car commuters, active and public transport commuters reported lower stress levels (Rissel et al., 2014). Second, the studies indicating public transport user boredom were conducted before the innovation and widespread diffusion of technologies such as smartphones and tablets. More recent evidence suggests these social and entertainment technologies help to counteract potential stress and boredom (Ettema et al., 2012). Third, several studies have shown that the fundamental ability to travel is associated with a greater degree of happiness, while inability to travel leads to a lower quality of life (Kolodinsky et al., 2013; Spinney et al., 2009). This has led to a focus on personal transport, particularly for people with disabilities or older adults participating in the community (Ravulaparthy et al., 2013; Spinney et al., 2009; van Roosmalen et al., 2010). With ageing populations in many countries, the ability to travel independently will largely depend on the provision of well-connected public transport services (see also Chapter 28).

Life satisfaction and community wellbeing Community wellbeing is often understood in terms of the level of social capital and social cohesion (Phillips, 2012). Transport can contribute to the social capital and cohesiveness of society, such as through access to community services (discussed in the next section). Increasing evidence also suggests neighbourhood design, in relation to traffic conditions, also impacts community wellbeing. Busy road traffic creates community severance and contributes to poorer quality of life (Anciaes et al., 2019; Foley et al., 2017; Gundersen et al., 2013). Increasing public transport services can therefore help to reduce many of the health impacts of local road traffic, including wellbeing. Public transport may, however, contribute to poorer wellbeing outcomes through noise pollution. Exposure to noise from road and rail transport contributes to noise annoyance and sleep disturbance, with negative impacts on quality of life, as well as cognitive development of 280

The health impacts of public transport

children and risk of cardiovascular diseases (Clark & Stansfeld, 2007). Interventions to reduce the health impacts of rail and road transport noise include regulations to limit noise emissions, road buffers, increasing green space, urban planning controls and new or closed infrastructure (Brown & Van Kamp, 2017)

Access to health care, community services and healthy food choices Poor access to health care facilities (e.g., general practices, hospitals and pharmacies) is associated with higher rates of avoidable hospital admissions (Ansari et al., 2006), while poor access to community facilities (e.g., libraries and community centres) is associated with poor mental health and vitality (Guite et al., 2006). Thus, improving access to community and health services is likely to improve these outcomes. Improving access to affordable healthy food is also important in reducing risk of overweight and obesity and associated health problems (GhoshDastidar et al., 2014). Food marketing in public transport stations and vehicles and in-station snack food vending machines are counteractive strategies for obesity prevention. One audit of train station advertisements across Sydney showed more than 80% of food advertisements to be for energydense discretionary foods (snack foods and sugar-sweetened beverages) (Sainsbury et al., 2017). A similar audit in New York showed subway station advertising of unhealthy foods and drinks targeted the most vulnerable population groups (Riley et al., 2018). Many cities have proposed or enacted bans on the advertising of unhealthy foods and beverages; however, the sale of unhealthy foods at stations remains a major challenge. Good-quality, affordable public transport helps make access to health care, community facilities and healthy food more equitable, especially for people living in rural or car-dependent suburban areas and unable to drive – the number of which is set to increase significantly in countries with ageing populations. However, public transport serving such populations often does not attract enough ridership to make them financially viable, and they depend on large government subsidies (Walker, 2012). For this reason, there is growing interest among public transport authorities and providers in flexible public transport services, whereby bus routes adjust to passenger demand  – with passengers requesting pick-up with a mobile phone app or phone call and smaller (lower-cost) vehicles typically used (Hensher, 2017). In some cases, public transport infrastructure might inhibit access to health care and community facilities. For example, an at-grade railway line with few crossings can act as a barrier between residents and facilities, significantly increasing network distances between them – even though the straightline distance between them may be short.

Infectious diseases “Infectious diseases” is an umbrella term for illnesses caused by bacteria, viruses and parasites – including tuberculosis, hepatitis, malaria, cholera, measles and influenza. Infectious diseases are spread by human-to-human contact, either directly or indirectly, or by animals to humans (zoonotic). Safe, effective and affordable vaccines and drug therapies have had an enormous impact on reducing or eliminating the spread of infectious diseases in high-income countries; however, many vaccine-preventable (e.g., typhoid, dengue, malaria) and drug-resistant (e.g., tuberculosis) diseases remain a major burden of disease in low-income countries (James et  al., 2018). The threat of emerging disease outbreaks, particularly airborne diseases, has become of increasing global concern given increasing globalisation, urban population densities and the role of human movement (travel behaviour patterns and accessibility). This presents 281

Melanie Crane and Christopher Standen

many challenges for public health, particularly with regard to controlling the global spread of diseases (Budd et al., 2009). Airborne respiratory virus outbreaks in recent years have spread quickly and widely, with high impacts on mortality – the most recent and largest scale being the COVID-19 coronavirus. Other recent airborne respiratory virus outbreaks include SARS (severe acute respiratory syndrome), the Middle East respiratory syndrome coronavirus (MERSCov), and H1N1 and H5N8 bird flu viruses.

Infectious disease transmission and public transport Contagious airborne diseases, like influenza, tuberculosis or coronaviruses, are transmitted by droplets (primarily) when speaking, coughing or sneezing. Places where a large number of people congregate, like crowded buses and train carriages or stations, produce environments for spreading these diseases (Mohr et al., 2012). A recent study modelling the impact of influenza-like illnesses in relation to the London Underground showed a high correlation between the use of public transport and spread of disease (Goscé & Johansson, 2018). Travel patterns showed that passengers travelling from areas with high rates of infection (transport hubs) interact with a high number of people, likely increasing the spread of infection. Another study, also in the United Kingdom, looked at cases of disease based on general practitioner patient records and data on bus and tram patronage (Troko et al., 2011). The authors revealed a very high association between acute respiratory infection and having travelled by public transport in the days before symptoms appeared. However, they also found that the risk of infection was reduced in the more frequent public transport users, plausibly because they had developed a level of immunity from repeated exposure. The majority of research investigating the role of transportation and infectious diseases has been in air transport, which appears to accelerate and escalate the spread of influenza-like diseases (Browne et al., 2016; Budd et al., 2009). Fewer case studies on ground public transport exist because of the challenges of contract tracing transmission from person to person, creating major gaps in our understanding of the role of public transport systems in transmitting and controlling the spread of these diseases (Browne et al., 2016; Mohr et al., 2012). Trip duration (length of time exposed), proximity to the infected person (face-to-face) and crowding likely increase the spread of contagious infectious diseases (Mohr et al., 2012); however, transmission rates (number of people infected) vary between diseases – for example, tuberculosis has more than twice the rate of infection as influenza yet it takes longer to incubate. Some viruses, like the COVID-19-causing coronavirus, are very successful in spreading because they can survive for a period on surfaces between human-to-human contact. This means that someone who has the disease may pass it on after touching mouth or nose and then touching a public surface, like a bus handrail. Increasing transport efficiency, sanitary conditions and ventilation can help to decrease the spread of disease, alongside isolating cases of infection (Xu et  al., 2013). Modelling studies (on airborne smallpox as an example) suggest that homes, offices and school environments create a higher risk for transmitting infections based on intimacy and contact time; closing schools and workplaces is therefore a key measure in controlling outbreaks (Zhang et al., 2016; Zhang et al., 2018). Suspending heavily used public transport depends on the contagiousness of the disease and if brief contact aids its propagation (Mohr et al., 2012; Zhang et al., 2016). Individual hygiene also plays an important role: not travelling when sick is the best way to help recovery and protect other people; covering one’s mouth when coughing (not with the hand); and washing hands. For controlling the community transmission of COVID-19, WHO guidelines encourage physical distancing and the wearing of face masks. What researchers know about the risk of public transport and its role in the spread of COVID-19 more broadly is based on correlation studies and the evidence is still emerging. 282

The health impacts of public transport

Opportunities and barriers to health The health impacts of public transport include direct and indirect impacts such as those mentioned previously. Calculating the health costs and benefits of new public transport infrastructure is important for minimising threats to vulnerable populations. Many studies have shown that a shift away from private vehicle transport to public transport, walking and cycling will result in population health gains, with the largest gain being a reduction of heart disease, stroke and type II diabetes due to an increase in travel-related physical activity (Mueller et al., 2017; Stevenson et al., 2016; Woodcock et al., 2009; Xia et al., 2015). Some initial increases in road injuries are likely in cities shifting from relatively high private vehicle mode share to walking and cycling. The benefits to respiratory health gained by shifting to higher public transport use are dependent on the change in transport-related particulate emissions, and that requires a change in fuel source and vehicle emissions. The opportunity for public transport to contribute to improving health by reducing the risk of chronic non-communicable diseases relies on decreasing the distance to public transport options and integration with walking and cycling infrastructure. Lack of public transport options and poor integration with walking and cycling infrastructure and mobility options for people with disabilities continue to encourage private motor vehicle transport and perpetuate poor health outcome.

Health impact assessment Health impact assessments (HIAs) are a tool for identifying potential health-related impacts of new project proposals, plans and policies. They can help in planning for public transport infrastructure provision by providing evidence about the potential health benefits or negative impacts that the infrastructure will have on health. What is reported varies widely, depending on how health is framed and if public health experts are engaged in transport planning decisions (Mindell et al., 2008; Riley et al., 2018). HIAs are sometimes voluntarily included as part of the environmental impact assessment process, and very few countries or their jurisdictions explicitly require HIAs to be conducted prior to construction. HIAs typically include five steps: screening, scoping, appraisal, reporting and monitoring of health impacts. Assessments of health impacts should generally include, but are not limited to, health impacts and outcomes explored in this chapter (primarily physical activity, air pollution and road injury), which have the largest direct impact on the burden of disease (Vos et al., 2017). HIAs should consider the population exposed, the health impact associated with the population exposed and equity impacts. Transport planners are unlikely to consider the full scope of health impacts and should look to involve public health authorities and researchers to help improve understanding of societal impacts.

Gaps in research evidence and practice Evidence regarding the indirect costs of transport to health, variations in exposure and the potential benefits that an integrated public transport and active travel system can achieve remain limited. The impact of inequalities of public transport provision and accessibility challenges of the poorer households and older adults also require further investigation and consideration. The role of public transport and transport behaviour in controlling infectious disease transmission is a widening area of concern, now more than ever, and requires further research. The majority of the research on public transport interventions (describing health impacts, mobility challenges and mechanisms and calculating health costs and benefits) comes from 283

Melanie Crane and Christopher Standen

high-income countries. Translation of evidence to low-to-middle income countries (LMICs) is challenging. One issue is informal transport services’ contribution to the health burden, including injury and air quality (Cervero, 2013). Research evidence and monitoring need to be improved to develop coordinated solutions for addressing rising health challenges associated with land use and urban transport in LMICs and in response to COVID-19 challenges.

Conclusion Public transport can have positive and negative impacts on health – it can support and encourage physical activity, thereby helping to reduce the risk of many chronic diseases associated with poor physical activity in the population. Provision of public transport services contributes to improving accessibility to health and community services, thereby contributing to social health. It has a role to play in reducing air and noise pollution and traffic-related injuries, and may provide some improvements to mental health and wellbeing. Efforts to improve public transport provision will lead to improvements in health and should be considered for the benefit of society.

References Abou-Zeid, M.,  & Ben-Akiva, M. (2011). The effect of social comparisons on commute well-being. Transportation Research Part A: Policy and Practice, 45(4), 345–361. Anciaes, P. R., Stockton, J., Ortegon, A., & Scholes, S. (2019). Perceptions of road traffic conditions along with their reported impacts on walking are associated with wellbeing. Travel Behaviour and Society, 15, 88–101. doi:10.1016/j.tbs.2019.01.006 Anderson, J. O., Thundiyil, J. G., & Stolbach, A. (2012). Clearing the air: A review of the effects of particulate matter air pollution on human health. Journal of Medical Toxicology, 8(2), 166–175. Ansari, Z., Laditka, J. N., & Laditka, S. B. (2006). Access to health care and hospitalization for ambulatory care sensitive conditions. Medical Care Research and Review, 63(6), 719–741. BBC News. (2018). Awareness drive sees rise in reports of London transport sex offences. Retrieved February 19, 2020, from www.bbc.com/news/uk-england-london-45902491 Bergstad, C. J., Gamble, A., Gärling, T., Hagman, O., Polk, M., Ettema, D., Friman, M., & Olsson, L. E. (2011). Subjective well-being related to satisfaction with daily travel. Transportation, 38(1), 1–15. Brown, A. L., & Van Kamp, I. (2017). WHO environmental noise guidelines for the European region: A systematic review of transport noise interventions and their impacts on health. International Journal of Environmental Research and Public Health, 14(8), 873. Browne, A., St-Onge Ahmad, S., Beck, C. R., & Nguyen-Van-Tam, J. S. (2016). The roles of transportation and transportation hubs in the propagation of influenza and coronaviruses: A systematic review. Journal of Travel Medicine, 23(1), tav002. Budd, L., Bell, M., & Brown, T. (2009). Of plagues, planes and politics: Controlling the global spread of infectious diseases by air. Political Geography, 28(7), 426–435. Burgess, A., & Horii, M. (2012). Risk, ritual and health responsibilisation: Japan’s ‘safety blanket’ of surgical face mask-wearing. Sociology of Health & Illness, 34(8), 1184–1198. Carse, A. (2011). Assessment of transport quality of life as an alternative transport appraisal technique. Journal of Transport Geography, 19(5), 1037–1045. Centers for Disease Control and Prevention. (2018). Hierarchy of controls. Retrieved January 31, 2020, from https://www.cdc.gov/niosh/topics/hierarchy/default.html Cepeda, M., Schoufour, J., Freak-Poli, R., Koolhaas, C. M., Dhana, K., Bramer, W. M., & Franco, O. H. (2017). Levels of ambient air pollution according to mode of transport: A systematic review. The Lancet Public Health, 2(1), e23–e34. Cervero, R. (2013). Linking urban transport and land use in developing countries. Journal of Transport and Land Use, 6(1), 7–24. Chan, L., Lau, W., Lee, S., & Chan, C. (2002). Commuter exposure to particulate matter in public transportation modes in Hong Kong. Atmospheric Environment, 36(21), 3363–3373.

284

The health impacts of public transport Clark, C., & Stansfeld, S. A. (2007). The effect of transportation noise on health and cognitive development: A review of recent evidence. International Journal of Comparative Psychology, 20(2). Delbosc, A. (2012). The role of well-being in transport policy. Transport Policy, 23, 25–33. De Vos, J., Schwanen, T., Van Acker, V., & Witlox, F. (2013). Travel and subjective well-being: A focus on findings, methods and future research needs. Transport Reviews, 33(4), 421–442. Ding, D., Gebel, K., Phongsavan, P., Bauman, A. E., & Merom, D. (2014). Driving: A road to unhealthy lifestyles and poor health outcomes. PloS ONE, 9(6). Ding, D., Lawson, K. D., Kolbe-Alexander, T. L., Finkelstein, E. A., Katzmarzyk, P. T., Van Mechelen, W., Pratt, M., & Committee, L. P. A. S. E. (2016). The economic burden of physical inactivity: A global analysis of major non-communicable diseases. The Lancet, 388(10051), 1311–1324. Duranton, G., & Turner, M. A. (2011). The fundamental law of road congestion: Evidence from US cities. American Economic Review, 101(6), 2616–2652. Elvik, R. (2019). Risk of non-collision injuries to public transport passengers: Synthesis of evidence from eleven studies. Journal of Transport & Health, 13, 128–136. Ettema, D., Friman, M., Gärling, T., Olsson, L. E., & Fujii, S. (2012). How in-vehicle activities affect work commuters’ satisfaction with public transport. Journal of Transport Geography, 24, 215–222. https:// dx.doi.org/10.1016/j.jtrangeo.2012.02.007 Ettema, D., Gärling, T., Olsson, L. E., & Friman, M. (2010). Out-of-home activities, daily travel, and subjective well-being. Transportation Research Part A: Policy and Practice, 44(9), 723–732. https://dx.doi. org/10.1016/j.tra.2010.07.005 Feng, Z., & Boyle, P. (2014). Do long journeys to work have adverse effects on mental health? Environment and Behavior, 46(5), 609–625. Foley, L., Prins, R., Crawford, F., Humphreys, D., Mitchell, R., Sahlqvist, S., Thomson, H., Ogilvie, D., & Team, M. S. (2017). Effects of living near an urban motorway on the wellbeing of local residents in deprived areas: Natural experimental study. PloS One, 12(4). Frank, L. D., Sallis, J. F., Saelens, B. E., Leary, L., Cain, K., Conway, T. L., & Hess, P. M. (2010). The development of a walkability index: Application to the neighborhood quality of life study. British Journal of Sports Medicine, 44(13), 924. doi:10.1136/bjsm.2009.058701 Gatersleben, B., & Uzzell, D. (2007). Affective appraisals of the daily commute: Comparing perceptions of drivers, cyclists, walkers, and users of public transport. Environment and Behavior, 39(3), 416–431. Gekoski, A., Gray, J. M., Horvath, M. A., Edwards, S., Emirali, A., & Adler, J. R. (2015). ‘What works’ in reducing sexual harassment and sexual offences on public transport nationally and internationally: A rapid evidence assessment. Project Report. Middlesex University; British Transport Police; Department for Transport, London. Ghosh-Dastidar, B., Cohen, D., Hunter, G., Zenk, S. N., Huang, C., Beckman, R.,  & Dubowitz, T. (2014). Distance to store, food prices, and obesity in urban food deserts. American Journal of Preventive Medicine, 47(5), 587–595. Giles-Corti, B., Vernez-Moudon, A., Reis, R., Turrell, G., Dannenberg, A. L., Badland, H., Foster, S., Lowe, M., Sallis, J. F., & Stevenson, M. (2016). City planning and population health: A global challenge. The Lancet, 388(10062), 2912–2924. Goscé, L., & Johansson, A. (2018). Analysing the link between public transport use and airborne transmission: Mobility and contagion in the London Underground. Environmental Health, 17(1), 84. doi:10.1186/s12940-018-0427-5 Gottholmseder, G., Nowotny, K., Pruckner, G., & Theural, E. (2009). Stress perception and commuting. Health Economics, 18, 559–576. Guite, H., Clark, C., & Ackrill, G. (2006). The impact of the physical and urban environment on mental well-being. Public Health, 120(12), 1117–1126. Gundersen, H., Mageroy, N., Moen, B. E., & Bratveit, M. (2013). Traffic density in area of residence is associated with health-related quality of life in women, the community-based Hordaland Health Study. Archives of Environmental & Occupational Health, 68(3), 153–160. Hajna, S., Ross, N. A., Brazeau, A.-S., Bélisle, P., Joseph, L., & Dasgupta, K. (2015). Associations between neighbourhood walkability and daily steps in adults: A systematic review and meta-analysis. BMC Public Health, 15(1), 768. Hall, C. M., & Ram, Y. (2018). Walk score® and its potential contribution to the study of active transport and walkability: A critical and systematic review. Transportation Research Part D: Transport and Environment, 61, 310–324. doi:10.1016/j.trd.2017.12.018

285

Melanie Crane and Christopher Standen Hensher, D. A. (2017). Future bus transport contracts under a mobility as a service (MaaS) regime in the digital age: Are they likely to change? Transportation Research Part A: Policy and Practice, 98, 86–96. Hess, D. B., Ray, P. D., Stinson, A. E., & Park, J. (2010). Determinants of exposure to fine particulate matter (PM2.5) for waiting passengers at bus stops. Atmospheric Environment, 44(39), 5174–5182. Hirsch, J. A., DeVries, D. N., Brauer, M., Frank, L. D., & Winters, M. (2018). Impact of new rapid transit on physical activity: A meta-analysis. Preventive Medicine Reports, 10, 184–190. James, S. L., Abate, D., Abate, K. H., Abay, S. M., Abbafati, C., Abbasi, N., Abbastabar, H., Abd-Allah, F., Abdela, J., & Abdelalim, A. (2018). Global, regional, and national incidence, prevalence, and years lived with disability for 354 diseases and injuries for 195 countries and territories, 1990–2017: A systematic analysis for the global burden of disease study 2017. The Lancet, 392(10159), 1789–1858. Kent, J. L., Mulley, C., & Stevens, N. (2019). Transport and wellbeing in a newly constructed greenfield estate: A  quantitative exploration of the commuting experience. Journal of Transport  & Health, 13, 210–223. Kheirbek, I., Haney, J., Douglas, S., Ito, K., & Matte, T. (2016). The contribution of motor vehicle emissions to ambient fine particulate matter public health impacts in New York City: A  health burden assessment. Environmental Health, 15(1), 89. Khreis, H., Kelly, C., Tate, J., Parslow, R., Lucas, K., & Nieuwenhuijsen, M. (2017). Exposure to trafficrelated air pollution and risk of development of childhood asthma: A  systematic review and metaanalysis. Environment International, 100, 1–31. https://doi.org/10.1016/j.envint.2016.11.012 Kohl 3rd, H. W., Craig, C. L., Lambert, E. V., Inoue, S., Alkandari, J. R., Leetongin, G., Kahlmeier, S., & Group, L. P. A. S. W. (2012). The pandemic of physical inactivity: Global action for public health. The Lancet, 380(9838), 294–305. Kolodinsky, J. M., DeSisto, T. P., Propen, D., Putnam, M. E., Roche, E., & Sawyer, W. R. (2013). It is not how far you go, it is whether you can get there: Modeling the effects of mobility on quality of life in rural New England. Journal of Transport Geography, 31, 113–122. https://dx.doi.org/10.1016/j. jtrangeo.2013.05.011 Landrigan, P. J. (2017). Air pollution and health. The Lancet Public Health, 2(1), e4–e5. Laurance, W. F., & Arrea, I. B. (2017). Roads to riches or ruin? Science, 358(6362), 442–444. Lee, I.-M., Shiroma, E. J., Lobelo, F., Puska, P., Blair, S. N., Katzmarzyk, P. T., & Group, L. P. A. S. W. (2012). Effect of physical inactivity on major non-communicable diseases worldwide: An analysis of burden of disease and life expectancy. The Lancet, 380(9838), 219–229. Lee, R. J., & Sener, I. N. (2016). Transportation planning and quality of life: Where do they intersect? Transport Policy, 48, 146–155. https://dx.doi.org/10.1016/j.tranpol.2016.03.004 Mindell, J., Boltong, A.,  & Forde, I. (2008). A  review of health impact assessment frameworks. Public Health, 122(11), 1177–1187. Mohr, O., Askar, M., Schink, S., Eckmanns, T., Krause, G., & Poggensee, G. (2012). Evidence for airborne infectious disease transmission in public ground transport – a literature review. Eurosurveillance, 17(35), 20255. Moreno, T., Pérez, N., Reche, C., Martins, V., De Miguel, E., Capdevila, M., Centelles, S., Minguillón, M., Amato, F., & Alastuey, A. (2014). Subway platform air quality: Assessing the influences of tunnel ventilation, train piston effect and station design. Atmospheric Environment, 92, 461–468. Morris, E. A. (2015). Should we all just stay home? Travel, out-of-home activities, and life satisfaction. Transportation Research Part A: Policy and Practice, 78, 519–536. doi:10.1016/j.tra.2015.06.009 Morris, E. A., & Guerra, E. (2014). Mood and mode: Does how we travel affect how we feel? Transportation, 42(1), 25–43. Morris, J. N., Heady, J. A., Raffle, P. A. B., Roberts, C. G., & Parks, J. W. (1953). Coronary heart disease and physical activity of work. The Lancet, 262(6796), 1111–1120. doi:10.1016/S0140-6736(53)91495-0 Mueller, N., Rojas-Rueda, D., Basagaña, X., Cirach, M., Cole-Hunter, T., Dadvand, P., DonaireGonzalez, D., Foraster, M., Gascon, M., & Martinez, D. (2017). Urban and transport planning related exposures and mortality: A health impact assessment for cities. Environmental Health Perspectives, 125(1), 89–96. Neupane, G., & Chesney-Lind, M. (2014). Violence against women on public transport in Nepal: Sexual harassment and the spatial expression of male privilege. International Journal of Comparative and Applied Criminal Justice, 38(1), 23–38. Newton, A. D. (2016). Why we shouldn’t dismiss the idea of women-only carriages. The Conversation. Novaco, R. W.,  & Gonzalez, O. I. (2009). Commuting and well-being. Technology and Well-Being, 3, 174–174.

286

The health impacts of public transport Patterson, R., Webb, E., Hone, T., Millett, C., & Laverty, A. A. (2019). Associations of public transportation use with cardiometabolic health: A systematic review and meta-analysis [Review]. American Journal of Epidemiology, 188(4), 785–795. doi:10.1093/aje/kwz012 Paulozzi, L. J. (2005). United States pedestrian fatality rates by vehicle type. Injury Prevention, 11(4), 232–236. Phillips, D. (2012). Quality of life: Concept, policy and practice. Taylor and Francis. Piercy, K. L., Troiano, R. P., Ballard, R. M., Carlson, S. A., Fulton, J. E., Galuska, D. A., George, S. M., & Olson, R. D. (2018). The physical activity guidelines for Americans. Jama, 320(19), 2020–2028. Ravindra, K., Wauters, E., Tyagi, S. K., Mor, S., & Van Grieken, R. (2006). Assessment of air quality after the implementation of compressed natural gas (CNG) as fuel in public transport in Delhi, India. Environmental Monitoring and Assessment, 115(1–3), 405–417. Ravulaparthy, S., Yoon, S., & Goulias, K. (2013). Linking elderly transport mobility and subjective wellbeing [Article]. Transportation Research Record, 2382, 28–36. doi:10.3141/2382-04 Riley, E., Harris, P., Kent, J., Sainsbury, P., Lane, A., & Baum, F. (2018). Including health in environmental assessments of major transport infrastructure projects: A documentary analysis. International Journal of Health Policy Management, 7(2), 144. Rissel, C., Crane, M., Wen, L. M., Greaves, S., & Standen, C. (2016). Satisfaction with transport and enjoyment of the commute by commuting mode in inner Sydney. Health Promotion Journal of Australia, 27(1), 80–83. Rissel, C., Curac, N., Greenaway, M.,  & Bauman, A. (2012). Physical activity associated with public transport use-a review and modelling of potential benefits [Article]. International Journal of Environmental Research and Public Health, 9(7), 2454–2478. doi:10.3390/ijerph9072454 Rissel, C., Petrunoff, N., Wen, L., & Crane, M. (2014). Travel to work and self-reported stress: Findings from a workplace survey in south west Sydney, Australia. Journal of Transport & Health, 1(1), 50–53. Saelens, B. E., & Handy, S. L. (2008). Built environment correlates of walking: A review. Medicine and Science in Sports and Exercise, 40(7 Suppl), S550. Sainsbury, E., Colagiuri, S., & Magnusson, R. (2017). An audit of food and beverage advertising on the Sydney metropolitan train network: Regulation and policy implications. BMC Public Health, 17(1), 490. Sallis, J. F., Floyd, M. F., Rodríguez, D. A., & Saelens, B. E. (2012). Role of built environments in physical activity, obesity, and cardiovascular disease. Circulation, 125(5), 729–737. Spinney, J. E., Scott, D. M., & Newbold, K. B. (2009). Transport mobility benefits and quality of life: A time-use perspective of elderly Canadians. Transport Policy, 16(1), 1–11. Stevenson, M., Thompson, J., de Sá, T. H., Ewing, R., Mohan, D., McClure, R., Roberts, I., Tiwari, G., Giles-Corti, B., & Sun, X. (2016). Land use, transport, and population health: Estimating the health benefits of compact cities. The Lancet, 388(10062), 2925–2935. St. Louis, E., Manaugh, K., van Lierop, D., & El-Geneidy, A. (2014). The happy commuter: A comparison of commuter satisfaction across modes. Transportation Research Part F: Traffic Psychology and Behaviour, 26, 160–170. Teschke, K., Harris, M. A., Reynolds, C. C., Shen, H., Cripton, P. A., & Winters, M. (2013). Exposurebased traffic crash injury rates by mode of travel in British Columbia. Canadian Journal of Public Health, 104(1), e75-e79. Thorpe, A., & Harrison, R. M. (2008). Sources and properties of non-exhaust particulate matter from road traffic: A review. Science of the Total Environment, 400(1–3), 270–282. Troko, J., Myles, P., Gibson, J., Hashim, A., Enstone, J., Kingdon, S., Packham, C., Amin, S., Hayward, A., & Van-Tam, J. N. (2011). Is public transport a risk factor for acute respiratory infection? BMC Infectious Diseases, 11(1), 16. van Roosmalen, L., Paquin, G. J., & Steinfeld, A. M. (2010). Quality of life technology: The state of personal transportation. Physical Medicine & Rehabilitation Clinics of North America, 21(1), 111–125. Vos, T., Abajobir, A. A., Abate, K. H., Abbafati, C., Abbas, K. M., Abd-Allah, F., Abdulkader, R. S., Abdulle, A. M., Abebo, T. A., & Abera, S. F. (2017). Global, regional, and national incidence, prevalence, and years lived with disability for 328 diseases and injuries for 195 countries, 1990–2016: A systematic analysis for the global burden of disease study 2016. The Lancet, 390(10100), 1211–1259. Walker, J. (2012). Human transit: How clearer thinking about public transit can enrich our communities and our lives. Island Press. Wener, R. E., & Evans, G. W. (2011). Comparing stress of car and train commuters. Transportation Research Part F: Traffic Psychology and Behaviour, 14(2), 111–116.

287

Melanie Crane and Christopher Standen Wheatley, D. (2014). Travel-to-work and subjective well-being: A study of UK dual career households. Journal of Transport Geography, 39, 187–196. Wong, L. P., Alias, H., Aghamohammadi, N., Ghadimi, A., & Sulaiman, N. M. N. (2017). Control measures and health effects of air pollution: A survey among public transportation commuters in Malaysia. Sustainability, 9(9), 1616. Woodcock, J., Edwards, P., Tonne, C., Armstrong, B. G., Ashiru, O., Banister, D., Beevers, S., Chalabi, Z., Chowdhury, Z., & Cohen, A. (2009). Public health benefits of strategies to reduce greenhouse-gas emissions: Urban land transport. The Lancet, 374(9705), 1930–1943. World Health Organization. (2010). Global recommendations on physical activity for health. WHO. World Health Organization. (2016). Ambient air pollution: A global assessment of exposure and burden of disease. WHO. World Health Organization. (2018a). Global action plan on physical activity 2018–2030: More active people for a healthier world. WHO. World Health Organization. (2018b). Global status report on road safety. WHO. Xia, T., Nitschke, M., Zhang, Y., Shah, P., Crabb, S., & Hansen, A. (2015). Traffic-related air pollution and health co-benefits of alternative transport in Adelaide, South Australia. Environment International, 74, 281–290. Xiao, C., Goryakin, Y., & Cecchini, M. (2019). Physical activity levels and new public transit: A systematic review and meta-analysis. American Journal of Preventive Medicine, 56(3), 464–473. Xu, F., Connell McCluskey, C.,  & Cressman, R. (2013). Spatial spread of an epidemic through public transportation systems with a hub, Mathematical Biosciences, 246(1), 164–175. doi:10.1016/j. mbs.2013.08.014 Zhang, N., Huang, H., Duarte, M., & Zhang, J. J. (2016). Dynamic population flow based risk analysis of infectious disease propagation in a metropolis. Environment International, 94, 369–379. https://dx.doi. org/10.1016/j.envint.2016.03.038 Zhang, N., Huang, H., Su, B., Ma, X., & Li, Y. (2018). A human behavior integrated hierarchical model of airborne disease transmission in a large city. Building and Environment, 127, 211–220.

288

21 DEMAND ESTIMATION FOR PUBLIC TRANSPORT NETWORK PLANNING Wenzhe Sun and Jan-Dirk Schmöcker

Introduction The classic public transport (PT) network planning process contains five main steps: (1) network route design, (2) frequency setting, (3) timetable development, (4) vehicle scheduling and (5) crew scheduling (Ceder & Wilson, 1986). Clearly, these steps are interdependent, and “loops” may be needed in the planning process. In particular, frequency setting and timetable development can often be integrated. Table 21.1 describes the framework of Ceder (2007) but in a simplified form. In Ceder (2007), multidirected interactions between the levels are illustrated and discussed. In this chapter, the aim is to emphasise that demand estimates are crucial for the first, and most influential, stages of the planning process. The latter two stages, vehicle and crew scheduling, mainly determine a balance between the demand-driven timetable and limited vehicle and human resources. The different ways to obtain demand estimates are the main focus of this chapter. Classic tools such as surveys are still important, but the availability of massive passive PT data sources such as automatic passenger counter (APC) and automatic fare collection (AFC) data has triggered notable advances in deriving reliable demand estimates (Kurauchi & Schmöcker, 2016). As a consequence of this, however, the gap in the service quality provided by the “data-rich” and the “data-poor” practitioner is enlarged due to fundamental differences in the planning process. A “big data”-driven planning process can accommodate fluctuating user demand by efficiently making use of limited land, capital and human resources. It is necessary to be aware of this gap and to understand the potential in the datasets that are more common and cheaper to collect. Two stages for PT demand estimation are defined. The first stage serves for the planning activities of initial network construction and initial timetable development. Since service performance data are not available at this stage, demand has to be inferred from population and traffic flows as well as comparative studies. If a service improvement for an existing service is required, actual demand data or demand estimates from the service in use can be obtained. In this chapter, both stages are discussed, but emphasis is placed on the second case, that is the demand estimation for an existing network, as the majority of the most recent research has focused on this stage.

289

Wenzhe Sun and Jan-Dirk Schmöcker Table 21.1 Public transport network planning process – adapted from Ceder (2007) Independent input

Planning activity

Output

Land use characteristics; Authority constraints; Demand by time of day and day of week Service standards; Time for first and last trips; Running times; Comparison measures Deadhead times; Recovery times; Schedule constraints; Cost elements Crew list; Crew work rules, rotation rules, constraints

Level A Network design

Interchanges and terminals; Fixed routes and stops

Level B Timetable development

Alternative setting of frequencies and headways; Selected public timetable

Level C Vehicle scheduling

Minimum fleet size; Vehicle schedules

Level D Crew scheduling

Crew schedules and duty rosters

The remainder of this chapter is as follows. The following section briefly reviews demand modelling based on four-step modelling, activity-based models and newer ideas that can contribute to demand estimation for new public transport services. This is followed by an introduction to the different levels of detail at which demand information can be obtained, as well as their role in the planning process. The chapter then focuses on the demand estimation methods with data from an existing PT service. For this, methods using emerging massive passive PT data to estimate route-level demand in terms of stop boarding/alighting flows and “leg-Origin Destination (leg-OD)” are discussed. Leg-OD refers to a combination of boarding and alighting points for a line. The discussion includes straightforward methods based on observed passenger flows (on-board surveys, APC and AFC data) and a recently proposed idea of using bus automatic vehicle location (AVL) data. The feasibility of using AVL data to estimate demand offers great potential benefits to the “data-poor” practitioners. The penultimate section provides methods estimating network-level demand in terms of origin-interchange-destination flows, referred to as “Journey-OD”. Finally, the remaining issues and future research directions are discussed.

Demand estimation for new services If a new service is planned, no actual demand data in any form are available, but it is the latent and induced demand that need to be estimated. Latent demand describes those attracted to the service from other modes and induced demand the additional new demand generated by a better transport service. Classic transport planning literature uses four-step models consisting of trip generation, trip distribution, mode choice and assignment to provide a basic estimation of demand (Ortúzar & Willumsen, 2011). For estimating latent demand, the existing traffic OD matrix of a city or region might be taken and then, assuming a specific public transport network design, discrete choice models can be applied to obtain modal splits to obtain public transport journey OD flows. Assignment models can then be subsequently applied to obtain leg-OD flows, expected boarding, alighting and on-board flows. A number of problems can arise, such as dealing with 290

Demand estimation for network planning

dynamics during a day and, over longer time periods, a consideration of user habits, as well as effects such as crowding and consideration of detailed timetable design that affect the attractiveness of the service which will require iteration between mode choice and assignment. There is a large body of literature; among others, Gentile and Noekel (2016) provide extensive coverage of the topic. Fonzone et al. (2016) furthermore contribute to the understanding of the circular effect of intelligent transport systems on demand patterns, discussing how information influences demand patterns during the different journey stages. Despite all its shortcomings, it is fair to say that various modified versions of the four-step models are still commonly used by practitioners to obtain an initial, rough demand estimate. An increasingly popular improvement to the approach is the use of agent-based models. Compared to the “macroscopic” four-step model, here the population is “microscopically” modelled for a range of supply-side changes. Such models include “activity-based” approaches including the four steps from trip generation to assignment. Thus, not only latent demand but also induced demand can be better reflected. Software solutions such as MATSim are currently used to model large city-wide scenarios based on public transport data, as in Fourie et al. (2016), but also to model demand for totally new systems, as seen in Bischoff and Maciejewski (2016). Finally, it can be noted that “big non-PT data” solutions are increasingly being used for demand estimation of new public transport services. For example, Bonnel et al. (2018) have been using mobile phone data to estimate OD matrices. Further, comparing the target city to a wide range of other cities and their socioeconomic, geographic, cultural and political situations can help to obtain initial overall demand estimates. Related to this is the potential of using map data to estimate latent and induced demand. Bast et al. (2015) have shown that very good population estimates can be achieved with map data. These and the distribution of “points of interest” in a city can then be used to estimate trip generation and trip distribution. Koca (2020) has been using OpenStreetMap data in combination with machine learning approaches to predict the OD matrix of taxi trip data for New York. Whether a similar approach to “learning” will also work to predict the demand for public transport needs further investigation. Compared to taxi trips that are less constrained by the infrastructure, public transport demand requires far more specifications in terms of network design and service characteristics. The next section focuses on the cases where a service is already existing so that (limited) data are available. The demand for PT planning is classified at the three aforementioned levels: stop flows, leg-OD flows and journey-OD flows.

Levels of demand information and their roles in planning Stop alighting/boarding flows and passenger loads Stop alighting/boarding flows and passenger loads play a critical role in timetable development. This information can be collected in a variety of ways, including (automatic or manual) counters at the boarding and alighting door, cameras on-board or at the vehicle stop or smart card readers. Passenger loads, in terms of on-board passengers as a vehicle arrives at a stop, can then be reliably derived by processing the data streams of passenger boarding and alighting flows at a low cost relative to direct observation. These three datasets are termed the “stop-level demand profile”. A simple time-tabling strategy is to schedule services with equally spaced headways (see also Chapter 32). In this case, headways can be determined based on the overall demand during the operational period of concern. An advanced alternative strategy is to equalise passenger loads per vehicle, considering the demand dynamics. The criteria for load balancing can be 291

Wenzhe Sun and Jan-Dirk Schmöcker

the average load over the route, the load at a predetermined checkpoint or the maximum load. Different timetables are usually assigned to holiday periods due to their specific demand profile. Assuming that boarding demand follows a uniform distribution, the demand can be interpreted as the arrival rate per minute or per hour in the interval and the calculated expected passenger waiting time given a timetable plan. Knowing the stop-level demand also supports the operator in inserting appropriate slack time into the timetable. The numbers of boarding and alighting passengers collectively illustrate the “busyness” of the PT stop. When large passenger flows frequently occur at a specific PT stop in an operating interval such as the morning peak, the delays due to the unplanned long dwell time are not random but predictable. They can therefore be removed by introducing slack time into the timetable in the planning stage. Ideally, the stop-level demand profile should not only include mean passenger numbers but also information on the expected variance. Finally, stop-level demand can help to assess and modify timetables based on passengerfocused criteria beyond unweighted service regularity aspects. For example, Chen et al. (2009) measure the passenger-weighted reliability for a bus transit network, using a punctuality index, a headway deviation index and a headway evenness index. The former is based on the route and the latter two on a stop. Watanabe et al. (2017) show the trade-off between relaxed and tightened schedules considering the demand at each stop.

At route level (leg-Origin Destination) Leg-OD captures the boarding and alighting stop of each PT ride and aggregates the rides by OD-pair for each line. Dominant OD-pairs can be observed. This information enables the operator to adjust capacity according to the expected route-section demand, by, for example, serving certain route sections during the day more frequently. It further allows for efficient short-turn strategies during disruptions and the potential of introducing express strategies for important OD pairs. Short turn is to schedule a portion of the fleet to serve short cycles on the route segments faced with high demand, and express services operate on the entire route but skip several normal stops (Cortés et al., 2011). Cortés et al. (2011) and Tirachini et al. (2011) distinguish passenger boarding demand at each stop based on whether the alighting stop is located within the short-turn cycle. This connection between boarding and alighting stops is only available in leg-OD or journey-OD data.

At network level (journey-OD) Journey-OD connects a series of PT rides conducted by a specific user within the time and distance threshold. It is useful for transfer-oriented timetable development. Ceder et al. (2001) develop a mixed integer linear program to optimise the timetable with the objective of maximising the number of simultaneous bus arrivals at transfer nodes. Ibarra-Rojas and Rios-Solis (2012) maximise the transfer synchronisation and meanwhile reduce unnecessary bunching between the buses of different lines. They propose an integer programming method extending the work of Ceder et al. (2001) and enhance the computational efficiency by eliminating a series of decision variables. In light of the network-level demand information provided by the journey-OD matrices, practitioners can furthermore optimise the bus network design, route configuration and stop location, such as increasing or decreasing overlapping route segments, combining the bus stops to reduce walking time for transfer passengers. Automatic fare collection technologies have made it possible to filter and chain the trips of users throughout the multimodal PT network, including bus, metro, railway and ferry. With advances in the integration 292

Demand estimation for network planning

of fare collection and the emergence of smart mobility platforms, it should be possible to create journey itineraries for individuals in an extended transport network with new feeder modes such as bicycle and car sharing. Thus, knowledge on PT users’ (true) origin and destination can support evolving PT network design even considering competition and cooperation between traditional PT modes and emerging feeder modes.

Time-varying mean and distribution of passenger flows For both stop and OD flows, variations over time are an important issue. Passive PT data make it easier for practitioners to collect the data from various operating intervals. Figure 21.1 illustrates the passenger demand profile in the morning and evening peak periods obtained from smartcard data from a bus line in Shizuoka, Japan. Significant differences between the demand patterns can be observed in these two periods. Bie et al. (2015) employ bus dwell time as the proxy for passenger alighting/boarding flows and develop a method to identify time-of-day operating intervals from AVL data for the bus operators. Planning is usually based on mean passenger flows or rates in an interval, though the knowledge of the distribution during a time interval can offer a fuller picture to the practitioner with respect to the degree and probability that the demand can be outside the expectation. Figure 21.1(c) reveals that in the morning peak, the load can exceed 50 passengers between Stops 17 and 24, which indicates severe crowding on some buses. However, the mean loads in this segment remain between 30 and 37 passengers, showing a normal on-board condition. Taking Stop 20 (the mean maximum load point) as the checkpoint, it can be observed that the passenger load exceeds 50 for six times in 30 sampled runs during the morning peak hour. This high probability that passenger load can deviate by more than 50% upward from the mean deserves the operator’s attention. For a bus route, passenger flows can be classified by run if the run ID is recorded when the passenger taps in. However, a passenger assignment model is required when it comes to a metro or railway route. The APC and AFC data for railway/metro route only record the time point when the passenger enters and leaves the PT station; thus, extra effort is needed to relate an individual passenger to a specific service. Zhu et al. (2017) develop a probabilistic approach for this. They connect AFC data and AVL data and attach the boarding probability to a series of candidate train vehicles for a single passenger (see also Chapters 33, 34, 35, 36). They presume that the time points of passengers entering/leaving the system and the arrival times of train

Figure 21.1a Observed boarding flows

293

Wenzhe Sun and Jan-Dirk Schmöcker

Figure 21.1b Observed alighting flows

Figure 21.1c Observed passenger loads

vehicles at origin/destination stations are given. With this information and the walking time from entrance to platform at the origin station and from platform to exit at the destination station, as well as the capacity constraints, they calculate the boarding probability. In this way, the flows and the leg-OD matrix for each vehicle are replicated.

Leg-OD estimation Following on from the classification introduced in the previous section, this section details approaches to estimating leg-OD flows from different PT data before discussing the chaining techniques to estimate journey-OD flows from AFC data or by combing several data sources in the next section. First, a review of the methods regarding estimating leg-OD matrices for a bus route from AFC or on-board survey data is shown, combining APC with survey data, as well as APC data by itself. Often, there is missing data, and so some discussion is included on the solutions used to address the missing origins or destinations in incomplete AFC data. Second, for practitioners who have difficulty in collecting data that directly observe passenger flows such as AFC and APC data, a recent method to infer expected OD flows using bus AVL data is provided. 294

Demand estimation for network planning

Estimation with survey, APC and AFC data Figure 21.2 illustrates the passenger flows for a general bus route. Consider that there are K buses in total in a time-of-day operating period for a bus route having N stops. For each bus run k in this period, let Bi,k, Ai,k and Oi,k denote the boarding flow, alighting flow and passenger load at stop i, with i and j indicating the origin and destination stops, leading to Qi,j,k, which represents specific OD flows for each bus run. The OD matrix of a general bus route is provided in Table 21.2, omitting the index of a bus run. There are N(N – 1)/2 non-zero elements in the leg-OD matrix. Boarding flows, alighting flow and passenger loads given observed OD flows are obtained using Equations (21.1)–(21.3). Note that BN and A1 are zero, as no passenger would board at the last stop or alight at the first stop. The passenger load is defined as the number of on-board passengers at the arrival, not the departure, of the bus; O1 is therefore also zero. Bi ,k =

N

∑Q

i , j,k

i = 1, 2,..., N −1

Equation 21.1

j =i +1 i−1

Ai ,k = ∑ Q j ,i,k i = 2, 3,...N

Equation 21.2

j=1

Oi ,k = Oi−1,k + Bi−1,k − Ai−1,k

Figure 21.2

i = 2, 3,...N

Equation 21.3

General illustration of passenger flows on a transit route as used similarly by e.g. Li (2009), Hazelton (2010)

Table 21.2 OD matrix for a transit route (omitting the bus run index k) Origin

Destination 1

1 2 i N-1 N Aggregates

AN ≡ 0

Aggregates 2

i

N-1

N

Q1,2

Q1,i Q2,i

Q1,N-1 Q2,N-1 Qi,N-1

Q1,N Q2,N Qi,N QN-1,N

A2

Ai

AN-1

AN

295

B1 B2 Bi BN-1

BN ≡ 0

Wenzhe Sun and Jan-Dirk Schmöcker

Hazelton (2001) analyzes the subtle distinction between replicating the OD trips and estimating mean OD trip rates (expectation). The discrepancy is demonstrated in the OD estimation for traffic flows. OD trips fluctuate in each homogeneous observational period, for example, different hours in the whole morning peak or the same hour in different days. Mean OD trips or OD trip rates can be employed to characterise these homogeneous periods in order to significantly reduce the estimation complexity. In the estimation for PT routes, the reconstruction problem, with the objective to accurately estimate the OD flows for each bus run in a time-of-day interval, should be distinguished from the estimation problem, with the objective to infer the expected OD flows in that period. For each time-of-day operating period, mean flows or arrival rates per minute (per hour) can characterise the demand pattern. The fluctuation in the flows of the bus runs in a same interval is usually the result of headway fluctuations. The OD estimation can therefore be paramaterised if headway data can be collected. Generally speaking, there are two approaches. The first one is intuitive. It models the OD flows as OD-pair arrival rates ai,j multiplied with the associated headway, as shown in Equation (21.4). Qi , j ,k = ai , j ∆i ,k

Equation 21.4

It is thus possible to directly estimate the rates if the leg-OD matrix is available for adequate bus runs, for example, collecting AFC data or large-scale on-board survey data. In an on-board survey, a leg-OD matrix sample can be constructed by asking the passengers of one bus trip to provide their boarding and alighting stop, but there might be “non-structural zeros” in the matrix sample (Ben-Akiva et al., 1985; McCord et al., 2010; Ji et al., 2015). These are the zero OD-flows that are not captured by the random survey, and they may still exist when aggregating surveys from different runs to create a less biased matrix unless the sample size is very large. Thus, the passenger OD rates drawn from these samples are likely to be biased. On the other hand, APC data provide passenger observations at a much finer scale, with the shortcoming that the passenger flows captured by the counter at boarding and alighting doors are not linked. Therefore, surveyed OD matrices can be taken as the base matrix and balancing techniques are employed to obtain the OD rates at a finer-scale. To make use of the boarding and alighting counts, which are reliable control totals, a second approach to model the OD flows is introduced. It assumes stop-based boarding rates and alighting probabilities. Let bi denote the arrival rate for stop i; ci the alighting probability, defined as the proportion of the alighting passengers over the number of on-board passengers at stop i and ai¢, j the conditional alighting probability at stop j given passenger boarding the bus at stop i. The flows can then be rewritten as in Equations (21.5) and (21.6). If A, B, O and Δ are observed, b and c can then be calibrated by statistical analysis. This means that, given boarding and alighting counts, the estimation problem is transformed into the estimation of the conditional alighting probability matrix a¢ as in (21.7) constrained by alighting proportion c. The right-hand side of (21.7) is then estimated with conditions (21.1) to (21.3). Bi ,k = bi∆i ,k

Equation 21.5

Ai ,k = c iOi ,k

Equation 21.6

Qi , j ,k = ai′, jbi∆i ,k

Equation 21.7

296

Demand estimation for network planning

Ben-Akiva et al. (1985), McCord et al. (2010) and Mishalani et al. (2011) estimate probability matrix a ¢ instead of arrival rate or volume matrix from unconnected boarding-alighting counts and base OD matrices derived from on-board survey data, using iterative proportional fitting (IPF). A problem remains, however, with the previously mentioned non-structural zeros. BenAkiva et al. (1985) point out that the non-structural zero entry to the surveyed matrix can be explained by a non-zero probability, which captures passenger travel patterns more correctly. McCord et  al. (2010) introduce a null matrix assuming that a boarding passenger has equal probability to alight at all the downstream stops, and the starting point of the probability matrix is then a matrix weighted by a null and surveyed matrix whose weights are dependent on the survey size. This also addresses non-structural zeros. Mishalani et al. (2011) discuss how increasing the sample size of the on-board OD survey improves the estimation performance; in their case, the sample size refers not to the number of sampled runs but the number of passengers that report the leg-trip they made on this specific bus route. Under a data environment with large quantities of passenger counts provided by APC data and base OD matrix, Ji (2011) proposes a heuristic expectation maximisation (HEM) approach and comprehensively compares it with IPF, expectation maximisation (EM) and conditional maximisation (CM) algorithms. Ji et al. (2017) use stop-based farebox data as the proxy of boarding counts and the base matrix collected by Wi-Fi sensors. The problem becomes increasingly difficult if the base matrix is totally unavailable. Consider a case when only APC data are available. As illustrated in Table 21.2, the number of unknown OD flows is N(N – 1)/2, while the number of observations and linear equations determined for the system is only 2N; the problem therefore becomes underspecified if N > 5. Several studies break down this problem by estimating the destination distribution for the boarding passengers at each stop. A naïve assumption is that the boarding passenger alights at each downstream stop with equal probability. Li and Cassidy (2007) assume a conditional alighting probability matrix and categorise all the bus stops into major and minor stops according to land use information and APC data. Li (2009) models the conditional alighting probability matrix with Markov chains. The probability of a passenger’s alighting at a specific stop is considered dependent only on whether s/he is on-board at the previous stop. Hazelton (2010) suggests that this is unrealistic and proposes a Markov chain Monte Carlo (MCMC) method using the transition probabilities derived by Li (2009)’s model as the initial “proposal distribution” in metropolis sampling. Ji et al. (2015) further find that the HEM method using only APC data can match the performance by IPF using APC data and a base matrix constructed by relatively few passengers. They confirm the performance match when more than 500 runs are collected on a bus route with 18 stops, 153 feasible OD pairs and 100 passengers being surveyed. Ji (2011) is recommended as additional reading material, as it comprehensively summarises the (up to then) existing leg-OD estimation methods considering their computational differences and similarities.

Inference of boarding and alighting stop with automatic fare collection data All of the previous research becomes obsolete if complete AFC data that contain both tap-in and tap-out are recorded. If records are incomplete, with APC data also available, it is possible to obtain the percentage of smart card users among all users and scale up demand accordingly by inversely applying the calibrated percentage as the expansion factor. There may be a need to correct for biases, though, given that regular users are more likely on certain OD pairs and are also more likely to possess smart cards. Accordingly, the percentage of smart card versus

297

Wenzhe Sun and Jan-Dirk Schmöcker

non-smart cards users may vary in different time-of-day periods, on different routes, and at different stops of a route due to the connectivity of the stop. The data regarding boarding or alighting stops could be missing for a couple of reasons. First, the completeness of data is highly dependent on the fare collection policy and transfer policy, as only one record, either tap-on or tap-off, is needed if there is a flat-fare system. Among distance and zonal fare, structures that charge according to the actual distance travelled (or zones traversed) and those charging for ODs independent of the route need to be distinguished. For the latter fare structure, the AFC records usually do not entail records on transfer points. This means that, again, leg-OD matrices are not straightforwardly available from the data. In that case, one can specify the path by combining AVL data to analyze the time interval between passenger arriving time to the stop, arrival and departure time of transit vehicle and passenger leaving time (see also Chapters 33, 34, 35, 36). For flat-fare PT networks, hence, the inference of boarding or alighting stops is required to complete the leg-OD matrix. Several studies propose methods by creating a database in which each passenger made at least two trips in one day (Barry et al., 2002; Trépanier et al., 2007; Munizaga & Palma, 2012). The missing information is inferred based on onward journeys as well as other day-to-day patterns, as also discussed in some more detail in the “journey-OD” section.

Estimation based on automatic vehicle location data Sun (2020) considers the problem from the perspective of “data-poor” practitioners without access to AFC and APC data and provides a new approach using bus AVL data as the main data source to infer passenger flows. AVL data are collected by an in-vehicle chip that keeps recording the spatial-temporal coordinates locally or uploads them to the server at a predetermined frequency, for example, every 10 seconds. No privacy issue relates to this data, and it is usually managed by the bus operators themselves or through their data providers. If the data collection frequency is high enough, or if it includes “event data” such as door opening, door closure and stand-still at a bus stop, then bus AVL data contain a rich detail level regarding the time components in the bus trips, from which passenger activity times (boarding and alighting times) can be extracted. With the help of a small survey to obtain the information on individual boarding/ alighting time, the connection between passenger flows and passenger activity times can be captured. However, it is in the form of a series of underspecified formulations; that is, the unique boarding and alighting number of passengers is not available. Figure 21.3 shows the model framework to estimate OD flows. The demand for OD-pair specific arrival rates are the estimated parameters. Once these are known, stop-level flows are also derived. ai,j is assumed to follow a modified gravity model, as in Equation (21.7): ai , j =

pig p aj ( j − i) +

a j −i

i = 1, 2,..., N − 1; j = 2, 3,..., N ;i < j

Equation 21.7

where pig denotes the generation power of stop i, and p aj denotes the attraction power of stop j. The first term in the denominator is the travel distance, and the second term is to acknowledge that a passenger is not likely to make a short bus trip travelling only one or two stops. The two vectors p g and pa , each of size N – 1, and parameter α substitute the arrival rate ai,j as the new unknown parameters. The number of unknown parameters to estimate is hence reduced from N(N – 1)/2 to 2(N – 1)+1. Sun (2020) turns to Bayesian inference, which has been shown 298

Demand estimation for network planning

Figure 21.3 A framework for OD estimation using bus AVL data. Items in grey boxes are the observations and input parameters; items in white boxes are the unknown and derived parameters

to be powerful in inferring OD trips given limited observations or observations with uncertainty (Maher, 1983; Hazelton, 2001, 2008, 2010). A Hamiltonian MCMC algorithm is then employed to estimate the unknown parameters. Results of this AVL-based approach are shown in Figure 21.4. Estimation results for a stop-level demand profile are illustrated in (a)–(c) in the form of the mean and 95% confidence interval, and the estimated versus observed OD flows are displayed in (d). The 299

Wenzhe Sun and Jan-Dirk Schmöcker

Figure 21.4a Boarding flows (observed and estimated from AVL data)

Figure 21.4b Alighting flows (observed and estimated from AVL data)

Figure 21.4c Passenger loads (observed and estimated from AVL data)

300

Demand estimation for network planning

Figure 21.4d OD flows (observed and estimated from AVL data)

distribution of the observations presents the actual demand fluctuation of the bus runs dispatched during a time-of-day operating interval, 7 am–8 am in this case. The distribution of estimation results, however, are the possible values for mean passenger flows given the distribution of mean passenger rates drawn by the MCMC algorithm and the observed headways. Therefore, the variation in estimated passenger flows is less pronounced than those for the observed passenger flows. Nevertheless, the mean estimates appear reliable and useful for service planning.

Journey-OD estimation Finally, the focus is on deriving complete journey itineraries across the multi-modal PT network for individual passengers in an observational period. If leg-ODs are available, estimating the journey-OD means to estimate transfer probabilities for all possible lines for alighting passengers. Similar to the discussion in the previous section, correlation between boarding and alighting flows at a stop and taking time differences between arriving and leaving services can be taken into account to “learn” transfer probabilities with Bayesian inference and other approaches. A different problem formulation arises from incomplete AFC data with tap-in records only. In that case, “panel data” of users are available in the form of subsequent boarding points. The objective then is to connect these data points to obtain complete user trajectories. Consider a multi-modal network consisting of railway lines under a distance-based fare structure and bus routes under flat fare structure so that tap-out information is not available for bus trips. For railonly trips, no inference is needed, but it is for the remaining four types of subtrips in a journey: single bus trips, trips with railway-to-bus or bus-to-rail transfers and trips with multiple bus legs. For a trip with a bus-to-railway transfer, the alighting point for the first leg is missing but can be easily inferred from the boarding point for the second leg. For the other three trip types, the origin of the next trip as information has to be leveraged so as to infer the alighting point of the previous trip. This might also include information derived from different days. A database containing one-day journey itineraries classified by passenger smart card ID can be feasibly constructed from AFC data. These basic principles of estimation are first proposed by Barry et al. (2002): 1) users will return after their activity to the alighting point of the previous trip; 2) in the same day, the destination of the last trip is identical to the origin of the first trip; 3) the unknown alighting location of a bus trip is the stop closest to the origin of the next trip. Trépanier et al. 301

Wenzhe Sun and Jan-Dirk Schmöcker

Figure 21.5 A simplified multimodal network, as also used in Gordon et al. (2018)

(2007) extend the database by constructing multiple-day itineraries, identifying mirrored journeys in a longer timespan. Zhao et al. (2007) apply maximum walking distances of 400 m or 5 min to detect the alighting stop given the next boarding stop in addition to the basic assumptions. Munizaga and Palma (2012) propose an algorithm minimising the total travel time of a journey, in particular the transfer time, filtering out the options not likely taken by transfer users. Alighting stop detection based on trip symmetry can be time consuming and data demanding. Alternatively, Gordon et al. (2018) relax the data requirement and attempt to build journey itineraries with the data in an observational period shorter than one day, such as a time-of-day period. As the boarding stop of the next leg is not available for the destination inference, they reduce the number of OD pairs in a PT network. They attach two action labels to each station (entry or exit at the station) and generalise for bus nodes by attaching two direction labels to a whole route (northbound or southbound). The journey OD flows are therefore captured and classified by all possible combinations of itineraries. A checklist-based method identifies each unique itinerary, and the flows are aggregated by itinerary. For the example network shown in Figure 21.5 there are 38 feasible itineraries, such as “Bus Route 3 northbound, Station C entry, Station B exit”. The base counts for these itineraries are directly observed from AFC data or inferred using the basic principles introduced previously. Control totals are collected at the railway station gates and bus fareboxes, which are used to obtain expansion factors. The journey expansion algorithm is based on IPF.

Conclusion The aim of this chapter has been to provide a systematic introduction on the demand estimation approaches serving PT network planning. The new service construction problem has been distinguished from the problem of estimating demand for existing services. For new services, demand estimation is mainly based on population and land use data as well as flows observed for existing transport modes. Information about existing “potential” demand can be estimated from an increasingly broad range of data. The role of map data has been discussed, but other data sources such as mobile phone records, Wi-Fi records, sales data or information that can be inferred from social media have not been discussed in detail. By estimating demand for existing services, the gaps between demand and supply can be identified for further improvement purposes. In this stage, actual passenger demand patterns are available in the form of massive passive PT data, such as AFC and APC data. Three basic levels 302

Demand estimation for network planning

have been defined, namely the stop-level demand profile for a transit route, leg-OD matrix for a transit route and journey-OD matrix for a PT network. The chapter then discussed approaches for each of these levels. There is a significant body of literature using AFC and APC data. It is pointed out that AVL data are also a potential information source if the higher-quality direct passenger flow data are not available. This has been the recent research topic of the authors, and an example is shown. As a practical note, low-cost actions can significantly enhance the quality of AVL data and further improve the quality of demand estimation. One example is automatically recording the stop and time information for each door opening and closing activity. This can help to extract precise passenger activity time and make it possible to estimate passenger flow, even with low-frequency bus AVL data. More generally, it is hoped that the chapter can “broaden the mind” of operators and researchers, showing how different (incomplete) data can be used and combined to obtain better information for service planning. The abundance of passive data in combination with machine-learning methods offers large potential that is starting to be “standard” via integration in mainstream planning software. A range of challenges still remain with respect to demand forecasting for network changes or if disruptions occur where a gap appears to remain between the approaches discussed in this chapter. The vast majority of studies using passive big PT data concern the reconstruction of demand. More recently, four-step and agent-based modelling has started to benefit from the findings of big data studies, but comprehensive frameworks as to how AVL, APC and AFC data can be used to derive sensitivities to key transport policy variables are still in their early stages and require further research. This chapter was largely written before the impact of the COVID-19 pandemic was appreciated. The methodologies discussed in this chapter are clearly not directly affected by the current crisis, but the implications of the pandemic around the globe, its lockdown and the emergence from lockdown are beginning to be realised. First, the avoidance of contact in general and in public transport in particular will encourage customers and operators to utilise cashless payment systems even more. For demand estimation, this is advantageous, as often only the electronically collected data are used for analysis. Furthermore, the increase in (anonymous) tracking data as well as the increased utilisation of pre-booking a seat in public transport services might be valuable data for activity estimation as well as demand prediction. Second, changes in passenger behaviour following the pandemic will lead to a need to re-calibrate demand estimation such as sensitivity in choice with respect to other alternatives. In this chapter, this has particular relevance to possible changes in the relationship between dwell time and demand. More social distancing might mean longer boarding time per passenger. For the demand estimation approach with dwell time data presented in this chapter, this would spell good news, as the effect of each passenger on the observed dwell time would be pronounced. Furthermore, the crowding-averse tendency simplifies the dwell time function in the estimation, as the boarding passengers are more willing to wait until the alighting process ends, and friction due to crowding might less often occur. As a result, post-COVID-19 models might be easier to calibrate.

References Barry, J. J., Newhouser, R., Rahbee, A., & Sayeda, S. (2002). Origin and destination estimation in New York City with automated fare system data. Transportation Research Record, 1817, 183–187. Bast, H., Storandt, S., & Weidner, S. (2015, November). Fine-grained population estimation. Proceedings of the 23rd SIGSPATIAL International Conference on Advances in Geographic Information Systems, pp. 1–10.

303

Wenzhe Sun and Jan-Dirk Schmöcker Ben-Akiva, M., Macke, P. P., & Hsu, P. S. (1985). Alternative methods to estimate route-level trip tables and expand on-board surveys. Transportation Research Record, 1037, 1–11. Bie, Y., Gong, X., & Liu, Z. (2015). Time of day intervals partition for bus schedule using GPS data. Transportation Research Part C: Emerging Technologies, 60, 443–456. Bischoff, J., & Maciejewski, M. (2016). Simulation of city-wide replacement of private cars with autonomous taxis in Berlin. Procedia Computer Science, 83, 237–244. Bonnel, P., Fekih, M., & Smoreda, Z. (2018). Origin-destination estimation using mobile network probe data. Transportation Research Procedia, 32, 69–81. Ceder, A. (2007). Public transit planning and operation: Theory, modeling and practice. Elsevier, Butterworth-Heinemann. Ceder, A., Golany, B., & Tal, O. (2001). Creating bus timetables with maximal synchronization. Transportation Research Part A: Policy and Practice, 35(10), 913–928. Ceder, A.,  & Wilson, N. H. (1986). Bus network design. Transportation Research Part B: Methodological, 20(4), 331–344. Chen, X., Yu, L., Zhang, Y., & Guo, J. (2009). Analyzing urban bus service reliability at the stop, route, and network levels. Transportation Research Part A: Policy and Practice, 43(8), 722–734. Cortés, C. E., Jara-Díaz, S.,  & Tirachini, A. (2011). Integrating short turning and deadheading in the optimization of transit services. Transportation Research Part A: Policy and Practice, 45(5), 419–434. Fonzone, A., Schmöcker, J.-D., & Viti, F. (2016). New services, new travelers and new models? Directions to pioneer public transport models in the era of big data. Journal of Intelligent Transportation Systems, 20(4), 311–315. Fourie, P., Erath, A., Ordonez, S. A., Chakirov, A., & Axhausen, K. W. (2016). Using smart card data for agent-based transport simulation. In F. Kurauchi & J. D. Schmöcker (Eds.), Public transport planning with smart card data (pp. 130–160). Springer (CSC Press). Gentile, G., & Noekel, K. (2016). Modelling public transport passenger flows in the era of intelligent transportation systems. Springer. Gordon, J. B., Koutsopoulos, H. N.,  & Wilson, N. H. (2018). Estimation of population origin–interchange–destination flows on multimodal transit networks. Transportation Research Part C: Emerging Technologies, 90, 350–365. Hazelton, M. L. (2001). Inference for origin–destination matrices: Estimation, prediction and reconstruction. Transportation Research Part B: Methodological, 35(7), 667–676. Hazelton, M. L. (2008). Statistical inference for time varying origin–destination matrices. Transportation Research Part B: Methodological, 42(6), 542–552. Hazelton, M. L. (2010). Statistical inference for transit system origin–destination matrices. Technometrics, 52(2), 221–230. Ibarra-Rojas, O. J., & Rios-Solis, Y. A. (2012). Synchronization of bus timetabling. Transportation Research Part B: Methodological, 46(5), 599–614. Ji, Y. (2011). Distribution-based approach to take advantage of automatic passenger counter data in estimating period route-level transit passenger origin–destination flows: Methodology development, numerical analyses and empirical investigations [Unpublished doctoral dissertation, The Ohio State University]. Ji, Y., Mishalani, R. G.,  & McCord, M. R. (2015). Transit passenger origin–destination flow estimation: Efficiently combining onboard survey and large automatic passenger count datasets. Transportation Research Part C: Emerging Technologies, 58, 178–192. Ji, Y., Zhao, J., Zhang, Z., & Du, Y. (2017). Estimating bus loads and OD flows using location-stamped farebox and Wi-Fi signal data. Journal of Advanced Transportation, 2017, 10. Article ID 6374858. https:// doi.org/10.1155/2017/6374858 Koca, D. (2020). OD matrix estimation by deep learning using map [Unpublished bachelor thesis, Kyoto University]. Retrieved from http://trans.kuciv.kyoto-u.ac.jp/its/pdf/2019_thesis_abst/thesis_abst_Danyel.pdf Kurauchi, F., & Schmöcker, J. D. (2016). Public transport planning with smart card data. CRC Press. Li, B. (2009). Markov models for Bayesian analysis about transit route origin–destination matrices. Transportation Research Part B: Methodological, 43(3), 301–310. Li, Y., & Cassidy, M. J. (2007). A generalized and efficient algorithm for estimating transit route ODs from passenger counts. Transportation Research Part B: Methodological, 41(1), 114–125. Maher, M. J. (1983). Inferences on trip matrices from observations on link volumes: A Bayesian statistical approach. Transportation Research Part B: Methodological, 17(6), 435–447. McCord, M. R., Mishalani, R. G., Goel, P., & Strohl, B. (2010). Iterative proportional fitting procedure to determine bus route passenger origin–destination flows. Transportation Research Record, 2145(1), 59–65.

304

Demand estimation for network planning Mishalani, R. G., Ji, Y., & McCord, M. R. (2011). Empirical evaluation of the effect of onboard survey sample size on transit bus route passenger OD flow matrix estimation using APC data. Transportation Research Record, 2246, 64–73. Munizaga, M. A., & Palma, C. (2012). Estimation of a disaggregate multimodal public transport origin– destination matrix from passive smartcard data from Santiago, Chile. Transportation Research Part C: Emerging Technologies, 24, 9–18. Ortúzar, J., & Willumsen, L. G. (2011). Modelling transport (4th ed.). John Wiley & Sons. Sun, W. (2020). Bus bunching prediction and transit route demand estimation using automatic vehicle location data. Doctoral thesis, Kyoto University. Retrieved from https://repository.kulib.kyoto-u.ac.jp/dspace/bitstream/2433/253498/2/dkogk04737.pdf Tirachini, A., Cortés, C. E., & Jara-Díaz, S. R. (2011). Optimal design and benefits of a short turning strategy for a bus corridor. Transportation, 38(1), 169–189. Trépanier, M., Tranchant, N., & Chapleau, R. (2007). Individual trip destination estimation in a transit smart card automated fare collection system. Journal of Intelligent Transportation Systems, 11(1), 1–14. Watanabe, Y., Nakamura, T., Schmöcker, J.-D., Uno, N., & Iwamoto, T. (2017, October 16–19). Adjusting bus timetables considering observed delays and passenger numbers. Proceedings of IEEE ITSC 2017; 20th International Conference on Intelligent Transportation Systems. Zhao, J., Rahbee, A., & Wilson, N. H. (2007). Estimating a rail passenger trip origin–destination matrix using automatic data collection systems. Computer-Aided Civil and Infrastructure Engineering, 22(5), 376–387. Zhu, Y., Koutsopoulos, H. N.,  & Wilson, N. H. (2017). A  probabilistic passenger-to-train assignment model based on automated data. Transportation Research Part B: Methodological, 104, 522–542.

305

22 THE FIRST/LAST MILE CONNECTION TO PUBLIC TRANSPORT Christoffel Venter

Introduction Interest is growing in the first and last mile of the public transport trip as an important component of the overall quality experienced by a public transport user. This reflects a growing appreciation of the public transport trip as inherently multi-modal, one where the effort and time required to access public transport from the origin of the trip, and to travel between the egress stop or station and the final destination, are seen as critical links in the trip chain. The first and last mile (1LM) connection is often the slowest, most unsafe, or most physically demanding part of the trip – especially where it involves walking in many walk-unfriendly cities – making it the weakest link that prevents large-scale shifts towards public transport. Public transport operators and city authorities find it challenging to deal with the 1LM in a coherent and integrated way. The 1LM environment is complex, encompassing a variety of modes with widely varying characteristics. A diversity of role-players affect the quality of 1LM connections, including public and private transport operators, urban design, street infrastructure, and traffic safety professionals, raising the difficulty of co-ordinating between them. Individual public transport operators rarely have control over the 1LM portion of their service, making them dependent on other actors for delivering whole-trip service quality to passengers. Local authorities seldom focus on public transport in a holistic, multi-modal way. The lack of attention to 1LM connections is partly because of an absence of tools for measuring, monitoring, and managing the quality of the 1LM environment in the context of the public transport system. Scholars have tended to focus on individual modes – for instance, a large body of work has evolved from the health and transport planning literatures on the relationships between the built environment, walkability, and walking behaviour (see also Chapter 23). Although a substantial part of this deals with walk access to public transport, there are no general methodologies for bringing the various modes into the same framework in order to explicitly capture the multi-modal nature of the 1LM problem (Phillips & Guttenplan, 2003). In addition, not enough is known about the needs and behaviour of the prospective public transport user regarding the 1LM part of their trip. A better understanding of 1LM issues will be useful for transport modelling, infrastructure planning, urban design, and the health research communities (Clifton & Muhs, 2012). And it will also help focus the attention of planners and implementers on 1LM issues as in themselves critical to the success of public transport. 306

The first/last mile connection

This chapter aims to help fill this gap by briefly reviewing the literature relating to the first and last mile connection to public transport and presenting a framework for the integrated measurement of the quality of the 1LM environment. The review highlights factors that have been found to affect 1LM quality and major approaches to measuring these factors. The proposed integrated framework is briefly explained and illustrated using a case study from South Africa. The chapter concludes with research needs for the future.

The first and last mile of the public transport trip The origins of the first and last mile idea lie in the “last-mile” concept in telecommunications and supply-chain management, where the last mile from hub to destination is the most challenging and costly from a logistical point of view. In public transport, the vast majority of 1LM trips have historically been made by walking (Jiang et al., 2012), but the number of other modes and services offering 1LM connectivity is growing. Bicycling is rising worldwide and is increasingly used for multi-modal trips, supported by the rise in cycle storage and bikeshare schemes at public transport stations (Liu et al., 2012; Wang & Liu, 2013). Newer forms of non-motorised transport, including e-scooters, personal mobility devices, e-bikes, and e-rickshaws, are proliferating, and although technologies and market models are still evolving, there is evidence that they could significantly alter the public transport experience by offering cheaper, faster, and more convenient access and egress options over larger distances (Kumar & Roy, 2019; Lesh, 2013), especially as a part of shared use schemes. 1LM trips are also increasingly served by other public transport modes as transport networks gain greater multi-modal integration (see also Chapter 6). This is reinforced by the shift in public transport service concepts over the last century in many parts of the world, which have deemphasised traditional local public transport dating back to the street car, providing uniform, slow service while minimising walk distances to origins and destinations in favour of more commuteroriented services with higher speeds and wider station spacings such as urban rail and bus rapid transit (BRT). The limited walking distances of users prevent such services from serving a majority of residents and jobs directly, especially in medium-density regions (Cervero, 1998). This makes the 1LM trip dependent on other modes like local buses and taxis and, especially in developing countries, a range of “intermediate” modes like cycle or motorised rickshaws and motorcycle taxis (Kumar & Roy, 2019). In more car-oriented environments, the private vehicle plays an important role in serving the 1LM trip, especially to and from rail stations in suburban areas (Cervero, 2001). This is promoted by the provision of large parking lots at stations, which is often seen as a critical way of enlarging public transport catchments. However, the environmental costs of serving 1LM trips primarily with automobiles are significant; the emissions associated with the cold start and hot evaporative soak (which is not eliminated) may well negate the benefits of patronising rail in the first place (Cervero, 2001; Hoehne & Chester, 2017). Thus, one of the primary goals of Transit Oriented Development (TOD) around rail stations is to shift the 1LM mode from private cars to more sustainable modes. Advances in vehicle, communications, and computational technologies hold promise for radically changing the manner in which 1LM trips are served while carving out a much greater role for the private sector (Canales et al., 2017). Transportation network companies offering mobility-on-demand services are already serving significant numbers of 1LM trips (Shaheen & Chan, 2016), although there is concern that they may divert entire trips away from public transport (Clewlow & Mishra, 2017). Simulation studies of flexible on-demand 307

Christoffel Venter

shuttles (also referred to as microtransit) serving an access/egress function to higher-capacity bus routes concentrated on the heaviest volume corridors suggest that these services can be scaled to deliver significant benefits in terms of extending the catchment area of public transport to include those not traditionally part of a service area, reducing costs of feeder bus services and parking infrastructure, reducing overall travel times, and reducing private vehicle use (Shaheen  & Chan, 2016; Zellner et  al., 2016). The long-term environmental impacts are still unclear and will largely depend on the rate of adoption of cleaner vehicles and fuels within these shared fleets. Wang et al. (2019) and Boarnet et al. (2017) argue that improving 1LM access could improve equity by enhancing regionwide accessibility to employment among excluded populations; Zuo et al. (2020) argue the same for enhanced use of bicycles for 1LM trips to public transport. Other authors have cautioned against the exclusionary effects that new technologies might have against those with limited access to bank accounts and smartphones (Yan et al., 2019). The 1LM landscape in cities is likely to change significantly in coming years and to grow in complexity. The implications for users, public transport operators, and cities are hard to predict. Knowledge of what users want from the 1LM experience, and how it affects their travel behaviour, is already substantial, as the next section shows. Yet it will need to evolve to take account of this new landscape.

Factors affecting the quality of the first and last mile connection The literature shows that the quality of the 1LM environment is systematically affected by a number of factors and that these can be measured. A wide variety of different approaches have been adopted, depending on the mode, context, and purpose of the analysis.

Access distance Walk catchments are particularly important for urban public transport systems, as walking is the main access mode for most public transport trips (Jiang et al., 2012). A number of studies have examined the distances that people are willing to walk to stops and stations. Conventional wisdom in public transport planning holds that the primary walk catchment of public transport is 400 m (a quarter mile). This is based on a synthesis of studies between the 1960s and 1980s on how the propensity to walk decreases with distance to the stop, showing that 75% to 80% of passengers walk 400 m or less to a local stop (Kittleson & Associates Inc. et al., 2013) (see Figure 22.1). At an average walking speed of 5km/h, this translates to a 5-minute walk. In the last decade, a renewed research interest in walking as a mode has produced new evidence on walk distances to public transport. Van Soest et al. (2020) provide a recent overview of this evidence and find that there is much larger variation in people’s willingness to walk to public transport than previously assumed. Morency et al. (2011) found in Montreal that about half of passengers walked more than 400 m to public transport, while in the Bay Area, the average distance to rapid rail was 800 m (Agrawal et al., 2008). A major source of variation relates to the mode that is being accessed: walk distances to rapid transit like rail have been found to be about double that for local bus (Agrawal et al., 2008; Guerra et al., 2012; Morency et al., 2011). The same seems to apply to BRT, although much fewer studies have focused on BRT than rail as a main mode (Jiang et al., 2012). The reasons for this are not clear, but it is likely related to the generally higher levels of service and longer distances of rail and BRT trips, indicating that

308

The first/last mile connection

Figure 22.1 Range of walking distances to bus stops from empirical studies Source: Based on: Kittleson & Associates Inc. et al., 2013

perhaps acceptable walking distances should be investigated as proportions of people’s total trip length rather than as absolute distances. Mode issues might be confounded with other attributes of the public transport service and with built environment factors that vary systematically with the mode. For instance, the function of the station plays a role: research in Jinan, China, has shown that average walk distances to terminal BRT stations are more than double those to other stations (Jiang et  al., 2012). Mulley et al. (2018) found people willing to walk longer to more frequent services, everything else being equal. In terms of built environment factors, good micro-level walkability expands acceptable walking distances (Dill, 2003; Park et al., 2015), as do macro-level land use variables like density and land use mix (Cervero, 2001; Saelens & Handy, 2008). Most research on the relationships between the built environment and walking has focused on walking propensity rather than walking distances per se. This research is discussed in the next section. Journey-related factors also affect the willingness to walk, including trip purpose, time of day, and trip length (Krygsman et al., 2004; Park & Kang, 2008). Although most studies focus on the home side of the multi-modal trip, there is some evidence that public transport users are willing to walk longer distances on the activity part of their trip than on the home side (Krygsman et al., 2004). Of course, if other modes are used for the 1LM trip, access distances could dramatically change. For instance, bicycle access to commuter rail typically peaks at 1.6 to 2  km in the United States (Park & Kang, 2008) and 1.5 to 3.5 km in the Netherlands (Keijer & Rietveld, 2000), with slightly shorter distances to access bus stops (Hochmair, 2015). Vuchic (2017) suggests that catchment distances double for Kiss & Ride modes (which could include taxis and e-hailing) as compared to walking, and then doubles again for Park and Ride, but cautions that this can vary widely depending on local conditions. All of this suggests that flexible catchment area definitions are needed and that service areas around public transport stops and stations vary based on the service offered as well as attributes of the people and places served (El-Geneidy et al., 2014; Jiang et al., 2012).

309

Christoffel Venter

Walking environment There is a continuously growing base of research that consistently reveals associations between walking behaviour and the built environment (Ewing  & Cervero, 2001; Saelens  & Handy, 2008). The bulk of this literature has examined walking in general, while an increasing number of studies have looked specifically at walking access to public transport. Three main methodological thrusts have been the use of perception surveys to examine preferences and behaviours around walkability of the built environment (e.g. Adkins et al., 2012; Agrawal et al., 2008), the estimation of walking mode choice models to study the role of built environment and other factors on walking (e.g. Loutzenheiser, 1997; Meng et al., 2016; Park et al., 2015), and the development of walkability assessment tools for use in evaluation and design (e.g. Frank et al., 2010; Khisty, 1994). The intention is not to review this large body of literature here; the interested reader can see Talen and Koschinsky (2013) and Grasser et al. (2013) for recent reviews in the land use/transport and health literatures, respectively. Initial studies of the role of the built environment focused on macro-level land use variables and showed that factors such as land use mix and diversity, residential density, and the presence of retail influence people’s willingness to walk to public transport (Cervero, 2001). Micro-scale design factors of the pedestrian environment were found to be secondary to macroscopic ones (Saelens & Handy, 2008). However, multiple recent studies have found solid evidence that micro-scale factors are equally important; these include street design (e.g. street connectivity, setbacks, and parking), pedestrian infrastructure (presence and condition of sidewalks, safe street crossings, conflicts with driveways), and aesthetic factors (e.g. buffers, benches, and shading). Safety from crime has also been found to be important (Tilahun et al., 2016) – in some cases second only to walking distance (Agrawal et al., 2008). There is also a correlation with personal factors such as gender: Kim et al. (2007) find, for instance, that security is more important for women than for men. Walkability indices have become popular as ways of capturing some or all of these factors in a single composite measure and have been used for evaluation, defining benchmarks, and prioritising and budgeting remedial actions (Khisty, 1994). Typical measures such as the walkability index developed by Frank et al. (2010) use a multicriteria rating to weight the different environmental factors. Other indices such as Walkscore use walking distances to points of interest to construct a gravity-based measure of access to opportunities (Carr et al., 2010). Some indices convert their results into a pedestrian level of service (LOS) indicator (on a scale between A [best] and F [worst]) to be compatible with traffic engineering practice. Validation studies have found some walkability indices correlate well with physical activity levels, mode choices, and obesity levels (Glazier et al., 2012; Manaugh & El-Geneidy, 2011).

Personal factors Despite some conflicting findings, factors like income, age, gender, disability, race, household characteristics, and vehicle ownership have been found to affect walking behaviour (Manaugh & El-Geneidy, 2011; Van Soest et al., 2020). Earlier studies suggested these factors were dominant (e.g. Loutzenheiser, 1997), but as the accuracy with which environmental factors – particularly micro-scale design factors – are measured has improved, the dominance of personal over urban design and station area factors has subsided. Similar personal factors affect bicycling to public transport (Givoni  & Rietveld, 2007), although gender gaps appear to be more pronounced (men are more likely to cycle than women) (Wang & Liu, 2013). Cultural factors also play a role. There is evidence that commuters in developing countries such as China (Jiang et al., 2012) and South Africa (Behrens, 2004) are willing to tolerate longer 310

The first/last mile connection

walking distances to public transport, perhaps because of different expectations and generally lower values of time. Perceptions of walkability have also been shown to vary by socioeconomics; for instance, Gebel et al. (2009) showed that adults with low socioeconomic status, who are overweight, or walking with children are more likely to perceive a high-walkable neighbourhood as low-walkable.

Bicycle compatibility As bicycling has gained popularity worldwide, so the number of studies have grown that examine bicycling access to public transport. Doolittle and Porter (1994) provide an early overview of the state of practice on the integration of bicycles and public transport, while Replogle (1993) points out that in Europe and Japan, bicycling has long played a key role to access rail and, to a lesser extent, bus services. Apart from cultural and historic reasons, this is attributable to “the great attention that has been given by local governments to making streets pedestrian and bicycle friendly” (Replogle, 1993, p. 76). A variety of stress level and compatibility indices or LOS measures exist to measure the quality of the street environment for bicycle users. One of the more well-known is the bicycle compatibility index developed for the Federal Highway Administration (Harkey et al., 1998), which assesses factors like the presence and width of bike lanes or paved shoulders, parking lanes, the type of roadside development, and the volume and speed of adjacent traffic. This index is considered well validated and useful for general evaluation purposes (Phillips & Guttenplan, 2003) but does not focus specifically on 1LM issues. Similar land use variables appear to play a role in bicycle compatibility as in walkability, although European studies have found biking to rail stations to increase for more suburban environments (Martens, 2007). Also, biking appears less attractive as an egress mode (on the activity side of the trip) than an access mode (on the home side) (Givoni & Rietveld, 2007; Martens, 2007).

Travel time accessibility The concept of accessibility to destinations has been used in a number of applications to include the speed and connectivity of the 1LM mode in the evaluation of public transport services. This makes behavioural sense, as accessibility to jobs (and presumably to other destinations) explains about four-fifths of public transport ridership (Owen & Levinson, 2015). Accessibility is measured in terms of the number of opportunities reachable by feeder service (e.g. walk or feeder bus) from stations in various lengths of time (Carr et al., 2010; Cheng & Agrawal, 2010). More complex approaches use gravity-based accessibility models (Chandra et al., 2013) or take temporal variations in service frequency into account to identify areas with poor spatial/temporal coverage (Liu et al., 2018; Transport for London, 2015). These access measures go beyond mere 1LM distances or catchments in evaluating the quality of the 1LM service offer, highlighting the importance of the density and spatial distribution of origins and destinations within the public transport catchment as well as the speed and routing of the 1LM mode. However, they do not include the qualitative environmental factors of the 1LM that are also important in the assessment.

Multi-modal level of service techniques The Transit Capacity and Quality of Service Manual (Kittleson & Associates Inc. et al., 2013) describes a standardised method for combining various qualitative and quantitative aspects of a 311

Christoffel Venter

multi-modal public transport trip into a single measure, which is then expressed as a LOS rating. It is meant to evaluate the quality of service of each mode using a street, including public transport, from the user’s perspective. It includes factors such as the availability and quality of the public transport service itself (e.g. frequency, capacity, and reliability) but also includes some factors related to the pedestrian environment like street connectivity, grade, shelters, and pedestrian crossings. The factors are included as correction factors to a series of equations when calculating the public transport LOS score. They are calibrated on empirical studies of public transport user perceptions in the United States and as such are very difficult to validate or transfer to other parts of the world. In summary, the large and growing body of work examining factors affecting the quality of the 1LM environment has delivered significant insights into what people value and how to evaluate it. Because of the multitude of factors, large variations exist across and even within cities in both the quality of 1LM environments and in users’ expectations of them (El-Geneidy et al., 2014; Van Soest et al., 2020). These variations are likely very context and culture specific, related to pedestrian and cyclist behaviour, differences in perceptions, and different uses of street space (Jiang et al., 2012). This suggests that any efforts to measure the quality of the 1LM environment systematically must be flexible enough to reflect these differences yet rigorous enough to allow meaningful comparison across locales.

Towards a framework for measuring the quality of the first and last mile environment This section presents some thoughts towards establishing a general framework for measuring the quality of the 1LM environment, building on the access-to-public transport literature. The point of departure is that a more precise measure is needed that acknowledges the specific nature of the 1LM environment, specifically its multi-modality and diversity, but that is robust enough to enable consistent measurement and comparison across different contexts to be useful for evaluation, benchmarking, and prioritisation. The section starts with a brief discussion of methodological issues, which are illustrated using data of the 1LM environment of the Gautrain urban rail system in the Tshwane-Johannesburg region in South Africa. More details are provided in Venter (2020).

Identification of first and last mile modes The methodology needs to be generic enough to be adaptable to a variety of modes operating in a specific 1LM environment. The first challenge is to define what exactly is meant by the first and last mile. It should by definition include the very first and last component to/from the origin and destination, typically a walking link. But which other modes are to be included depends on the problem at hand: if the interest is in the 1LM to/from a rail system, it might include local buses and taxis serving rail stations. If the interest is in the 1LM connection of that local bus system, it might only include the walking mode between bus stops and origins/ destinations. It follows that even within the same geographic area, the definition of 1LM might vary depending on the goal of the analysis, which is important to clarify at the outset. The 1LM evaluation methodology should ideally allow multiple modes to be assessed within the same framework for the sake of comprehensiveness and comparability. The selected modes should also take into account differences between the access and egress side of public transport trips. As an illustration, Figure 22.2 shows the access and egress modes of a sample of rail passengers of the Gautrain rail system, based on passenger surveys at the two 312

The first/last mile connection

Figure 22.2 Access and egress modes of Gautrain passengers, Hatfield and Park stations Source: Based on data supplied by Gautrain Management Agency

terminal stations. Modal asymmetry is clear: walking and feeder bus play a much larger role as last-mile modes for egress trips, while park-and-ride dominates for the first-mile trip to the station. In this case, walking and feeder bus were chosen as 1LM modes for analysis.

Delimiting the service area and identifying routes Two approaches are possible to determine the 1LM area of interest: either a discrete distance band can be defined around each station or stop to delimit the service area, or the spatial coverage of 1LM trips can be identified directly through surveys or direct observation, regardless of distance travelled to/from the station. As discussed previously, no consensus exists among practitioners or researchers regarding a uniform standard for, nor uniform approach to estimating, catchment area size (Jiang et al., 2012). The more flexible second approach seems preferable, as it reflects the actual travel patterns of passengers. Agrawal et al. (2008) and Jiang et al. (2012) have demonstrated the feasibility of identifying 1LM routes by asking passengers to trace the route of their walk trips on a map, which then become the basis of the 1LM evaluation. If the costs of conducting local surveys are prohibitive, it is suggested that the expert knowledge of people familiar with the area, together with geographic information system mapping of points of interest, could supplement or replace passenger surveys. Further research is needed on costeffective ways of using various data sources to identify 1LM routes. The proposed procedure delivers a set of routes along which 1LM trips take place, together with a set of origins and destinations that are the non-public transport end of the 1LM trip. It is thus a route-based, rather than an area-based, measure. If a single 1LM measure is calculated for all the routes serving a particular station or set of stations, then it might be preferable to weight the different routes according to the importance of the route. This could potentially be done using the number of origins/destinations served by a route or the number of people using a given route.

Attributes for assessment The walkability literature has proven that the environmental quality of an area can be “unpacked” into its components (Adkins et al., 2012) for purposes of evaluation. The component factors 313

Christoffel Venter

affecting 1LM quality for a particular area need to be selected according to the modes under consideration and local conditions. The selection can draw on the research described previously but needs to be tailored for each place, perhaps using local surveys and expert experience. In the Gautrain case, the researchers used a combination of global and local research to select the 19 quality attributes listed in Table 22.1. The chosen attributes emphasise micro-level design factors of the walking environment and bus infrastructure, while macro-level land use factors such as density and access to opportunities are incorporated indirectly through distance and time measures of the 1LM trip. The quality of routes can be assessed using a variety of tools, including Google Street View images, audits, user questionnaires, and checklists (Aghaabbasi et al., 2017). The purpose is to provide a rating against each of the previous factors for each route, link (e.g. sidewalk section), or node (e.g. intersection). Audits are the most common, as they can reliably be conducted by trained auditors (Adkins et al., 2012; Agrawal et al., 2008), as long as care is taken to ensure consistency and reliability (Clifton et al., 2007). In the case of Gautrain, trained volunteers assessed each route on foot, scoring each element against the previous attributes on a scale

Table 22.1 Categories, attributes, and user-derived weights included in first/last mile quality assessment of Gautrain, South Africa Broad category of first/last mile attributes

Specific attributes

Importance weighting derived from user surveys

Safety from crime while waiting for and walking to buses

o o o o

0.11 0.11 0.10 0.11

Comfort of waiting areas

Ease of finding information Safety from traffic accidents while waiting and walking Sidewalk comfort and quality

Time and distance of access trip

Cost (fare) of access trip

CCTV monitoring Visible security and police Emergency call box Lighting quality at sidewalks and bus stops o Overhead shelter (rain and sun) o Resting facilities o Wi-Fi provision at waiting areas o Info on bus delays and arrivals o Alternate route information o Safe road crossing o Safe place to walk and wait for buses o Walkways wide and obstruction free o Clean and pleasant street environment o Walkways flat, even, and neat o Short walking distance to/from bus stop o Short waiting time for bus o Short travel time to station o Cost of feeder trip o Pay point machines at bus stops TOTAL

Source: Venter, 2020

314

0.07 0.05 0.03 0.05 0.03 0.03 0.03 0.004 0.01 0.001

0.04 0.06 0.04 0.06 0.05 1.00

The first/last mile connection

between –2 and +2, with a score of 0 being fixed as the minimum acceptable level of quality. For instance, the attribute walkway width and obstructions received a score of +2 if the walkway was at least 1.8 m wide and obstacle free, 0 if it was 1.2 m at its narrowest point (the minimum width prescribed in South Africa (Department of Transport, 2014), and –2 if it was less than 1.2 m wide or had obstacles like street furniture. A total of 129 1LM routes around ten rail stations were assessed, covering 350 km of walk routes. Figure 22.3 shows an example of attribute scores for the 1LM environment of two stations. It is clear that, while considerable heterogeneity exists across the factors, categories like comfort of waiting areas (bus stops) and time/distance and cost score consistently lower. This is especially so in the more suburban station environments.

Figure 22.3 Audit scores for two example Gautrain stations

315

Christoffel Venter

Incorporation of local attitudes regarding relative importance of attributes In order to reflect user needs, the score of each attribute was weighted with an importance weight derived from user surveys. Various techniques have been used to examine the relative weighting of attributes, including analytical network processes of experts’ opinions (Naharudin et  al., 2017), correlation analysis (e.g. Ji  & Gao, 2010), and importance rating surveys (e.g. Aghaabbasi et al., 2017; Agrawal et al., 2008). In the Gautrain case, an intercept survey was used to obtain importance ratings from a sample of 250 rail and BRT users. The sample was slightly skewed towards younger users but provided fairly representative data. The results shown in Figure 22.4 suggest that passengers have a very clear ranking of 1LM attributes. Safety from crime is considered the most important by far, followed by the cost of the access trip. The dominance of personal safety concerns mirrors findings in the United States (Agrawal et al., 2008; Tilahun et al., 2016). The concern with cost is novel, as it has not been included in walkability research, but is clearly important where public or private for-hire modes enter the 1LM scene, especially so in low-income communities. Sidewalk quality and comfort is the least important issue to this group of South African users, unlike typical findings elsewhere (Aghaabbasi et al., 2017; Agrawal et al., 2008).

Definition of summary indices Indices summarising the audited quality of an individual 1LM route, subsets of routes, all routes serving a station, or all stations in a system can be calculated as the weighted average attribute score. Attribute weights were derived by converting results from the importance survey to weights that add up to 1.0, as shown in Table 22.1 for the Gautrain case. The results, still scaled between –2 and +2, can be compared to identify trends and concerns. Figure  22.5 shows, for example, the frequency distribution of weighted route scores for all Gautrain stations but separated by location type. It indicates, first, that the majority of 1LM environments perform

Figure 22.4 User survey: Percentage of respondents identifying a 1LM attribute as among top three most important

316

The first/last mile connection

Figure 22.5 Frequency distribution of 1LM route scores around Gautrain stations, South Africa

relatively poorly (i.e. below the minimum acceptable level indicated by a score of zero). Second, suburban environments perform worse than urban ones, due to a combination of relatively poorer sidewalk and bus stop infrastructure and low densities (necessitating long and costly bus feeder trips).

Conclusions and further research As awareness grows of the importance of the 1LM as an integral part of the public transport journey, so does the need to better understand how it affects passengers’ (and prospective passengers’) behaviour. One potentially fruitful strand of research is to use qualitative techniques to explore intersections between demographics, modal experiences and options, and 1LM needs, such as in the work by Hickman and Vecia (2016) in London. It might lead to new insights into the equity and environmental impacts of various 1LM strategies in terms of their contribution to achieving sustainable transport goals. Evidence suggests that passengers’ 1LM needs and behaviours vary significantly across places and cultures; this variation needs to be better understood. This is especially important as cities are standing on the cusp of major changes in 1LM offerings brought about by shared mobility, micromobility, and new market models, the implications of which are not yet understood. The majority of 1LM studies to date investigate rail as main mode; access to newer modes like bus rapid transit is understudied. Qualitative research should supplement quantitative studies exploring how these offerings are likely to change travel behaviour broadly, public transport mode choice specifically, and mobility in cities. While some visions of the future foresee new technologies drastically reducing the role of both walking and local bus operations in the 1LM environment, this is speculative and deserves critical analysis. Behavioural studies need to be supported by better data collection of the 1LM trip. While the collection of walking and cycling data in travel surveys has improved, multi-modal trips are still often underreported (Clifton  & Muhs, 2012), hampering our ability to characterise the 317

Christoffel Venter

1LM trip to/from public transport. Big data techniques using smart cards and in-vehicle navigation might exacerbate the problem, as these might miss the 1LM trip altogether. Better techniques are also needed for assessing the quality of the 1LM trip using in-field or remote sources. The chapter described an audit-based technique that takes the subjective requirements of 1LM users into account when evaluating 1LM quality. The suggested framework is completely generic with regard to modes, catchment areas, attributes or factors, and how they should be weighted and may easily be customised to fit individual situations. It proposes a route-based rather than area-based approach for the evaluation of 1LM connections to specific public transport services. Its strength lies in its ability to reflect the multi-modal nature of the 1LM environment, as the attributes of any combination of 1LM modes may be included. It has been applied only to assess walking and feeder bus modes, but may easily be extended to include cycling, taxis, e-hailing, and park-and-ride. More research is needed on how to incorporate informal 1LM modes prevalent in developing countries, as it may be more difficult to identify 1LM routes and stops. Potential difficulties that need further research arise when users of different 1LM modes attach different importance weights to attributes, which might well happen if 1LM modes serve different socioeconomic groups. Further research is also needed on how to properly reflect accessibility to opportunities as an attribute, as this is a critical aspect of the utility of 1LM regimes but only incorporated indirectly. Further work is needed to test the transferability and robustness of the technique to ensure that similar ratings in different places carry the same qualitative meaning, especially given the potential subjectivity that is introduced by the use of trained auditors for conducting the environmental ratings. It might also be useful to relate the scores explicitly to a standardised scale such as level of service ratings between A and F. Last, most research has focused on the first mile, or the access part of the public transport trip. Research shows that the last mile might be quite different in terms of its characteristics and user behaviour; future research should differentiate between first- and last-mile issues more explicitly.

Acknowledgements This work was partly funded by the Gautrain Management Agency and the Volvo Research and Educational Foundation via the BRT+ Centre of Excellence.

References Adkins, A., Dill, J., Luhr, G., & Neal, M. (2012). Unpacking walkability: Testing the influence of urban design features on perceptions of walking environment attractiveness. Journal of Urban Design, 17(4), 499–510. Aghaabbasi, M., Moeinaddini, M., Shah, M. Z., & Asadi-Shekari, Z. (2017). A new assessment model to evaluate the microscale sidewalk design factors at the neighbourhood level. Journal of Transport & Health, 5, 97–112. Agrawal, A. W., Schlossberg, M., & Irvin, K. (2008). How far, by which route and why? A spatial analysis of pedestrian preference. Journal of Urban Design, 13(1), 81–98. Behrens, R. (2004). Understanding travel needs of the poor: Towards improved travel analysis practices in South Africa. Transport Reviews, 24(3), 317–336. Boarnet, M. G., Giuliano, G., Hou, Y., & Shin, E. J. (2017). First/last mile transit access as an equity planning issue. Transportation Research Part A: Policy and Practice, 103, 296–310. Canales, D., Bouton, S., Trimble, E., Thayne, J., Da Silva, L., Shastry, S., Knupfer, S.,  & Powell, M. (2017). Connected urban growth: Public-private collaborations for transforming urban mobility. Coalition for Urban Transitions c/o World Resources Institute, 10, 39–45. Retrieved, June 25, 2020, from http:// newclimateeconomy.net/content/cities-working-papers.

318

The first/last mile connection Carr, L. J., Dunsiger, S. I., & Marcus, B. H. (2010). Walk Score™ as a global estimate of neighborhood walkability. American Journal of Preventive Medicine, 39(5), 460–463. Cervero, R. (1998). The transit metropolis: A global inquiry. Island Press. Cervero, R. (2001). Walk-and-ride: Factors influencing pedestrian access to transit. Journal of Public Transportation, 3(4), 1. Chandra, S., Bari, M. E., Devarasetty, P. C., & Vadali, S. (2013). Accessibility evaluations of feeder transit services. Transportation Research Part A: Policy and Practice, 52, 47–63. Cheng, C.-L., & Agrawal, A. W. (2010). TTSAT: A new approach to mapping transit accessibility. Journal of Public Transportation, 13(1), 55–72. Clewlow, R. R., & Mishra, G. S. (2017). Disruptive transportation: The adoption, utilization, and impacts of ride-hailing in the United States: Research Report UCD-ITS-RR-17–07, Institute of Transportation Studies, University of California, Davis. Clifton, K. J.,  & Muhs, C. D. (2012). Capturing and representing multimodal trips in travel surveys: Review of the practice. Transportation Research Record, 2285(1), 74–83. Clifton, K. J., Smith, A. D. L., & Rodriguez, D. (2007). The development and testing of an audit for the pedestrian environment. Landscape and Urban Planning, 80(1–2), 95–110. Department of Transport. (2014). Non-motorised transport facility guidelines: Policy and legislation, planning, design and operations (Draft). Pretoria, South Africa. Dill, J. (2003). Transit use and proximity to rail: Results from large employment sites in the San Francisco, California, Bay Area. Transportation Research Record, 1835(1), 19–24. Doolittle, J., & Porter, E. K. (1994). Integration of bicycles and transit. TCRP Synthesis 4. Transportation Research Board. El-Geneidy, A., Grimsrud, M., Wasfi, R., Tétreault, P., & Surprenant-Legault, J. (2014). New evidence on walking distances to transit stops: Identifying redundancies and gaps using variable service areas. Transportation, 41(1), 193–210. Ewing, R., & Cervero, R. (2001). Travel and the built environment: A synthesis. Transportation Research Record: Journal of the Transportation Research Board, 1780, 87–114. Frank, L. D., Sallis, J. F., Saelens, B. E., Leary, L., Cain, K., Conway, T. L., & Hess, P. M. (2010). The development of a walkability index: Application to the Neighborhood Quality of Life Study. British Journal of Sports Medicine, 44(13), 924–933. Gebel, K., Bauman, A.,  & Owen, N. (2009). Correlates of non-concordance between perceived and objective measures of walkability. Annals of Behavioral Medicine, 37(2), 228–238. Givoni, M., & Rietveld, P. (2007). The access journey to the railway station and its role in passengers’ satisfaction with rail travel. Transport Policy, 14(5), 357–365. Glazier, R., Weyman, J., Creatore, M., Gozdyra, P., Moineddin, R., Matheson, F., & Booth, G. (2012). Development and validation of an urban walkability index for Toronto, Canada. Toronto Community Health Profiles Partnership, 1–21. Grasser, G., Van Dyck, D., Titze, S., & Stronegger, W. (2013). Objectively measured walkability and active transport and weight-related outcomes in adults: A  systematic review. International Journal of Public Health, 58(4), 615–625. Guerra, E., Cervero, R.,  & Tischler, D. (2012). Half-mile circle: Does it best represent transit station catchments? Transportation Research Record, 2276(1), 101–109. Harkey, D. L., Reinfurt, D. W., & Knuiman, M. (1998). Development of the bicycle compatibility index. Transportation Research Record, 1636(1), 13–20. Hickman, R., & Vecia, G. (2016). Discourses, travel behaviour and the ‘last mile’ in London. Built Environment, 42(4), 539–553. Hochmair, H. H. (2015). Assessment of bicycle service areas around transit stations. International Journal of Sustainable Transportation, 9(1), 15–29. Hoehne, C. G., & Chester, M. V. (2017). Greenhouse gas and air quality effects of auto first-last mile use with transit. Transportation Research Part D: Transport and Environment, 53, 306–320. Ji, J.,  & Gao, X. (2010). Analysis of people’s satisfaction with public transportation in Beijing. Habitat International, 34(4), 464–470. Jiang, Y., Zegras, P. C., & Mehndiratta, S. (2012). Walk the line: Station context, corridor type and bus rapid transit walk access in Jinan, China. Journal of Transport Geography, 20(1), 1–14. Keijer, M., & Rietveld, P. (2000). How do people get to the railway station? The Dutch experience. Transportation Planning and Technology, 23(3), 215–235.

319

Christoffel Venter Khisty, C. J. (1994). Evaluation of pedestrian facilities: Beyond the level-of-service concept. Transportation Research Record, 1438, 45–50. Kim, S., Ulfarsson, G. F., & Hennessy, J. T. (2007). Analysis of light rail rider travel behavior: Impacts of individual, built environment, and crime characteristics on transit access. Transportation Research Part A: Policy and Practice, 41(6), 511–522. Kittleson & Associates Inc., Parsons Brickherhoff, KFH Group Inc., Texas A&M Transportation Institute, Arup. (2013). Transit capacity and quality of service manual (3rd ed.). TCRP Report 165. Transit Cooperative Research Program. Krygsman, S., Dijst, M., & Arentze, T. (2004). Multimodal public transport: An analysis of travel time elements and the interconnectivity ratio. Transport Policy, 11(3), 265–275. Kumar, A., & Roy, U. K. (2019). E-rickshaws as sustainable last mile connectivity in an urban dilemma: Case of Delhi. Paper presentation. The International Conference on Transportation and Development 2019: Innovation and Sustainability in Smart Mobility and Smart Cities. Lesh, M. (2013). Innovative concepts in first-last mile connections to public transportation. In S. Jones (Ed.), Urban public transportation systems (pp. 63–74). American Society of Civil Engineers (ASCE). Liu, X., Porter, R., Zlatkovic, M., Fayyaz, K., & Taylor, J. (2018). First and last mile assessment for transit systems. Mountain-Plains Consortium. Liu, Z., Jia, X., & Cheng, W. (2012). Solving the last mile problem: Ensure the success of public bicycle system in Beijing. Procedia-Social and Behavioral Sciences, 43, 73–78. Loutzenheiser, D. R. (1997). Pedestrian access to transit: Model of walk trips and their design and urban form determinants around Bay Area Rapid Transit stations. Transportation Research Record, 1604(1), 40–49. Manaugh, K.,  & El-Geneidy, A. (2011). Validating walkability indices: How do different households respond to the walkability of their neighborhood? Transportation Research Part D: Transport and Environment, 16(4), 309–315. Martens, K. (2007). Promoting bike-and-ride: The Dutch experience. Transportation Research Part A: Policy and Practice, 41(4), 326–338. Meng, M., Koh, P. P., & Wong, Y. D. (2016). Influence of socio-demography and operating streetscape on last-mile mode choice. Journal of Public Transportation, 19(2), 3. Morency, C., Trépanier, M., & Demers, M. (2011). Walking to transit: An unexpected source of physical activity. Transport Policy, 18(6), 800–806. Mulley, C., Ho, C., Ho, L., Hensher, D., & Rose, J. (2018). Will bus travellers walk further for a more frequent service? An international study using a stated preference approach. Transport Policy, 69, 88–97. Naharudin, N., Ahamad, M. S. S., & Sadullah, A. F. M. (2017). Assessing criteria of the pedestrian-friendly first/last mile transit journey by using analytical network process (ANP) group judgment. Paper presentation. Proceedings of the ICOSH-UKM 2017. Owen, A., & Levinson, D. M. (2015). Modeling the commute mode share of transit using continuous accessibility to jobs. Transportation Research Part A: Policy and Practice, 74, 110–122. Park, S., Deakin, E., & Jang, K. (2015). Can good walkability expand the size of transit-oriented developments? Transportation Research Record, 2519(1), 157–164. Park, S., & Kang, J. (2008). Factors that influence walking and biking to the station: Modeling commuter rail user’s access mode choice. Paper presentation. The Transportation Research Board 87th Annual Meeting. Phillips, R. G., & Guttenplan, M. (2003). A review of approaches for assessing multimodal quality of service. Journal of Public Transportation, 6(4), 4. Replogle, M. (1993). Bicycle access to public transportation: Learning from abroad. Transportation Research Record, 1396, 75. Saelens, B. E., & Handy, S. L. (2008). Built environment correlates of walking: A review. Medicine and Science in Sports and Exercise, 40(7 Suppl), S550. Shaheen, S., & Chan, N. (2016). Mobility and the sharing economy: Potential to facilitate the first-and last-mile public transit connections. Built Environment, 42(4), 573–588. Talen, E., & Koschinsky, J. (2013). The walkable neighborhood: A literature review. International Journal of Sustainable Land Use and Urban Planning, 1(1). Tilahun, N., Thakuriah, P. V., Li, M., & Keita, Y. (2016). Transit use and the work commute: Analyzing the role of last mile issues. Journal of Transport Geography, 54, 359–368. Transport for London. (2015). Assessing transport connectivity in London. Transport for London. Van Soest, D., Tight, M. R., & Rogers, C. D. (2020). Exploring the distances people walk to access public transport. Transport Reviews, 40(2), 160–182.

320

The first/last mile connection Venter, C. J. (2020). Measuring the quality of the first/last mile connection to public transport. Research in Transportation Economics, 83, 100949. Vuchic, V. R. (2017). Urban transit: Operations, planning, and economics. John Wiley & Sons. Wang, F., Ross, C. L., & Karner, A. (2019). Understanding the influence of mobility as a service (MAAS) on job accessibility and transportation equity. Paper presentation. The Transportation Research Board 98th Annual Meeting. Wang, R., & Liu, C. (2013). Bicycle-transit integration in the United States, 2001–2009. Journal of Public Transportation, 16(3), 95–119. Yan, X., Zhao, X., Han, Y., Van Hentenryck, P.,  & Dillahunt, T. (2019). Mobility-on-demand versus fixed-route transit systems: An evaluation of traveler preferences in low-income communities. arXiv preprint arXiv:1901.07607. Zellner, M., Massey, D., Shiftan, Y., Levine, J., & Arquero, M. (2016). Overcoming the last-mile problem with transportation and land-use improvements: An agent-based approach. International Journal of Transportation, 4(1), 1–26. Zuo, T., Wei, H., Chen, N., & Zhang, C. (2020). First-and-last mile solution via bicycling to improving transit accessibility and advancing transportation equity. Cities, 99, 102614.

321

23 PUBLIC TRANSPORT AND THE BUILT ENVIRONMENT Murtaza Haider and Ahmed El-Geneidy

Introduction The economic, cultural, and social success of large and small cities, to an extent, depends upon the ease of mobility and accessibility to destinations. Modern metropolises are also known for their advanced transportation systems that enable accessibility and mobility. A  glance at the populace and economically vibrant cities in Southeast Asia, Europe, and North America would reveal several common factors that contribute to their success. One of those factors is an efficient public transport system that provides comprehensive spatial coverage throughout the urban landscape. Densely populated cities like New York and Hong Kong cannot function without the elaborate public transport spatial networks that provide access to employment hubs during peak periods of travel. Whereas comprehensive and efficient public transport systems have contributed to the economic and social success of cities across the globe, the presence of public transport-based mobility continues to be concentrated in a small number of large cities. In North America, despite the significant contribution of public transport-based mobility to enhancing economic and social outcomes in populous cities, public transport use is somewhat limited in mid- to smallsized cities. Meanwhile, the experience in mid- to small-sized European cities is different from the one in America such that commutes by public transport continue to account for a sizable proportion of all trips made. Public transport use is more pronounced in cities of varying sizes in Europe but not as much in North America. Researchers have identified several reasons for the differences. The most cited difference between cities in Europe and North America is their respective built environments. European cities are known for their high-density, compact, mixed land uses, which researchers believe have enabled more extensive adoption of public transport. In comparison, except for larger and more populous cities, a limited number of North American cities, urban land use in North American cities is characterised by mid- to low-density, sprawling, and single-purpose land uses. Researchers point out that vintage matters for land use. The land use in European cities, most of which were predominantly built before the emergence of the private automobile or rail-based public transport, reflects the prevalent transport technology of their times. It has been argued that the maximum desirable commute throughout human history, irrespective of the transportation technology, has been 45 minutes (Garreau, 1991). This 322

The built environment

implies that land use and the physical extent of the city from ancient Rome (when the mode of transportation was predominantly walking) to the modern mixed-mode transportation cities should be so that the desirable (average) commute times are less than 45 minutes. It is argued that in addition to vintage, the scarcity of developable land is more pronounced in European cities than their North American counterparts, which made post-war development transpire at much higher densities in Europe. At the same time, comparatively higher gasoline taxes, parking charges, and other levies in Europe make mobility by private automobile significantly more expensive than it is in North America. This chapter focuses, because of the availability of suitable data, on the relationship between the built environment and mobility by public transport in selected North American cities. The aim of this chapter is not to solely focus on how one can increase mobility by public transport in urban settings where travel by public transport has taken a backseat. Instead, this inquiry also focuses on the circumstances that make public transport not the preferred alternative for travellers, particularly because trips by public transport are, on average, much longer in duration than the comparable trips by private automobile, and with individuals assumed to be utility maximisers, the desire to optimise their utility often overshadows the desire to have the better societal outcomes that would come with public transport rather than car trips. This chapter comprises two parts. First, the chapter undertakes a systematic literature review to determine how research in the past has addressed the relationship between the built environment and public transport. In the second part, the chapter focuses on empirical evidence, mostly in the form of spatially disaggregated demographic data, complemented with spatial measures of accessibility and travel behaviour to answer more nuanced questions. For instance, is public transport use in the newly built part of the traditionally public transport-dominated cities significantly higher than neighbourhoods of similar vintage in other cities not known for higher public transport use? The chapter concludes by identifying the enabling factors for better public transport outcomes and avenues for further exploration in the future.

What is known from the literature A symbiotic relationship can be imagined between the built environment and public transport use. Public transport-supportive built form, the specifics of which are discussed later, can promote public transport use. Consequently, and over time, newly built higher-order public transport could influence the built environment in proximity of public transport stations. Hence, a bidirectional model of the built environment and public transport use emerges. The literature presented here, though not exhaustive, presents the essential themes that have come to define the prevalent “conventional wisdom” as it relates to how the built environment and public transport use are connected. Slightly divergent views from the literature are presented when empirical tests reveal weak or no association between the built environment and public transport use. A cursory look at the academic and professional literature suggests that the contemporaneous correlation between proxies of the built environment and travel behaviour has been the most frequent output of studies exploring the determinants of public transport use. The built environment attributes, namely population or employment densities measured as persons or jobs per unit area at the neighbourhood, city, regional, or national level; housing density, intersection density, road-length or sidewalks per unit area, land use mix, and prevalence of non-residential land uses, such as retail, in the walking distance have served as proxies to describe built environment at varying spatial scales. Others have broadened the list by including public transport 323

Murtaza Haider and Ahmed El-Geneidy

supply attributes, namely public transport service frequency and proximity to public transport stations, among others. Research has mostly documented that a compact built environment discourages travel by private automobile and encourages travel by public transport (Ewing & Cervero, 2017). Cervero and Kockelman (1997) analyzed the travel mode choices for non-work trips for the residents in the San Francisco Bay Area and concluded that the built environment attributes, clustered as density, land-use diversity, and design (3Ds), are correlated with travel behaviour. They observed that high population and employment density, diversity in land uses, and urban design that favoured mobility by non-motorised modes were able to reduce trip generation rates and to increase travel by non-motorised modes. The strength of the relationships they uncovered was weak, as they concluded that the influence of the 3Ds on travel behaviour appeared to be modest at best, concluding travel behaviour might be influenced by compact and diverse built form that favours mobility by non-motorised modes. The 3Ds highlighted as a bundled set of attributes have, either individually or collectively, dominated the discourse on the determinants of travel behaviour. Earlier, Newman and Kenworthy (1989) highlighted the need to reduce automobile dependence by focusing on factors that help reduce automobile and gasoline use. Compact urban form or density emerged as the central theme since then and was later reiterated in a series of subsequent publications that demonstrated lower automobile use in jurisdictions with higher population densities (Kenworthy & Laube, 1999). Bivariate comparisons of population densities and automobile use were instrumental in framing policies on how to increase public transport ridership while reducing dependence on the private automobile. However, as was demonstrated by others (Handy et al., 2006), comparing a proxy of the built environment and one for travel behaviour in isolation ignores the other mitigating factors that influence the evolution of the built environment and associated human behaviour. Often, high-density cities are older than low-density cities, such that the higherdensity neighbourhoods were primarily built earlier before the use of automobiles became ubiquitous. Controlling for the age of a place is likely to demonstrate the inherent limitations in bivariate correlations, which ignore the structural differences amongst jurisdictions. This vintage argument is considered again later. Neighbourhood design attributes that influence connectivity are also associated with travel behaviour, especially the use of non-motorised modes. At the same time, the built environment attributes that promote walk or bicycle modes might be associated with lower public transport mode share, thus exposing the trade-off between non-motorised modes and public transport. A study from Beijing demonstrated that “[h]igher destination accessibility, a higher number of exclusive bicycle lanes, a mixed environment and greater connectivity between local streets tend to increase the use of the bicycle” (Zhao, 2014, p. 1019). Two additional findings were equally relevant. One was that residential density had “no significant effects on the use of a bicycle for commuting” (p. 1019). Second, an increase in the supply of public transport was associated with a “decrease rather than increase [in] bicycle commuting” (p.  1019). Zhao (2014) found that “drastic changes in the built environment are a major reason for the demise of ‘the kingdom of bicycles’ in China” (p.  1019). This implies that a very walkable built environment might not necessarily be equally supportive, or facilitating, of travel by public transport. In fact, at extremely high population densities, one sees diminishing returns to density for public transport as the mode of travel switches in favour of walking and cycling (Haider, 2019). The correlation between the walkability of a place and greater use of public transport has been reported independently in different jurisdictions (Frank et al., 2006). However, such a line of inquiry often makes certain assumptions implicitly or explicitly and, at times, ignore factors 324

The built environment

that may influence the statistical significance and magnitude of the association between attributes of the built environment and travel behaviour. Glazier et al. (2014), for instance, assume that the presence of walkable destinations and residential density could be achieved by modifying built form. The authors assume that, as an outcome of public policy interventions, modifying the built environment by changing its density or land use is doable. However, transforming existing land uses in the short term requires political and financial capital, changing demographics, and the resulting change in cultural preferences, all of which cannot be readily accomplished by recommending “for use by policy-makers, planners and public health officials”. The other associated challenge is with self-selection, where those who are more likely to have opted for non-motorised modes of travel or public transport would self-select themselves in such locations that permit the desired travel behaviour Handy et  al. (2006), Cervero and Duncan (2008), Cao et al. (2009), Mokhtarian and Cao (2008), Cao et al. (2006), Næss (2009). After accounting for self-selection, the strength of the relationship between built environment attributes and travel behaviour might be weaker than expected. When it comes to travel mode choice decisions, Cervero and Duncan (2008) show “that residential self-selection accounts for approximately 40 percent of the rail-commute decision” (p. 1) (see also Chapter 4).

The geography of intellectual curiosity While reviewing research on the linkages between the built environment and public transport use, a small number of cities can be identified that have served as experimental laboratories. Influential research from the small number of cities has been instrumental in devising policies linking the built environment and public transport. A question emerges about the relevance of these studies for other cities with distinct demographics, topology, climate, and economy. A large number of highly influential studies, for instance, are based on data from Bay Area in San Francisco (Cervero  & Landis, 1997); Portland, Oregon (Li et  al., 2008; Dong & Zhu, 2015); New York (Chen et al., 2008); King County (Frank et al., 2006); and Beijing (Zhao, 2014). Furthermore, few international comparisons of built environment-public transport use nexus could be found in the literature. Also, how sampled cities were selected and what was done to account for the order-of-magnitude difference between the sampled cities is not known. A comparative study of the built environment of 15 cities located in 12 countries found considerable differences in the built form (Adams et  al., 2014). The authors found “a 38-fold difference in median residential densities, a 5-fold difference in median intersection densities and an 18-fold difference in median park densities” (p. 1). Hence the question that emerges is whether one can generalise findings from a relatively high or very high-density place, such as Hong Kong, to North Shore, New Zealand, which depicts much lower development densities. At the same time, few if any studies consider the relationship between the built environment and public transport for European cities. The relative scarcity of research exploring the linkages between the built environment and public transport use from European cities needs some reflection. Perhaps similarities in the built environment among European cities, the similar vintage of their construction and development over the past few centuries, and much higher use of public transport throughout the European continent are the reasons such exploration was not needed. Also, the sociopolitical frameworks commonly found in Europe, for instance, acceptance of higher levels of taxation in return for higher levels of public services and social security safety nets, are inherently different from those prevalent in the United States and Canada. Hence, intercity comparisons of jurisdictions across the Atlantic would ignore not just the fact that most North American urban centres, unlike the European counterparts, 325

Murtaza Haider and Ahmed El-Geneidy

developed primarily in the past 100  years, but also that sociodemographic and economic frameworks across the continents differ considerably.

Old versus new cities Comparative studies of the built environment and transportation often control for the systematic differences between and among cities. For instance, empirical analysis controls for household demographics, income levels, automobile ownership, price of gasoline, and the like. What is often not controlled for is vintage. The built environment, to a large extent, is a product of the prevalent transportation technology. Since human beings started living in formal settlements, the shape and size of the settlements have been influenced by transportation technology. The spatial extent of the mostly pedestrian communities in ancient Rome and Egypt was defined, to a large extent, but the distances covered comfortably by foot. As transportation technology improved first with domesticated animals and later with motorised transportation, the spatial extent and structure of communities changed considerably. The built environment seen in the modern European metropolises with narrow streets, tree-lined boulevards, and low-rise multiresidential buildings supporting very high population densities was a reflection of the modes of transportation available when these places were first built and settled over the past few hundred years. In comparison, the North American metropolises were mostly developed when the private automobile and rail-based, higher-order public transport were readily available. The resulting built form in North America is thus a reflection of the automobility of its time. Neighbourhood vintage influences the built environment, which in turn influences public transport use. For comparative research, one must, therefore, attempt to control for vintage. Comparing cities built mostly in the 19th century or before with those developed primarily in the 20th century must include explicit controls for vintage to tease out the nuanced relationship between the built environment and public transport. Even comparing areas of different vintage within the same cities must not ignore the timing of development. For instance, Dong and Zhu (2015), in a study of smart growth developments in Portland and Los Angeles, observed that older neighbourhoods, when compared with newer builds, tend to be more advanced on smart growth metrics. The next section presents empirical (stylised) findings of the relationship between the built environment and public transport. The analysis is informed by the discussion presented in the literature review. Questions as to why public transport use is higher in some jurisdictions and not in others are addressed. The literature reviewed in this chapter demonstrates various benefits of higher public transport use. Still, at least in North America, public transport use pales in comparison with that of the private automobile. Of course, efficient and reliable public transport is a prerequisite for the mass adoption of public transport. In places where public transport service is unreliable, not efficient, or not comprehensive in its coverage, people will likely use other modes for mobility. However, in many large North American cities that are characterised by the public transport-supportive built environment, public transport use still lags travel by private automobile. The analysis presented here explores the barriers to higher public transport use by exploring some factors that have not attracted sufficient attention in the past.

Spatial analysis of the built environment – public transport linkages The empirical analysis presented here is based mainly on Census-based data from Canada and the United States. A  new database of the built environment and public transport proximity 326

The built environment

indicators for almost 480,000 neighbourhoods across Canada is used to explore the relationship between the built environment and public transport use.

‘Public transport is better, but cars are faster’ Statistics Canada, Canada’s official statistical agency, for the first time in 2011 included a question about the duration of daily commutes in the National Household Survey (NHS) that replaced the long-form census. The respondents, employed individuals 15 years of age or older, were asked the following question: “How many minutes did it usually take this person to get from home to work?”1 The public-use microdata file was subsequently released in 2013, which prompted an op-ed in Canada’s national newspaper, the Globe and Mail. The op-ed revealed that commutes by public transport, irrespective of the local demographics, were significantly longer than commutes by the private automobile (Haider, 2013). On average, across Canada, public transport commutes were 81% longer in duration than those by car (Haider, 2013). Even for cities known for high public transport use, public transport fared poorly for travel time. The commute to work data, released as part of the NHS, showed that in a public transport-friendly city like Montréal, travel to work by car on average took 26.5 minutes. However, travel by public transport on average took 42 minutes. Thus, even in cities known for public transport use and infrastructure, commute by public transport was 58% longer. The op-ed revealed nothing new to the transportation planning fraternity. However, for the public transport enthusiasts, and the broader public, public transport times being considerably slower was a revelation. Four days after the op-ed was published, the newspaper dedicated a section to publishing feedback from readers who had reacted with hundreds of messages and comments. The exercise revealed that many who advocate for public transport earnestly believed that public transport commutes were faster than those by the private automobile and that if urban workers were moved in large numbers from using cars to public transport, the systemwide travel times would improve. Nevertheless, Haider (2013) noted that commute to work data challenges the notion that building more public transit will save travel time by shifting commuters from cars to public transit. How is it possible that transferring commuters from a faster mode of travel to a slower one will shorten travel times? Simple arithmetic and common sense suggest that system-wide travel times will instead be longer when more people commute by the slower mode, i.e., public transit. (p. A13) The more pertinent question to ask is even if public transport service is available and travel costs are not the deciding factor, why would one opt to commute by a slower mode to and from work? Urban commuters, especially those who are time poor, are likely to prefer the mode of travel that is faster. Hence, public transport does not have to be just cheaper or reliable; it must also offer travel times that are competitive with those by cars. In 2016, the long-form census in Canada reconfirmed what was observed five years earlier in the NHS. Public transport commutes on average were significantly longer than by cars in Canada’s eight most populous Census Metropolitan Areas (CMAs), which are collectively home to 18 million people, about 50% of Canada’s population. Commutes by public transport were, for example, 70% longer in Calgary than by car (Figure 23.1). At the lower end, public transport commutes were 58% longer by public transport in Winnipeg. Despite public transport commutes being significantly longer than those by cars, Canada’s most populous urban regions 327

Murtaza Haider and Ahmed El-Geneidy

Figure 23.1 The difference in average commute times by public transport and car (percentage)

reported high public transport use, with Toronto reporting a public transport mode split of 24.7%. Quebec City was at the lower end with 11.4% (Statistics Canada, 2019). A similar relationship is observed for urban regions in the United States. Higher public transport use is associated with longer commute times. The three-dimensional association between population density (a proxy for the built environment), commute times, and public transport mode share is compared for the core-based statistical areas (CBSAs). A CBSA is essentially a conurbation developed around an urban centre with a minimum population of 10,000. The 31 CBSAs are necessarily conurbations that stretch beyond state boundaries. The most populous CBSA with a population of 14.3 million is linked with New York City and includes areas in New Jersey and White Plain. The least populous in the dataset is San Rafael, CA, with a population of 259,358 individuals. The results are shown in the next two figures. Figure 23.2 provides evidence in support of the oft-cited observation about the built environment and public transport mode share that higher-density regions are associated with higher public transport mode share. Figure 23.3 illustrates the argument being made here that higher public transport use is associated with longer average commute times. CBSAs with public transport mode share of greater than 10% are associated with average commute times of 30 minutes or more. The size of circles in Figure 23.3 depicts the population densities. What is interesting is that some high-density CBSAs, characterised by large circles, report lower public transport mode share of 5% or less and commute times of less than 30 minutes. On the other hand, one sees some high-density CBSAs associated with higher public transport mode share and consequently longer average commute times. Thus, one could infer from Figure  23.3 that population density does not automatically associate with higher public transport mode share. This observation has also been made in the literature, such that higher density is a prerequisite but not a sufficient condition for higher public transport use. However, the primary inference remains the same as was observed for the Canadian data; that is, higher public transport mode share correlates with longer commute times. CBSAs are agglomerations of urban areas with significant diversity in demographics, where the most populous regions such as New York and Los Angeles have populations over 10 million, while smaller areas such as Gary, Indiana, have populations of 0.7 million. Similarly, the 328

The built environment

Figure 23.2 Population density and public transport mode share across CBSAs in the United States

Figure 23.3 Average commute time and public transport mode share across CBSAs in the United States

spatial footprint of these areas varies significantly, such that Washington-Arlington-Alexandria covers an area of 5,000 square miles, whereas Boston, Massachusetts, stretches over 1,160 square miles. One can argue that comparing CBSAs with such diversity in demographics and spatial extent could suffer from aggregation bias that may hide the nuanced local differences within and across CBSAs. 329

Murtaza Haider and Ahmed El-Geneidy

To address this limitation, a similar analysis performed at the census tract (C.T.) level for the New York region using data from the 2000 census is presented. Mainly, the 5,000-plus C.T.s constituting the New York region are divided into six discrete regions based on distance from downtown Manhattan. Figure 23.4, therefore, contains six panels, each representing a unique subset of C.T.s based on their distance from downtown Manhattan. The panel labelled nearest covers only those C.T.s that are nearest to downtown Manhattan and are situated within 10 km of the Central Business District (CBD). Other panels accordingly present data for C.T.s that are located at greater distances from downtown Manhattan. A breakdown of the C.T.s distances is presented in Table 23.1. The discrete slices of space allow an exploration of any change in the relationship based on the location of the C.T.s. The x-axis for the six panels represents public transport mode share, whereas the y-axis represents the average commute time in minutes. Also, the markers are customised to categorise each C.T. into low- (diamond), medium- (triangle), or high-income (square) neighbourhoods.

Figure 23.4 Population density, household income, and public transport mode share in the New York region

330

The built environment Table.23.1 Distance thresholds for the six neighbourhood categories Distance from Manhattan CBD (km)

mean

min

max

Nearest Nearer Medium near Medium far Farther Farthest

6.59 12.63 18.70 30.01 54.28 97.35

0.28 10.05 15.57 22.44 40.17 71.79

10.04 15.55 22.40 40.16 71.36 179.80

The panel labelled nearest presents a scatter plot between public transport mode share and median commute times for C.T.s situated within 10 km of the Manhattan CBD. One sees a positive correlation between public transport mode share and median commute times. Even for the neighbourhoods that are closest to downtown Manhattan, which is served by fast-moving subways on dedicated rights-of-way, commute times on average are higher for the neighbourhoods with higher public transport use. The positive correlation persists for C.T.s that are depicted in panels labelled as nearer, medium near, medium far, or farther from the CBD. The correlation weakens for only those neighbourhoods that are located at least 70 km away from downtown Manhattan. The shape of markers in the scatter plot, which accounts for the neighbourhood income level, reveals that low-income C.T.s (diamond) often report higher levels of public transport mode share and longer commute times. This is more evident for C.T.s located nearest or nearer to the CBD but not for remotely located suburban C.T.s. A key finding from Figure 23.4 is that a higher level of public transport mode share is associated with longer commute time. The relationship holds for both a comparative analysis of spatially aggregated data for CBSAs and also the spatially disaggregate data at the neighbourhood level (C.T.) for the New York region. This exposes a key challenge for increasing public transport use in North America. Even in places where the built environment is conducive to higher-order public transport systems and places with an ample supply of diverse public transport modes offering efficient public transport service, public transport commute times are longer. To compete with the private automobile, public transport must offer competitive travel times.

Does vintage matter? In the literature review section earlier, the relevance of vintage for the built environment and associated public transport use was discussed. Cities built before automobile use became ubiquitous were designed to facilitate mobility by non-motorised modes of travel. Specifically, such cities were known for higher population densities and mixed land uses, and the destinations were closely placed. Pre-auto era cities portrayed a compact urban form that was devoid of sprawl, which is characteristic of automobile-dependent communities. In this section, the relationship between vintage, specifically the age of a neighbourhood, is compared with the associated built environment and its association with public transport use. Data from the 2016 Census in Canada is presented for two urban regions, Montréal and Calgary. Data for respective Census Metropolitan Areas is used, where a CMA consists of “one or more neighbouring municipalities situated around a core. A census metropolitan area must have a total population of at least 100,000 of which 50,000 or more live in the core”.2 331

Murtaza Haider and Ahmed El-Geneidy

A spatially disaggregate approach is adopted where Calgary and Montréal CMAs are analyzed at the census tract level. C.T.s are “small, relatively stable geographic areas that usually have a population between 2,500 and 8,000 persons. They are located in census metropolitan areas and in census agglomerations that had a core population of 50,000 or more in the previous census”.3 The purpose of this analysis is to compare population densities, a proxy for the built environment, and public transport mode share in similar vintage neighbourhoods. Montréal is known as a public transport city and famous for its European-styled urban design (Racine, 2019). On the other hand, Calgary reports a much lower population density and public transport use. Competitive analysis of public transport mode share across Canada often highlights the apparent fact that places like Montréal, because of their compact built form, have been able to achieve much higher public transport use than places like Calgary, whose built environment is categorised as sprawling. Such intercity comparisons, as argued earlier, often ignore vintage. Calgary, unlike Montréal, is a newer city where a majority of the dwelling units enumerated in 2016 were constructed after 1980. In comparison, Montréal was settled much earlier in the 18th century and grew in space and population over the past 200 years. Using Calgary and Montréal as opposite poles of the built environment, the role, if any, that vintage may have played in their respective built environments and consequent public transport mode share is investigated. However, instead of comparing the entire urban landscape in one metropolis with that in the other, the public transport mode share and built environment for respective neighbourhoods that are differentiated by age are compared. A C.T. is categorised as new if most of the dwellings in the C.T. were constructed after 1980. Similarly, a C.T. is categorised as old if most dwellings were built before 1981. Hence, the comparative analysis presented here contrasts population densities and public transport mode shares for older neighbourhoods in Montréal with older counterparts in Calgary and vice versa. The results are presented in Table 23.2. The Montréal CMA has a population of 4 million, with 1.73 million dwellings. Calgary, on the other hand, is a smaller urban region with a population of 1.4 million inhabitants and

Table 23.2 A comparison of age differentiated neighbourhoods in Calgary and Montréal CMA

Dwellings in Built since 2016 1981

Montréal 1,727,215 Calgary 519,775 CMA

731,225 328,880

Built since 1981 (%)

42.3% 63.3%

New if at least 50% dwellings built since 1981 (%)

Population

Old New neighbour- neighbourhoods hoods

Total

Old New neighbour- neighbourhoods hoods

66.5 44.3

100 100

2,471,730 1,627,197 515,399 877,210

Employed Public transit Public transit mode 15-plus mode split split (%) (commuters) (%) Old New neighbour- neighbourhoods hoods

Montréal 1,883,920 Calgary 684,260

25.0 14.5

30.2 16.6

13.9 12.8

332

33.5 55.7

Population density (persons/sq. km) Old neighbourhoods

New neighbourhoods

7,125 2,777

2,352 2,605

 

 

   

   

The built environment

520,000 dwellings. Apart from the differences in size, the two cities are inherently different in the age of the constituting neighbourhoods, such that 63% of the dwellings in Calgary were constructed after 1980. In Montréal, post-1980 dwellings represented 42% of the dwellings. Whereas one in three neighbourhoods were predominantly built since 1981 in Montréal, 56% of the neighbourhoods were of post-1980 vintage in Calgary. The public transport mode split in Montréal at 25% for commute trips is 72% higher than the one in Calgary. However, the difference in public transport mode splits when compared for the entire urban areas ignores the vintage of the constituting neighbourhoods. To address this limitation, public transport mode splits between Montréal and Calgary are compared by dissecting the cities into old and new neighbourhoods. As expected, parts of Montréal characterised as old neighbourhoods reported a much higher public transport mode split of 30.2% compared to older parts of Calgary with the corresponding statistic at 16.6%. Unlike older neighbourhoods, a comparison of newer neighbourhoods between the two cities revealed that Montréal, with a 13.9% public transport mode split, was marginally better than 12.8% observed for Calgary. Interestingly, the significant difference in public transport mode splits observed for older neighbourhoods between the two cities almost disappears for newer neighbourhoods. A comparison of the means test revealed that the difference in mode splits for newer neighbourhoods was not statistically significant (p = 0.2870, Table 23.3). The difference in average public transport ridership for older parts of the two cities was statistically significant (p < 0.000). To a large extent, the previously mentioned results can be explained by the differences and similarities in the built environment between the two cities. Working with population density as a proxy for the built environment, one sees a significant difference in population densities of old neighbourhoods in Montréal and Calgary. However, that difference disappears, in fact, reverses, for newer neighbourhoods. Consider that the average population density in the older neighbourhoods of Montréal was recorded at 7,100 persons per square kilometre compared to 2,777 persons per square kilometre for older neighbourhoods in Calgary. However, when neighbourhoods built predominantly after 1980 in Montréal are compared with similar-vintage neighbourhoods in Calgary, then, at 2,605 persons per square kilometre, newer neighbourhoods in Calgary reported higher density than their newer counterparts in Montréal. These results present an interesting story. Public transport mode share is higher for a public transport-supportive built environment characterised by higher development densities, compact built form, and mixed land uses, to name a few. When differences in the built environment exist between cities, one also sees a difference in public transport mode shares. However, the differences in the built environment are more pronounced in parts of the cities built earlier. Recently developed neighbourhoods, even in public transport-friendly cities, have similar built environments like the ones found in newer neighbourhoods in cities not known for public transport use. As the difference in the built environments reduces or disappears over time, public transport Table 23.3 Comparison of means test for public transport mode share in new neighbourhoods in Calgary and Montréal Group

Obs

Mean

Std. err.

Std. dev.

[95% Conf. interval]

Montréal Calgary Combined Diff

306 141 447

13.88549 12.84501 13.55728 1.040476

0.611841 0.550887 0.453677 0.976152

10.70285 6.541415 9.5918

12.68152 11.75588 12.66567 –0.87796

333

15.08945 13.93414 14.44889 2.958916

Murtaza Haider and Ahmed El-Geneidy

mode shares also become similar across jurisdictions. Regrettably, the trend one sees here is of less reliance on public transport in newly built parts of cities, irrespective of the public transport use prevalent in the well-established, older parts of the city.

A national database of public transport accessibility Comparative analysis of the built environment and public transport use requires comparable data or metrics to explore the association between the built environment and public transport use. Developing such metrics is resource and time intensive. Comprehensive multijurisdictional studies of the built environment across countries or continents are few and often involve a handful of jurisdictions. The financial and resource constraints in data gathering and metric development are the reason built environment studies are confined to a select few cities in North America where there has been a tradition of collecting such datasets. National-level public-sector agencies have the capacity and the resources to develop crosscountry metrics for proximity, built environment, and public transport-supportive land uses. One such database is the recently released (April 2020) Proximity Measure Database (PMD) by Statistics Canada.4 In collaboration with the Canada Mortgage and Housing Corporation, Statistics Canada developed a nationwide database of 10 proximity/built environment-related metrics at the spatially disaggregated scale of dissemination blocks such that the database comprises half a million (mutually exclusive and almost exhaustive) observations. Statistics Canada defines the dissemination block (D.B.) as follows: A dissemination block is an area bounded on all sides by roads and/or boundaries of standard geographic areas. The dissemination block is the smallest geographic area for which population and dwelling counts are disseminated. Dissemination blocks cover all the territory of Canada.5 This extensive database makes it possible to answer questions about the relationship between the built environment and public transport at the national level. As pointed out earlier, most previous research has focused mainly on local urban markets, from which inferences were drawn for policymaking at the national, provincial, or regional levels that might not be relevant beyond the study areas. In the following paragraphs, a brief discussion of some of the relevant metrics from PMD is presented to determine the extent of public transport-supportive built environment at the national level and to compare relevant metrics amongst nine populous urban centres in Canada accounting for more than 50% of Canada’s population. Canada’s capital, Ottawa, spreads across two provinces, Ontario and Quebec, and therefore the metrics are presented separately for the two parts. The journey-to-work data collected as part of the census in 2016 showed that public transport mode split for commutes is highest in the Toronto Census Metropolitan Area, followed by Montréal and Vancouver such that the three most populous urban regions report at least one in five work-related trips being made by public transport (Figure 23.5). The public transport mode share declines to a low of under 10% for Hamilton, a neighbouring CMA to Toronto and part of the Toronto commuter belt. The public transport mode splits differ across the urban regions in the same way their demographics differ. Toronto, Montréal, and Vancouver are more populous, and their respective built environments, proxied by population and employment densities, compactness of urban form, diversity in land uses, and the like are more likely to facilitate commuting by public transport. Proximity to public transport infrastructure offering reliable and regular public transport service is a prerequisite for higher levels of public transport use. Often, proximity to public transport use is measured as the network or straight-line distance from trip origins to the nearest public transport station or stop. However, the public transport proximity measure developed by 334

The built environment

Figure 23.5 Public transport mode share in Canada’s most populous Census Metropolitan Areas, 2016

Statistics Canada not only accounts for spatial proximity to public transport infrastructure but also normalises proximity by the aggregate trip making activity in the morning peak period and public transport operating frequencies derived from the General Transit Fleet Specification. The measure is defined as follows: Proximity to public transit measures the closeness of a dissemination block to any source of public transportation within a 1 km walking distance. This measure is derived from the number of all trips between 7:00 a.m. – 10:00 a.m. from a conglomeration of 95 General Transit Feed Specification (GTFS) data sources. (Statistics Canada, 2020) Figure  23.6 presents the distribution of proximity to public transport index. A  quick comparison of Figures 23.5 and 23.6 reveals a lack of one-to-one correspondence between higher public transport proximity and higher public transport mode shares. Whereas Toronto CMA reported the highest public transport mode split, the highest proximity to public transport infrastructure is reported for Winnipeg, which reported a much lower public transport mode split. Similarly, whereas Edmonton and Calgary, the two most populous cities in Alberta, have ranked lower in public transport proximity, their corresponding public transport mode shares are higher than other regions that reported better proximity to public transport infrastructure. This suggests that whereas proximity to public transport matters, it is not a sufficient condition for higher public transport use. Another point to realise is that not all public transport is created equal. Accessibility to higher-order public transport, either rail or bus, but having a dedicated right-of-way, could be an essential factor in determining the extent of public transport mode share. Consider that the highest public transport mode shares are reported for Toronto, Montréal, and Vancouver, 335

Murtaza Haider and Ahmed El-Geneidy

Figure 23.6 Proximity to public transport infrastructure in large CMAs in Canada

which are the three cities with rail-based public transport operating on dedicated rights-of-way. Montréal and Toronto are served by underground rail in their core municipalities, whereas Vancouver is served by surface rail operating on the dedicated right-of-way. Following these three cities in public transport mode share is Ottawa, which until recently operated one of the most successful bus rapid transit systems in North America, with an overall public transport mode split for commute trips of about 19%. Recently, Ottawa has replaced its bus rapid transit system with a rail-based system. Following Ottawa is Calgary, which also operates a well-subscribed surface rail system that serves part of the urban core (see also Chapter 13). Figure 23.6 shows that Winnipeg reports the highest accessibility to public transport, but that it is not among the cities with higher public transport mode points to the fact that accessibility to any type of public transport is not sufficient. Instead, proximity to higher-order public transport that offers competitive or near competitive travel times to automobile is seen to have a higher payoff for public transport mode share. Since the public transport mode share reported here is for work trips only, one would expect that higher accessibility to employment destinations should correlate with higher public transport use. Figure 23.7 offers corroborative evidence. Cities with higher accessibility to employment are also the ones with the highest public transport mode shares. However, this relationship does not hold for all cities, where Calgary and Winnipeg report higher accessibility to employment, yet higher-order public transport facilities in Ottawa are partly responsible for higher public transport mode split. The numerous proxies for the built environment, including measures of compactness, diversity, and accessibility, may be collapsed into one aggregate measure to classify places. As mentioned earlier, Statistics Canada developed eight distinct metrics to capture proximity to destinations, including employment, public transport neighbourhood parts, educational facilities, grocery stores, and pharmacies. In addition, Statistics Canada combined these disaggregate metrics into one aggregate measure as an ordinal variable categorising each neighbourhood as being a high, low, or medium amenity-dense neighbourhood.6 Figure 23.8 presents the map for Vancouver, where each dissemination block has been coded as a high amenity (dense grey-coloured dots), medium amenity (dark grey), and low amenity 336

The built environment

Figure 23.7 Proximity to employment in large CMAs in Canada

Figure 23.8 Ordinal depiction of neighbourhoods being of high (light-grey clustered), medium (dark grey), or low (scattered dots [grey]) amenity density

337

Murtaza Haider and Ahmed El-Geneidy

Figure 23.9 Categorising almost 500,000 neighbourhoods across Canada into three amenity density clusters

density (sparse grey-coloured dots). Of the vast area that makes up Vancouver, only a small segment of neighbourhoods is categorised as high-amenity (dense grey-coloured dots) neighbourhoods. Most neighbourhoods are categorised as medium (dark grey) or low (sparse greycoloured dots) amenity density. If a public transport-friendly city like Vancouver could boast only a small minority of its neighbourhoods as high amenity density, what can one say of large and small towns whose built environment is not compact and public transport mode share is considerably small? Another relevant question to pose is what segment of the population lives in high amenity density neighbourhoods. Since the Proximity Measures Database covers all dissemination blocks across Canada, one may try to answer this question by aggregating the population for dissemination blocks per their categorisation for amenity density. The results are presented in Figure 23.9. Primarily, across Canada, fewer than 3% of Canadians live in areas that could be categorised as amenity-dense neighbourhoods. An overwhelming majority of Canadians making up more than 80% of the population reside in neighbourhoods that are categorised as lower amenity density. One can infer from these results the potential for public transport use across Canada. The current built environment of neighbourhoods across Canada such that more than 80% of the population resides in places where the built environment does not meet the prescribed standards for supporting reliable and efficient public transport. The fewer than 20% of Canadians who live in medium amenity density neighbourhoods are the ones who could be targeted for improvement in public transport mode share.

Conclusions This chapter has explored the relationship between the built environment and public transport use. The near-consensus in the reviewed literature is that higher density, diversity, and other urban design dimensions of the built environment are associated with higher use of public transport. However, no one factor is a sufficient determinant of higher public transport use. At the same time, built environment metrics are correlated with each other such that any association observed between a built environment metric and public transport use might ignore other indicators, not controlled for explicitly in the analysis, which may also be influential and yet 338

The built environment

missing from the analysis. For instance, higher population density neighbourhoods share several other built environment traits, such as higher intersection or road densities, and demographic traits, such as smaller housing units and small-sized households. Thus, population density, if used as an indicator of the built environment in isolation, might reveal a more pronounced impact on public transport usage. In contrast, other supporting built environment attributes, not explicitly controlled following the analysis, may also be influential on the outcome and, when included in the analysis, might reduce the estimated impact of population density. The empirical part of the chapter explores the question of why, even when public transport service is available, a large number of commuters, at times the majority, travel by non-public transport modes. This chapter identifies the average travel time differences between public transport and automobile as a reason for the lower use of public transport. Data from Canada’s most populous cities shows that even for cities where higher-order public transport operating on a dedicated right-of-way exists, average travel times by public transport are considerably longer than those by the private automobile. The empirical analysis presented in the chapter shows that neighbourhood vintage is a critical determinant of the built environment that consequently influences public transport mode share. A comparison of public transport use in Montréal, Canada’s second-most-populous urban centre, also known for higher public transport mode split for commutes, and Calgary, which is not known as a public transport-oriented city, showed significant differences in their built environment and public transport use. Such comparisons, though, ignore the differences in vintage between the two cities. Future comparisons of cross-jurisdictional differences in built environment attributes and public transport use must also be informed by the differences in the vintage of neighbourhoods and cities. Public policy recommendations emerging from research on the linkages between the built environment and public transport use often assume that the existing built environment characteristics, if not supportive of higher public transport use, could be modified. For instance, when researchers find higher densities to be correlated with higher public transport use, their findings imply that population densities be increased in low-density areas above the minimum threshold needed for operating frequent public transport service. Such recommendations are based on the assumption that the built environment of existing neighbourhoods could be modified. However, such interventions are costly for political and financial reasons. Hence, examples of changes in the built environment of existing neighbourhoods are quite rare. Hence, recommendations for a higher population density or compact built form to support mobility by public transport will be more useful for planned developments.

Future research Research on this chapter was initiated before the onset of COVID-19. Since March  2020, restrictions on mobility and assembly have disrupted all forms of local, regional, and international travel. Even in the cities known for higher use of public transport, ridership has declined precipitously, especially during the COVID-19-mandated lockdowns. International travel, even in August 2020, is allowed only under exceptional circumstances. Whereas restrictions on urban mobility have been relaxed, and public transport agencies can operate service, public transport ridership in developed economies is only a fraction of its pre-pandemic levels (Verma, 2020). A related response to COVID-19 has been a growing preference for suburban locations where larger homes are available at relatively lower prices than the urban core. If the increase in the preference for suburbs continues, even at a slower rate, suburban population growth will imply even a larger share of the population residing in public transport-deficient areas. 339

Murtaza Haider and Ahmed El-Geneidy

At the same time, travel to and from employment hubs in downtown, which are often served efficiently by public transport, has declined significantly because of COVID-19. Even though workers are now being allowed to return to their offices, the uptake is significantly less than anticipated, which further contributes to a decline in travel by public transport to destinations that have been served efficiently by public modes of transportation. Future research exploring the relationship between the built environment and public transport must account for the aforementioned changes in the socioeconomic spheres resulting from the restrictions related to COVID-19. In addition to COVID-19, other technological innovations in transportation are also influencing the way people plan and execute urban travel. The expected arrival of fully autonomous vehicles in the future and the rise of ride-hailing apps and services – for example, Uber and Lyft – have already started to impact how people travel (Tirachini & del Río, 2019). Technology-driven innovations are influencing the relationship between the built environment and public transport. In the future, research in public transport must be mindful of the technological and intergenerational shifts in tastes, where millennials have shown diverging behaviours for home and automobile ownership compared to the generations that preceded them (Klein & Smart, 2017). Previous research has often adopted cross-sectional approaches to study the relationship between the built environment and public transport. Future research, though, should focus on longitudinal analysis to improve the understanding of the dynamic relationship between the built environment and public transport. Furthermore, interjurisdictional comparisons of public transport use should account for the vintage of built form, which, to a large extent, is determined by the prevalent modes of transportation. The relationship between the built environment and public transport is not stationary. It has evolved and manifested differently across divergent urban spaces. Changes in travel behaviour will be better understood with a broader research agenda to embrace the traditional and emerging determinants of travel behaviour and land use.

Notes 1 2 3 4 5 6

https://www23.statcan.gc.ca/imdb-bmdi/instrument/5178_Q2_V1-eng.pdf https://www150.statcan.gc.ca/n1/pub/92-195-x/2011001/geo/cma-rmr/cma-rmr-eng.htm. https://www150.statcan.gc.ca/n1/pub/92-195-x/2011001/geo/ct-sr/def-eng.htm https://www150.statcan.gc.ca/n1/pub/17-26-0002/172600022020001-eng.htm https://www12.statcan.gc.ca/census-recensement/2016/ref/dict/geo014-eng.cfm https://www150.statcan.gc.ca/n1/pub/17-26-0002/172600022020001-eng.htm

References Adams, M. A., Frank, L. D., Schipperijn, J., Smith, G., Chapman, J., Christiansen, L. B., . . . Sallis, J. F. (2014). International variation in neighborhood walkability, transit, and recreation environments using geographic information systems: The IPEN adult study. International Journal of Health Geographics, 13(43), 1–17. doi:10.1186/1476-072X-13-43 Cao, X., Handy, S. L., & Mokhtarian, P. L. (2006). The influences of the built environment and residential self-selection on pedestrian behavior: Evidence from Austin, TX. Transportation, 33(1), 1–20. Cao, X., Mokhtarian, P. L., & Handy, S. L. (2009). Examining the impacts of residential self-selection on travel behaviour: A focus on empirical findings. Transport Reviews, 29(3), 359–395. Cervero, R., & Duncan, M. (2008). Residential self selection and rail commuting: A nested logit analysis. University of California Transportation Center, Working Papers, University of California Transportation Center. https://escholarship.org/uc/item/72p9n6qt Cervero, R., & Kockelman, K. (1997). Travel demand and the 3Ds: Density, diversity, and design. Transportation Research Part D-Transport and Environment, 2(3), 199–219.

340

The built environment Cervero, R.,  & Landis, J. (1997). Twenty years of the Bay Area Rapid Transit system: Land use and development impacts. Transportation Research Part A: Policy and Practice, 31(4), 309–333. doi:10.1016/ S0965-8564(96)00027-4 Chen, C., Gong, H.,  & Paaswell, R. (2008). Role of the built environment on mode choice decisions: Additional evidence on the impact of density. Transportation, 35(3), 285–299. doi:10.1007/ s11116-007-9153-5. Dong, H., & Zhu, P. (2015). Smart growth in two contrastive metropolitan areas: A comparison between Portland and Los Angeles. Urban Studies, 52(4), 775–792. doi:10.1177/0042098014528396. Ewing, R., & Cervero, R. (2017). “Does compact development make people drive less?” The answer is yes. Journal of the American Planning Association. American Planning Association, 83(1), 19–25. doi:10.108 0/01944363.2016.1245112 Frank, L. D., Sallis, J. F., Conway, T. L., Chapman, J. E., Saelens, B. E., & Bachman, W. (2006). Many pathways from land use to health – Associations between neighborhood walkability and active transportation, body mass index, and air quality. Journal of the American Planning Association. American Planning Association, 72(1), 75–87. Garreau, J. (1991). Edge city: Life on the new frontier. Anchor Books. https://play.google.com/store/books/ details?id=bhubIafmBv8C Glazier, R. H., Creatore, M. I., Weyman, J. T., Fazli, G., Matheson, F. I., Gozdyra, P., . . . Booth, G. L. (2014). Density, destinations or both? A comparison of measures of walkability in relation to transportation behaviors, obesity and diabetes in Toronto, Canada. PLoS ONE, 9(1), e85295. doi:10.1371/ journal.pone.0085295 Haider, M. (2013, July 2). Public transit is better, but cars are faster. The Globe and Mail. Haider, M. (2019). Diminishing returns to density and public transit. In Transport findings. doi:10.32866/10679 Handy, S., Cao, X. Y., & Mokhtarian, P. L. (2006). Self-selection in the relationship between the built environment and walking – Empirical evidence from northern California. Journal of the American Planning Association. American Planning Association, 72(1), 55–74. Kenworthy, J. R., & Laube, F. B. (1999). Patterns of automobile dependence in cities: An international overview of key physical and economic dimensions with some implications for urban policy. Transportation Research Part A: Policy and Practice, 33(7), 691–723. doi:10.1016/S0965-8564(99)00006-3 Klein, N. J., & Smart, M. J. (2017). Millennials and car ownership: Less money, fewer cars. Transport Policy, 53, 20–29. doi:10.1016/j.tranpol.2016.08.010. Li, F. Z., Harmer, P. A., Cardinal, B. J., Bosworth, M., Acock, A., Johnson-Shelton, D., & Moore, J. M. (2008). Built environment, adiposity, and physical activity in adults aged 50–75. American Journal of Preventive Medicine, 35(1), 38–46. doi:10.1016/j.amepre.2008.03.021. Mokhtarian, P. L., & Cao, X. (2008). Examining the impacts of residential self-selection on travel behavior: A focus on methodologies. Transportation Research Part B: Methodological, 42(3), 204–228. Næss, P. (2009). Residential self-selection and appropriate control variables in land use: Travel studies. Transport Reviews, 29(3), 293–324. Newman, P. G., & Kenworthy, J. R. (1989). Cities and automobile dependence: An international sourcebook. https://trid.trb.org/view/351194 Racine, F. (2019). The influence of urban design theories in the transformation of urban morphology: Montreal from 1956 to 2018. Journal of Urban Design, 24(6), 815–839. doi:10.1080/13574809.2019.1 601994 Savage, K. (2019). Results from the 2016 census: Commuting within Canada’s largest cities (Catalogue no. 75-006-X). Statistics Canada. https://www150.statcan.gc.ca/n1/en/pub/75-006-x/2019001/ article/00008-eng.pdf?st=N5gu61LA Statistics Canada. (2020). Proximity measures database. https://www150.statcan.gc.ca/n1/pub/17-26-0002/ 172600022020001-eng.htm Tirachini, A., & del Río, M. (2019). Ride-hailing in Santiago de Chile: Users’ characterisation and effects on travel behaviour. Transport Policy, 82, 46–57. doi:10.1016/j.tranpol.2019.07.008 Verma, P. (2020, August 15). Public transit cuts felt deepest in low-income areas. The New York Times. www.nytimes.com/2020/08/15/us/virus-transit-congress.html Zhao, P. (2014). The impact of the built environment on bicycle commuting: Evidence from Beijing. Urban Studies, 51(5), 1019–1037. doi:10.1177/0042098013494423

341

24 INTELLIGENT MOBILITY AND BIG DATA FOR PLANNING, TRUST, AND PRIVACY Caitlin D. Cottrill

Introduction According to the UK’s Transport Systems Catapult, At its simplest, Intelligent Mobility is about moving people and goods around in an easier, more efficient and more environmentally-friendly way. It uses new ideas and new technologies to look beyond traditional, infrastructure-heavy approaches to transport, and instead to come up with innovative ways to improve mobility and make journeys better and accessible to all. (Transport Systems Catapult, n.d.) While much of the focus of intelligent mobility (IM) is on the technologies that enable efficient movement, the collection, integration, and use of data generated by such technologies are of equal importance. In the case of public transport, a shift towards mobile ticketing and fare payments and app-based journey planners has resulted in the development of rich and robust data resources, which have been used for a variety of purposes, ranging from day-to-day operations to more strategic long-term planning activities (Pelletier et al., 2011). While in aggregate, such data represent a valuable addition to the public transport toolkit, it must also be acknowledged that individual data records possess the ability to directly identify users’ identities and habits as demonstrated through their travel behaviours and the linking of personal financial information. Such potential was demonstrated in 2018, when Public Transport Victoria, Australia, released a dataset containing 1.8  billion travel records covering the period between June 2015 and June 2018. The records, which were generated by 15.1 million users of Victoria’s myki public transport smartcard, contained tap-on and tap-off data (including date, time, and location) from users of the state’s tram, train, and bus network. Researchers from the University of Melbourne were able to use the ‘de-identified’ data (at times combined with other contextual information, such as Twitter posts) to successfully identify a number of travellers, including a Victorian MP (Taylor, 2019; Culnane et al., 2019). It is evident, then, that the use of emerging data resources in public transport offers great promise; however, the protection and security of such data are also critical considerations. In this chapter, the use of big data resources in the public transport sector, with particular focus 342

Intelligent mobility and big data

on methods of collection and areas of use, is first explored. Next, the privacy and data protection implications of such use, including considerations of new regulatory requirements and the importance of trust as a mitigating factor, are addressed. Finally, the chapter discusses how together these considerations may impact the use of data resources in the public transport sphere, and issues related to data privacy in the context of MaaS are highlighted.

The use of data in the public transport sector Effective public transport planning and operations have long relied on an understanding of passenger needs and preferences, routing efficiencies, and origin-destination patterns to determine optimal timing and locations of services. Traditionally, a number of data resources have been used in these planning and operations activities, including household travel surveys, on-board passenger counts, passenger intercept surveys, and, increasingly, forms of automatic vehicle location (AVL) data from vehicles (White, 2016; Behrens & Schalekap, 2010; Richardson et al., 1995; van Oort et al., 2015; Hounsell et al., 2012). While such resources have provided useful and fairly reliable data to be used in ongoing public transport operations, management, and planning, they have often proved insufficient for fully understanding longitudinal aspects of public transport use or linked trips., In addition, these traditional resources have often been temporally deficient in terms of ongoing needs for route modifications due to changing origins and destinations and modification to roadway networks (Bagchi  & White, 2005; Chapleau et al., 2008; Wilson et al., 2008). A fundamental technology that has had myriad impacts in the public transport realm in recent decades are global positioning system (GPS)-enabled devices, such as mobile phones, GPS receivers, in-vehicle navigation systems, and others. Their decreasing cost and increasing uptake has greatly expanded the availability of location-based services (LBS), defined as “computer applications (especially mobile computing applications) that deliver information tailored to the location and context of the device and the user” (Huang et al., 2018). A variety of types of such services have been developed, including: • Orientation and localisation • Navigation • Search • Identification • Event check (Steiniger et al., 2006) In the public transport realm, the use of GPS data has been frequently cited in a number of areas, including providing insights in travel time variability (Mazloumi et al., 2010), improving public transport journey planners (Allulli et al., 2014), and using collected data to improve policy and planning decisions (Gschwender et al., 2016). In addition, standalone or smartphonebased GPS devices are increasingly used in household travel surveys, which has been seen to improve the robustness of data collected on public transport use, including access between the trip origin and public transport stop and the point of disembarkment and the destination (Bohte & Maat, 2009). GPS has provided a solid basis for the collection of location data; however, in recent years, the data landscape for public transport has widened even further, largely as a result of the increasing use of both smart cards and location-sensing mobile apps (see also Chapter 36). Smart cards provide an alternative medium to cash for public transport fare payments, a 343

Caitlin D. Cottrill

function which is particularly important during the COVID-19 pandemic. According to Bagchi and White (2005), Each smart card can be identified by a unique serial number. The cards can be registered to a given individual, or they can be anonymous. On these cards can be placed electronically a range of fare options such as travelcards or stored value (a monetary amount credited to the card which is debited as and when journeys are made). (p. 464) Depending on the system, users may ‘tap-in’ to the system to record a flat fare payment, or ‘tap-in’ and ‘tap-out’ for distance or zone-based fares. For each tap-in or -out, the following data are generated: card number, date, time, validation status, and stop number (Agard et al., 2006). In addition, a determination is made if there are adequate funds available to pay for the trip and, if so, a deduction is made. In some cases, users may set up their cards to allow for direct debit from a linked financial account when funds are running low (Mezghani, 2008). Though many public transport agencies are moving away from bespoke, single-purpose smart cards and toward electronic payment using general debit or credit cards (Keitel, 2011), the expectations remain for public transport operators to have access to the relevant data generated. Regarding mobile apps, cellular phones are some of the most ubiquitous devices in the world, and smartphones (defined as mobile phones with advanced computing capabilities) have quickly gained a large share of the market. Though adoption rates vary, largely according to national development levels (according to research conducted by the Pew Research Center, “a median of 76% across 18 advanced economies surveyed have smartphones, compared with a median of only 45% in emerging economies” [Tyler & Silver, 2019]), the rapidity of smartphone penetration rates has made them a key component of emerging public transport business models, discussed subsequently. While there is no comprehensive estimate of the percentage of public transport systems that make use of smartphone apps, a 2017 study from Deloitte reported that of survey participants with a smartphone, 16% of persons in developing and 18% of those in developed countries had paid for public transport with a smartphone (Wigginton et al., 2017), indicating a degree of adoption worldwide. In terms of functionality, Gössling (2018) undertook an extensive review of transport- and mobility-related apps, finding the following represented (amongst others): travel information, planning, and routing; service sharing; payment and price comparisons; and participatory transport systems. More recently, Mobility as a Service (or MaaS) apps have emerged in the marketplace and aim at their simplest level to allow users to ‘bundle’ their mobility choices using a single app for planning, purchase, and delivery of multi-modal tickets and services (Li & Voege, 2017; Wong et al., 2020, see also Chapter 3). The data collected via use of smart cards and smartphone apps and services have been demonstrated to have great benefits for the public transport sector. Creating more detailed, robust, and timely data on passenger movements, adherence to timetables, and information requests contributes to the overall potential for public transport service providers to more effectively meet the needs of their customers. Smartphone-generated location data can additionally be used for operational purposes (particularly in the absence of AVL data) and to support the provision of real-time passenger information (Corsar et al., 2013). Such applications also enhance the potential for providers to maximise the potential efficacy of public transport services (Schmitz et al., 2016; Edwards et al., 2011; Simonyi et al., 2014). Access to more detailed, accurate, and, potentially, representative data streams may also contribute to efforts to better respond to a

344

Intelligent mobility and big data

greater variety of user needs (Filippi et al., 2013). Given these benefits, it is evident why public transport agencies worldwide have begun to focus more time, staff, and resources on implementing smart card-based fare payment systems and developing smartphone-based technologies that provide continuing and comprehensive data streams on the actions and events of persons and places.

Data and privacy The benefits of the technologies and the data they generate discussed previously are myriad. However, as noted in the introduction, they are not without concern. A key issue here is privacy – as noted in a report commissioned by the US Federal Highways Administration: With many transportation apps tracking and storing sensitive information, such as user location data, concerns are emerging on how to safeguard user privacy . . . and on how this might impact usage of transportation apps and services. . . . Community-based navigation app, Waze; public transport app, Moovit; and community-based running and cycling app, Strava, are transportation apps revolving around the collection and analysis of user data. They raise questions about how these data are being shared, particularly as the three apps partner and share data with cities. (Shaheen et al., 2016) Data generated through dedicated GPS receivers, smart cards, and mobile apps are highly identifiable. The presence of unique identifiers (such as a smart card or mobile identification numbers) combined with robust and detailed location data indicates that without adequate data and privacy protection, the potential for users’ private data to be detected or inferred is substantial (Narayanan et al., 2016; Rossi et al., 2015; Riederer et al., 2016). In recent years, the public have become more aware of the potential for privacy violations to occur in the mobility sector, largely as a result of incidents such as the myki data release discussed previously; a 2017 reveal that the ride-hailing company Uber had been tracking its customers for up to five minutes following completion of a ride (Wamsley, 2017); and a 2018 investigative report by the Associated Press that revealed that many Google services on Android devices and iPhones store users’ location data, even if the user has enabled a privacy setting that is intended to prevent Google from doing so (Nakashima, 2018). Such violations, and their reporting in the popular media, have begun to impact user awareness and perceptions of privacy, with a 2019 report undertaken by the location data technology company Blis finding that, “Nearly two in three [American] consumers are more aware of how their personal information is being used today than they were just a year ago, and 83% of people are aware that their location is tracked” (Blis, 2019). Given such awareness, it is unsurprising that previous research has indicated that privacy concerns can be a barrier to the intent to download and use mobile apps (Gu et al., 2017) and that privacy expectations may be violated by mobile app practices (Lin et al., 2012). More specifically, with reference to travel apps, Dastjerdi et al. (2019) found that information privacy concerns contributed negatively to the likelihood of Danish consumers creating a registered account that would utilise personal information and location tracking to provide incentives and motivations to engage in ‘green’ travel. In this climate, a number of new policies, regulations, and directives have emerged that directly address data privacy. While perhaps the most notable is the EU’s General Data Protection Regulation [GDPR–(EU) 2016/679], which became enforceable in May  2018 and

345

Caitlin D. Cottrill

superseded the EU’s Data Protection Directive (95/46/EC), a number of others have also been enacted worldwide. Determann (2020) notes, In June  2018, California enacted the California Consumer Privacy Act (CCPA), a novel and extremely broad and prescriptive law against data selling . . . Maine and Nevada already followed with their own state laws against data selling. In August 2018, India introduced a new bill and Brazil enacted its first data protection law, both modelled after the GDPR. According to the United Nations Conference on Trade and Development (UNCTAD), as of 2020, “107 countries (of which 66 were developing or transition economies) have put in place legislation to secure the protection of data and privacy” (UNCTAD, 2020). Of particular note in these developments is that location data are increasingly considered to be ‘personal’ data. For example, under Article 4 of the GDPR: “Personal data” means any information relating to an identified or identifiable natural person (‘data subject’); an identifiable natural person is one who can be identified, directly or indirectly, in particular by reference to an identifier such as a name, an identification number, location data, an online identifier or to one or more factors specific to the physical, physiological, genetic, mental, economic, cultural or social identity of that natural person. (General Data Protection Regulation, 2016) The inclusion of both location data and online identifiers in the definition of ‘personal’ data indicates recognition of the ability of these data resources to identify individuals, and is generating increasing awareness in the public transport industry (Cottrill, 2020; Sørensen & Kosta, 2019). The importance of this was also acknowledged in a 2018 report from the Open Data Institute, which stated, journey data is transforming relationships between companies and customers, drastically increasing the importance of trust, and raising critical questions of ethics, equity and engagement which cannot go unanswered. The impact of these trends will only increase as people grow more aware of these issues and as they gain more rights over data about them (p. 4) Under such circumstances, issues of transparency and communication of data practices in public transport, as well as trust in the system and providers, are critical to address.

Transparency, communication, and trust The concept of transparency is critical in terms of data privacy and protection, with the GDPR defining it as follows: “The principle of transparency requires that any information addressed to the public or to the data subject be concise, easily accessible and easy to understand, and that clear and plain language and, additionally, where appropriate, visualisation be used” (General Data Protection Regulation, 2016). Of note is that methods of communication are also critical in terms of ensuring transparency, as many data privacy policies and practices have

346

Intelligent mobility and big data

been criticised for being overly technical, unclear in application, or time-consuming to read (McDonald & Cranor, 2008; Cottrill & Thakuriah, 2014; Lederman et al., 2016). In terms of both transparency and communication, Kitchin (2016) uses Transport for London (TfL), which oversees the provision of public transport in London, as an example of good practice, stating that, “TfL has adopted a transparent approach to data privacy and data protection policies, which are published on their website. These policies are short, clear and unambiguous, written in plain English that avoids dense legal language” (p. 57). Given the complexities of technologies and related data described previously, such approaches are becoming increasingly important, particularly given the influence of trust in the willingness of consumers to utilise public transport technologies such as smart cards and mobile apps. Trust has been indicated as a mediating factor in privacy concerns by researchers such as Joinson et al. (2010), Miltgen and Smith (2015), and Kehr et al. (2015). In general, researchers have found that as trust in the collecting agency grows, so too does the willingness of consumers to provide data. This indicates a need for public transport agencies to ensure that they continue to be trusted to obtain and use private data through approaches such as privacy-by-design (Cavoukian, 2009; Avoine et al., 2014) in development of data-collecting technologies and transparency in their data practices. Finally, it is critical that public transport agencies be transparent not only about their collection of data but also in how they are used. Research has indicated that consumers are more likely to feel comfortable with the idea of sharing personal or potentially identifiable data when they see a clear nexus between these data and the intent of use, particularly if this may confer direct benefits to them (Cottrill & Thakuriah, 2015). In the case of public transport, providing users with clear, easily understandable information regarding what collected data may be used for (e.g. route planning, service provision, or modification of schedules) may provide adequate impetus for sharing to take place.

Mobility as a Service in the spotlight Issues of data privacy are particularly relevant in the case of Mobility as a Service, as demonstrated in Cottrill (2020). Given the multiplicity of actors involved in MaaS systems (including transport operators, data providers, technology and platform providers, insurance companies, and regulatory organisations [Kamargianni & Matyas, 2017]), providing transparency regarding how data resources are collected and shared is of critical importance. Data required for MaaS systems to function are likewise wide ranging, including not only those related to the transport system and public transport options (including where customers can access/egress transport assets and services, pricing information, asset characteristics, routes, and timetables [Datson, 2016; Kamargianni  & Goulding, 2018]) but also user data, including origin and destination locations, acceptable transport options, and financial data for payments. Such a complex ecosystem of data and actors results in increased potential for privacy violations to occur, making data and privacy protection key considerations. When developing MaaS systems, utilising a privacy-by-design approach will be helpful in ensuring trust of both customers and involved operators, as it will provide for considerations of how the system is enacted with respect to data sharing and use across involved actors. Utilising a trusted third party for data storage may be considered here, as this will provide protections in terms of access and data linkage, thus minimising the potential for privacy violations to occur. Ensuring that relevant data protections are built into the system and clearly and transparently presented to users will further enhance trust and respond to applicable regulation (such as GDPR in European contexts).

347

Caitlin D. Cottrill

Conclusion As public transport increasingly moves towards the use of digital technologies for providing passenger information, the purchase of tickets and fares, and collecting information for planning purposes, it is important to consider the implications of the data generated on consumer privacy and trust. While these technologies have the potential to generate significant benefits in terms of improving service provision and the ease of customer travel, it is critical to ensure that the data practices of providers are clearly communicated to the user in plain, non-technical language; that data protection is built in to the development of systems; and that data uses are aligned with customer expectations. By taking such precautions, it is anticipated that consumers will continue to view public transport agencies with trust and be willing participants in the intelligent mobility ecosystem. As the use of technology increases in the public transport sector, further research will be required on how best to build in privacy protections that consider both regulatory requirements and the enhancement of trust by travellers. In the wake of the COVID-19 pandemic, issues of data privacy and protection have become only more critical. With an increasing number of countries and governments at all levels looking to contact-tracing smartphone applications to help manage the spread of the virus, considerations of how data will be treated and protected following collection have not always been fully addressed. There are clear parallels seen in privacy and data protection within the public transport realm, particularly given the spatio-temporal richness of data collected. In addition to generalised contact-tracing apps, some countries have also implemented measures for specifically tracking public transport journeys, given concerns of the close proximity of passengers and drivers in enclosed spaces. In China, for example, passengers are required to scan a QR code for entrance onto public transport vehicles, which both indicates their ability to travel under China’s ‘traffic light’ system (with green, amber, and red codes indicating likely health status) and allows tracking of public transport trips (Law, 2020). In many locations, the use of cash for fare payment is also being discouraged, and passengers are encouraged to book travel in advance to limit crowding on vehicles (UK Department for Transport, 2020; Australian Government Department of Health, 2020). While such measures are appropriate in the context of the virus, they also have the potential to introduce additional personal data into the public transport ecosystem, and how these data are treated and protected requires additional consideration.

References Agard, B., Morency, C., & Trépanier, M. (2006). Mining public transport user behaviour from smart card data. IFAC Proceedings Volumes, 39(3), 399–404. doi:10.3182/20060517-3-FR-2903.00211 Allulli, L., Italiano, G. F., & Santaroni, F. (2014). Exploiting GPS data in public transport journey planners. In J. Gudmundsson & J. Katajainen (Eds.), Experimental algorithms. SEA 2014: Lecture notes in computer science (pp. 295–306). Springer. doi:10.1007/978-3-319-07959-2_25 Australian Government Department of Health. (2020). Principles for COVID-19 public transport operations. Retrieved July  20, 2020, from https://www.infrastructure.gov.au/transport/files/covid19_public_ transport_principles_29052020.pdf Avoine, G., Calderoni, L., Delvaux, J., Maio, D., & Palmieri, P. (2014). Passengers information in public transport and privacy: Can anonymous tickets prevent tracking? International Journal of Information Management, 34(5), 682–688. doi:10.1016/j.ijinfomgt.2014.05.004 Bagchi, M., & White, P. R. (2005). The potential of public transport smart card data. Transport Policy, 12(5), 464–474. doi:10.1016/j.tranpol.2005.06.008 Behrens, R., & Schalekap, H. (2010). Public transport mode satisfaction in Cape Town. Findings of passengers intercept survey. Southern African Transport Conference Proceedings, Proceeding (No. 29). Blis. (2019). The ‘currency’ of data: Quantifying the value of consumer information in 2019. Blis.

348

Intelligent mobility and big data Bohte, W., & Maat, K. (2009). Deriving and validating trip purposes and travel modes for multi-day GPSbased travel surveys: A large-scale application in the Netherlands. Transportation Research Part C: Emerging Technologies, 17(3), 285–297. doi:10.1016/j.trc.2008.11.004 Cavoukian, A. (2009). Privacy by design: The 7 foundational principles. Information and privacy commissioner of Ontario, Canada. www.ipc.on.ca/wp-content/uploads/resources/7foundationalprinciples.pdf Chapleau, R., Trépanier, M., & Chu, K. K. (2008, May 25–31). The ultimate survey for transit planning: Complete information with smart card data and GIS. Paper presented at the 8th international conference on survey methods in transport: Harmonisation and data comparability, Annecy, France. Corsar, D., Edwards, P., Baillie, C. C., Markovic, M., Papangelis, K., & Nelson, J. D. (2013). GetThere: A rural passenger information system utilising linked data & citizen sensing. In Proceedings of the 12th international semantic web conference (ISWC 2013) posters & demonstrations track. CEUR-WS. Cottrill, C. D. (2020). MaaS surveillance: Privacy considerations in mobility as a service. Transportation Research Part A: Policy and Practice, 131, 50–57. doi:10.1016/j.tra.2019.09.026 Cottrill, C. D., & Thakuriah, P. V. (2014). Privacy in context: An evaluation of policy-based approaches to location privacy protection. International Journal of Law and Information Technology, 22(2), 178–207. doi:10.1093/ijlit/eat014 Cottrill, C. D., & Thakuriah, P. V. (2015). Location privacy preferences: A survey-based analysis of consumer awareness, trade-off and decision-making. Transportation Research Part C: Emerging Technologies, 56, 132–148. doi:10.1016/j.trc.2015.04.005 Culnane, D., Rubinstein, A., Benjamin, I. P., & Teague, A. (2019). Stop the open data bus, we want to get off. arXiv preprint arXiv:1908.05004 Dastjerdi, A. M., Kaplan, S., e Silva, J. D. A., Nielsen, O. A., & Pereira, F. C. (2019). Participating in environmental loyalty program with a real-time multimodal travel app: User needs, environmental and privacy motivators. Transportation Research Part D: Transport and Environment, 67, 223–243. doi:10.1016/j. trd.2018.11.013 Datson, J. (2016). Mobility as a service: Exploring the opportunity for mobility as a service in the UK. Transport Systems Catapult. https://ts.catapult.org.uk/wp-content/uploads/2016/07/Mobility-as-a-Service_ Exploring-the-Opportunity-for-MaaS-in-the-UK-Web.pdf Determann, L. (2020). Determann’s field guide to data privacy law: International corporate compliance. Edward Elgar Publishing. Edwards, P., Corsar, D., Nelson, J., Velaga, N., Beecroft, M., Pan, J., & Sripada, S. (2011). Building an information ecosystem for public transport in rural areas. Proceedings of Digital Engagement. Filippi, F., Fusco, G., & Nanni, U. (2013). User empowerment and advanced public transport solutions. Procedia-Social and Behavioral Sciences, 87, 3–17. doi:10.1016/j.sbspro.2013.10.590 General Data Protection Regulation. (2016). Regulation (EU) 2016/679 of the European Parliament and of the Council of 27 April 2016 on the protection of natural persons with regard to the processing of personal data and on the free movement of such data, and repealing Directive 95/46. Gössling, S. (2018). ICT and transport behavior: A conceptual review. International Journal of Sustainable Transportation, 12(3), 153–164. doi:10.1080/15568318.2017.1338318 Gschwender, A., Munizaga, M., & Simonetti, C. (2016). Using smart card and GPS data for policy and planning: The case of Transantiago. Research in Transportation Economics, 59, 242–249. doi:10.1016/j. retrec.2016.05.004 Gu, J., Xu, Y. C., Xu, H., Zhang, C.,  & Ling, H. (2017). Privacy concerns for mobile app download: An elaboration likelihood model perspective. Decision Support Systems, 94, 19–28. doi:10.1016/j. dss.2016.10.002 Hounsell, N. B., Shrestha, B. P., & Wong, A. (2012). Data management and applications in a world-leading bus fleet. Transportation Research Part C: Emerging Technologies, 22, 76–87. doi:10.1016/j.trc.2011.12.005 Huang, H., Gartner, G., Krisp, J. M., Raubal, M., & Van de Weghe, N. (2018). Location based services: Ongoing evolution and research agenda. Journal of Location Based Services, 12(2), 63–93. doi:10.1080/ 17489725.2018.1508763 Joinson, A. N., Reips, U. D., Buchanan, T., & Schofield, C. B. P. (2010). Privacy, trust, and self-disclosure online. Human–Computer Interaction, 25(1), 1–24. doi:10.1080/07370020903586662 Kamargianni, M., & Goulding, R. (2018). The mobility as a service maturity index: Preparing the cities for the mobility as a service era. Transport Research Arena, 7, 1–10. Zenodo. doi:10.5281/zenodo.1485002 Kamargianni, M., & Matyas, M. (2017). The business ecosystem of mobility-as-a-service. 96th Transportation Research Board (TRB) Annual Meeting. Washington, DC.

349

Caitlin D. Cottrill Kehr, F., Kowatsch, T., Wentzel, D., & Fleisch, E. (2015). Blissfully ignorant: The effects of general privacy concerns, general institutional trust, and affect in the privacy calculus. Information Systems Journal, 25(6), 607–635. doi:10.1111/isj.12062 Keitel, P. L. (2011). The electronification of transit fare payments: Examining the case for partnerships between payments firms and transit agencies. FRB of Philadelphia-Payment Cards Center Discussion Paper. Kitchin, R. (2016). Getting smarter about smart cities: Improving data privacy and data security. Data Protection Unit, Department of the Taoiseach. http://mural.maynoothuniversity.ie/7242/ Law, E. (2020, April  24). Coronavirus: China’s contact tracing app touted as helping to contain outbreak. The Straits Times. Retrieved July 20, 2020, from https://www.straitstimes.com/asia/east-asia/ coronavirus-chinas-contact-tracing-app-touted-as-helping-to-contain-outbreak Lederman, J., Taylor, B. D., & Garrett, M. (2016). A private matter: The implications of privacy regulations for intelligent transportation systems. Transportation Planning and Technology, 39(2), 115–135. doi: 10.1080/03081060.2015.1127537 Li, Y., & Voege, T. (2017). Mobility as a service (MaaS): Challenges of implementation and policy required. Journal of Transportation Technologies, 7(2), 95–106. doi:10.4236/jtts.2017.72007 Lin, J., Amini, S., Hong, J. I., Sadeh, N., Lindqvist, J., & Zhang, J. (2012, September). Expectation and purpose: Understanding users’ mental models of mobile app privacy through crowdsourcing. Proceedings of the 2012 ACM Conference on Ubiquitous Computing, pp. 501–510. doi:10.1145/2370216.2370290 Mazloumi, E., Currie, G.,  & Rose, G. (2010). Using GPS data to gain insight into public transport travel time variability. Journal of Transportation Engineering, 136(7), 623–631. doi:10.1061/(ASCE) TE.1943-5436.0000126 McDonald, A. M., & Cranor, L. F. (2008). The cost of reading privacy policies. I/S: A Journal of Law and Policy for the Information Society, 4, 543–568. Mezghani, M. (2008). Study on electronic ticketing in public transport. European Metropolitan Transport Authorities (EMTA), European Metropolitan Transport Authorities (EMTA). Miltgen, C. L., & Smith, H. J. (2015). Exploring information privacy regulation, risks, trust, and behavior. Information & Management, 52(6), 741–759. doi:10.1016/j.im.2015.06.006 Nakashima, R. (2018). AP Exclusive: Google tracks your movements, like it or not. Retrieved January  14, 2020, from https://apnews.com/828aefab64d4411bac257a07c1af0ecb/AP-Exclusive:-Google-tracksyour-movements,-like-it-or-not Narayanan, A., Huey, J., & Felten, E. W. (2016). A precautionary approach to big data privacy. In Data protection on the move (pp. 357–385). Springer. doi:10.1007/978-94-017-7376-8_13 Open Data Institute. (2018). Personal data in transport: Exploring a framework for the future. Open Data Institute. Pelletier, M. P., Trépanier, M.,  & Morency, C. (2011). Smart card data use in public transit: A  literature review. Transportation Research Part C: Emerging Technologies, 19(4), 557–568. doi:10.1016/j. trc.2010.12.003 Richardson, A. J., Ampt, E. S., & Meyburg, A. H. (1995). Survey methods for transport planning. Eucalyptus Press. Riederer, C., Kim, Y., Chaintreau, A., Korula, N., & Lattanzi, S. (2016). Linking users across domains with location data: Theory and validation. Proceedings of the 25th International Conference on World Wide Web, ACM, pp. 707–719. Rossi, L., Walker, J., & Musolesi, M. (2015). Spatio-temporal techniques for user identification by means of GPS mobility data. EPJ Data Science, 4(1), 1–16. doi:10.1140/epjds/s13688-015-0049-x Schmitz, C., Bartsch, S., & Meyer, A. (2016). Mobile app usage and its implications for service management – Empirical findings from German public transport. Procedia-Social and Behavioral Sciences, 224, 230–237. doi:10.1016/j.sbspro.2016.05.492 Shaheen, S. A., Cohen, A. P., Zohdy, I. H., & Kock, B. (2016). Smartphone applications to influence travel choices: Practices and policies. Federal Highway Administration. Simonyi, E., Fazekas, Z., & Gáspár, P. (2014). Smartphone application for assessing various aspects of urban public transport. Transportation Research Procedia, 3, 185–194. doi:10.1016/j.trpro.2014.10.104 Sørensen, J., & Kosta, S. (2019). Before and after GDPR: The changes in third party presence at public and private European websites. The World Wide Web Conference, pp. 1590–1600. doi:10.1145/3308558.3313524 Steiniger, S., Neun, M., & Edwardes, A. (2006). Foundations of location based services. Lecture Notes on LBS, 1(272), 1–28. Taylor, J. (2019, August 15). Myki data release breached privacy laws and revealed travel histories, including of Victorian MP. Retrieved June  20, 2020, from www.theguardian.com/australia-news/2019/aug/15/ myki-data-release-breached-privacy-laws-and-revealed-travel-histories-including-of-victorian-mp

350

Intelligent mobility and big data Transport Systems Catapult. (n.d.). What is intelligent mobility? Retrieved January 10, 2020, from https:// ts.catapult.org.uk/intelligent-mobility/introduction/ Tyler, B., & Silver, L. (2019). Smartphone ownership is growing rapidly around the world, but not always equally. Retrieved June  20, 2020 from www.pewresearch.org/global/2019/02/05/smartphone-ownershipis-growing-rapidly-around-the-world-but-not-always-equally/ UK Department for Transport (2020). Coronavirus (COVID-19): safer travel guidance for passengers. Retrieved 20 July  2020 from https://www.gov.uk/guidance/coronavirus-covid-19-safer-travel-gui dance-for-passengers UNCTAD. (2020). Data protection and privacy legislation worldwide. Retrieved February  25, 2020, from https://unctad.org/en/Pages/DTL/STI_and_ICTs/ICT4D-Legislation/eCom-Data-ProtectionLaws.aspx van Oort, N., Sparing, D., Brands, T., & Goverde, R. M. (2015). Data driven improvements in public transport: The Dutch example. Public Transport, 7(3), 369–389. doi:10.1007/s12469-015-0114-7 Wamsley, L. (2017, August 29). Uber ends its controversial post-ride tracking of users’ location. Retrieved January 12, 2020, from www.npr.org/sections/thetwo-way/2017/08/29/547113818/uber-ends-its-controversialpost-ride-tracking-of-users-location White, P. (2016). Public transport: Its planning, management and operation. Taylor & Francis. Wigginton, C., Curran, M., & Brodeur, C. (2017). Global mobile consumer trends. Deloitte Touche Tohmatsu Limited. Wilson, N. H., Zhao, J., & Rahbee, A. (2008). The potential impact of automated data collection systems on urban public transport planning. In N. H. Wilson & A. Nuzzolo (Eds.), Schedule-based modeling of transportation networks (pp. 75–99). Springer Science & Business Media. Wong, Y. Z., Hensher, D. A., & Mulley, C. (2020). Mobility as a service (MaaS): Charting a future context. Transportation Research Part A: Policy and Practice, 131, 5–19. doi:10.1016/j.tra.2019.09.030

351

PART D

Specific public transport delivery issues

25 THE PROVISION OF SERVICE INFORMATION FOR PUBLIC TRANSPORT Nigel Halpern

Introduction The provision of information is of great importance for public transport. It can help to achieve objectives related to creating awareness; informing target markets; influencing attitudes and behaviour; and encouraging preference, repeat business, and loyalty. Depending on the objectives, information can be about the service or the brand. Service information primarily relates to the journey (e.g. to inform target markets about routes, schedules, pricing), while brand information can also be non-journey related (e.g. to raise the profile or image of the service, persuade passengers to switch from alternative transport, encourage loyalty). The main focus of this chapter is on the provision of service information. This is closely linked to marketing communications, which is concerned with the messages and media used to communicate with target markets. Messages are essentially the information that is provided, while media are the tools used to store and deliver information. In a public transport context, service-related messages typically include information on travel planning, routes and schedules, tickets and prices, service changes, disruption and delays, and customer services and practical information such as Wi-Fi access and onboard entertainment; food and drink; accessibility and assistance; and travelling with animals, baggage, or special items. Media typically includes signage, printed materials, information kiosks or displays, television, radio, telephone, newspapers, newsletters, exhibitions or visits, help points, information offices, call centres, ticket offices or machines, public address systems, websites, social media, and mobile technologies. This is by no means an exhaustive list of messages and media. However, it provides an indication of the many options available. In the past, service information was only really provided to broad audiences via traditional messages and media, for instance, printed route maps and published timetables and fares. This is static information because it is only updated occasionally and when changes to the service are made. Traditional messages and media are still widely used. However, they have been increasingly replaced with new approaches. For instance, Mulley et al. (2017) explain how Sydney trains have increased the use of digital versus printed messages and media. This is because the communications landscape has changed dramatically during the last few decades and continues to do so for two main reasons. First, prior to the COVID-19 pandemic, there were growing levels of mobility in society and a greater focus on the use of public transport, for instance, as a more sustainable alternative to the car. Many countries and larger cities in particular world-wide 355

Nigel Halpern

have experienced increased demand for public transport and from a greater proportion of the population. There are now greater demands on service quality, and markets are more fragmented (e.g. in terms of their needs and expectations). Second, the digital revolution continues to impact society and change the way people consume information. It provides opportunities for public transport operators to invest in new messages and media such as real-time information (RTI) systems. More recently, there has been growing interest in user engagement where operators seek to target more narrow audiences, or even specific individuals, with customised information and interact with users, for instance, via experiential initiatives or digital media rather than treating them as passive recipients of information. This chapter reviews current knowledge and key trends on the provision of service information for public transport. The next section considers the traditional approach of providing static information and the importance of standards. The growing interest in RTI is then considered. This is followed by a look at user engagement – interactive and more personal approaches to the provision of information. The chapter concludes by highlighting several areas of interest for future research.

User information Static information Users of public transport require clear and accurate service information that is readily available and easy to understand (EMBARQ, 2011). To achieve this, public transport operators develop user information systems that offer messages via a range of media located throughout the transport network, for instance, consisting of signage, displays, maps, timetables, information kiosks, ticket offices and machines, and help points. This has traditionally consisted of static information that rarely changes. It may be viewed by users (e.g. at a station or stop, on board the vehicle or in the street), or it may be printed and available to users to take away, as is the case with a printed timetable. Information should be located where users need it most, for instance, where decision-making is most likely to take place. The information should be simple yet informative. It should also be meaningful, intuitive, and easy for all to use, taking into account the needs of people with disabilities, and the need to be presented in multiple languages. Design considerations are important (EMBARQ, 2011). These typically include choices regarding the font and colours to be used, how the information should be presented (e.g. in terms of layout), whether it should be provided in visual and/or audio format, what type and quality of materials should be used, whether it needs to be illuminated (e.g. at night), how it will be kept clean (e.g. from graffiti or bird droppings), and whether it will have a modern or more traditional feel to it. Standards should therefore be applied throughout the system. This is important for maintaining a clear and consistent brand. To achieve this, many operators publish guidelines. For instance, Transport for London (TfL) is responsible for public transport in Greater London. TfL publishes a set of design standards that are essential to maintaining a high level of quality and consistency in its provision of information (see Transport for London, 2020). The standards allow TfL to ensure the correct positioning of branding elements, the correct use of corporate typefaces, and compliance with the UK Equality Act 2010 (UK Government, 2010). They also aid in promoting safety by making sure the messages help passengers use the transport network with ease and confidence. TfL is responsible for various modes of transport across Greater London, so it produces a range of design standards for each mode. These can be used by staff, suppliers, and design agencies. While the guidelines for each mode vary, there are also some common standards.

356

Service information for public transport

Real-time information The provision of RTI has been made possible by advancements in technology during the last few decades and is increasingly provided by public transport operators. It allows users to make more informed decisions about their trip and is recognised as a necessary component of major public transport networks, for instance, in Europe (Politis et al., 2010) and the United States (Harmony & Gayah, 2017). The latter found that almost 70 percent of the 58 public transport operators they surveyed in the United States offer RTI. RTI is typically delivered via any electronic media with display screens and automated public address systems being common choices at stations or stops, or onboard vehicles. Information provided typically includes expected departure or arrival times based on data collected from automatic vehicle location systems and is often combined with data collected from systems that capture other information, for instance, on traffic congestion, weather conditions, or incidents. Data allows the information system to compare the real-time situation with the published timetable. Information provided typically includes the next available service but also several later ones, providing both published and expected times. The cause of any disruption or delays and advice on connecting services or alternative travel arrangements may also be provided. As with static information, the needs of people with disabilities or the need for multiple languages should be considered when deciding how to present the information. Increasingly, systems are used to provide a wider range of RTI such as a list of stops, vehicle length or number of carriages (e.g. of a train) and therefore the expected location of it on the platform, vehicle type and capacity, and seat availability. However, the most valued types of information tend to be related to the location and subsequent departure or arrival time of the vehicle (Harmony & Gayah, 2017). Investment and maintenance costs for RTI systems can be high, and lack of funding is typically the main constraint for operators seeking to introduce or improve such systems (Harmony & Gayah, 2017). It is therefore important to understand the value of them. Dziekan and Kottenhoff (2007) conduct a mind-map exercise on possible effects of RTI displays at public transport stops. They identify seven possible effects: reduced wait time; positive psychological factors such as reduced uncertainty, increased ease of use and a greater feeling of security; increased willingness to pay; adjusted travel behavior such as better use of wait time or more efficient travelling; mode choice effects; higher customer satisfaction; better image. (Dziekan & Kottenhoff, 2007, p. 1) Furthermore, the availability of RTI and the quality of information provided has been found to enhance perceived usefulness and ease of use of public transport, which subsequently influence use intentions (Kaplan et al., 2017). Politis et al. (2010) found that the provision of RTI increases bus usage by 1.8 percent or approximately 320,000 new passenger trips with an estimated total value of €160,000, which goes some way towards offsetting the investment and maintenance costs for the system. Brakewood et al. (2015) report an increase in weekday bus usage of 1.7 percent as a result of the provision of mobile-based RTI. However, the increase is concentrated to larger routes. Tang and Thakuriah (2012) found that the provision of RTI increases bus usage; however, the average increase is rather modest, so while the provision of RTI is generally found to have a positive effect on use of public transport, there is some uncertainty over how great an impact it has.

357

Nigel Halpern

The provision of RTI does, however, seem to have a significant effect on waiting time. Watkins et al. (2011) conducted a study on the impact of real-time next bus countdown information. Their study found that perceived wait time is greater than actual wait time for those without RTI. However, those using RTI did not perceive their wait time to be longer than their actual wait time. RTI users reported an average wait time of 7.5 minutes versus 9.9 minutes for those using traditional arrival information – a difference of about 30 percent. Brakewood et al. (2014) also quantify the benefits of RTI for bus users, comparing users of RTI with non-users. Wait times are almost 2 minutes less for users of RTI compared to non-users (this is similar to the time difference observed by Watkins et  al., 2011). In addition, users of RTI have lower levels of anxiety and frustration when waiting for the bus compared to non-users. They also have higher levels of satisfaction with perceived waiting time and frequency of the service. The findings therefore suggest that RTI improves the passenger experience of waiting, which is a common pain-point for bus journeys. The provision of RTI also plays an important role in times of service disruption or delay (Papangelis et al., 2016). Public transport services do not always run exactly to their published timetable. Disruption and delays can occur that affect the user’s journey not only in terms of when they depart or arrive at their destination but also in terms of whether they make any planned connections with the same or other modes of transport. Disruption and delays can be a great cause of frustration and affect the user’s confidence in using the service, which may subsequently reduce the likelihood of their repeat business and loyalty. This is supported by Brakewood et al. (2014, p. 1) who state that: “Public transit agencies often struggle with service reliability issues; when a bus does not arrive on time, passengers become frustrated and may be less likely to choose transit for future trips”. The constraint of funding RTI systems was mentioned earlier. However, if the provision of RTI can combat perceptions of unreliability, it may be a cheaper alternative to other efforts to improve on-time performance (Watkins et al., 2011).

User engagement Sarker et al. (2019) mention how the traditional approach to providing public transport information involves operators as active communicators of information, while users are passive recipients. However, the changed communications landscape means that it is no longer enough to simply communicate to users of public transport. Instead, operators need to interact and engage with them at different stages of their journey. This represents a shift from mass to niche communications and is based on a more continuous approach that seeks to nurture long-term relationships. It compares to traditional approaches discussed so far in this chapter that are characterised by disconnected, point-in-time communications. By interacting and engaging with users, operators can encourage reciprocity and active involvement by enabling them to share information which may subsequently increase information quality, use, and loyalty.

Experiential initiatives Engagement may be encouraged in offline spaces, for instance, via experiential initiatives that provide live interactions. These are sometimes used by operators to inform users about new public transport systems or changes to existing systems as part of a user education programme (see EMBARQ, 2011). The House of the Tramway and Major Projects provides a good example of this. It was launched in 2017 to inform target markets of work related to the 2019 tramway project in Caen la mer and of other major projects in the city of Caen, France. The house was a physical place of information and exchange and allowed the public to access practical 358

Service information for public transport

information and find out more about planned works. A large space in the house was dedicated to a permanent exhibition about the future tram service. In addition, ambassadors were present to interact and engage with the public (Communauté urbaine Caen la mer, 2017). Prototypes are also used to show how new infrastructure or services might work in practice. For example, the city of Umeå in Sweden commissioned a prototype bus stop to be featured at the EU Arctic Forum, which was held in the city in 2019. The bus stop was designed to improve waiting conditions for passengers using public transport in the Arctic regions where adverse weather conditions are common. The stop has lights and sounds to alert passengers about approaching buses, thus allowing them to relax and take shelter rather being exposed to the elements while waiting for the bus (Ravenscroft, 2019).

Websites In terms of engagement in online spaces via digital media, the main options used by public transport operators are websites, social media, and mobile applications. Websites are commonly used, but there has been little coverage of them in the literature. One exception is EMBARQ (2011), who mention how websites can be used to feature important information, for instance, route maps and schedule information, including potential disruptions or other factors that may affect journeys. However, they note that the information itself is not enough and that websites also need to be user friendly and well designed and conform to brand standards. Although rather outdated, Transportation Research Board (2002) provides a synthesis of best practice on the effective use of websites by public transport operators. The report has a chapter on next directions for public transport websites where it lists trip planners, RTI, customer relationship management (e.g. through subscription to and mailing of e-newsletters with a variety of information), and e-commerce (e.g. sale of tickets or other services or merchandise) as key trends for the future. All of these are focused on by public transport operators today. In addition, websites increasingly offer links to social media for two-way communications, and there is growing interest in the provision of live online chats with staff or chatbots – the latter being an artificial intelligence that customers can communicate with.

Social media One of the first social media sites was the social networking site Six Degrees, launched in 1997. There are now 2.5  billion people worldwide using Facebook alone (Statista, 2020a). Public transport operators have responded to the growth in social media by joining key sites, and some of the largest operators now have an impressive number of followers. For instance, TfL has 2.5 million followers on Twitter, 425,000 on Facebook, 163,000 on LinkedIn, 125,000 on Instagram, and 47,000 on YouTube. Liu et al. (2016) examine social media site use at 43 of the top 50 public transport operators in the United States. The three most popular are Twitter (used by 100 percent of operators), Facebook (93 percent), and YouTube (81 percent). Other popular sites are Instagram and LinkedIn. Uses of social media are covered extensively in National Academies of Sciences, Engineering, and Medicine (2012). The report is based on a survey of 34 public transport operators in the United States and Canada. According to the findings, public transport operators adopt social media for five main reasons: (1) to provide timely updates (e.g. with RTI and advisories), (2) to provide the public with information (e.g. about services, prices, and planned works), (3) citizen engagement (to take advantage of the interactive nature of social media to connect with users in an informal way), (4) employee recognition (for recognising existing employees 359

Nigel Halpern

and the recruitment of new employees), and (5) entertainment (to display a personal touch and entertain users). Similarly, Liu et al. (2016) find that social media is most commonly used by public transport operators to respond to comments and to notify people about transport system changes and transport options available. Social media is cheap, quick, and easy to use. This means that even operators with small networks and limited finances can set up social media accounts. The speed and ease of using social media, along with its potential reach, means that it is particularly useful during disruptions or emergency situations when messages need to be communicated to target markets as quickly as possible (e.g. see Cottrill et al., 2017 for an example of how Twitter was used to manage disrupted services during a major sporting event). However, the relative speed and ease of use means that messages can be posted too quickly by staff without proper control over facts and quality of content (see also Chapter 24). Managing staff access to social media is also difficult given that the distinction between personal and professional lives is increasingly blurred. In addition, it is not possible to control comments from users that may be misleading to others or negative about the brand. Indeed, in their sentiment analysis of posts on Twitter, Collins et al. (2013) find that users are more inclined to post negative as opposed to positive messages. Any negative messages can spread quickly and damage the brand, especially if managed poorly. But at the same time, quick and effective responses from management can turn negative situations into positive ones. Social media can also be time consuming to manage and maintain, especially for operators that have a large following on multiple sites, such as TfL, with approximately 3.3 million followers on five different sites. According to National Academies of Sciences, Engineering, and Medicine (2012), other issues associated with social media include accessibility (e.g. for people with disabilities or for those that do not use social media), security (in terms of added exposure to cyber threats), personal privacy (which is a challenge given that social media is typically governed by the privacy policy of the platform rather than the public transport operator), archiving and records retention (to comply with rules that apply to other electronic or paper records), and keeping up with change in what is a dynamic environment. There are also challenges associated with integration – coordinating activities and information on social media with other platforms and activities, identifying revenue opportunities associated with social media, and developing metrics to evaluate the costs and benefits associated with using social media. Regarding the latter, Liu et al. (2016) find that public transport operators commonly use metrics such as the number of subscribers; number of people that receive RTI; number of people that provide feedback; and the perceptions and sentiments of people, for instance, regarding reliability of service and perceived environmental friendliness. However, more advanced social media metrics are needed that provide a deeper understanding of the benefits.

Mobile applications There are approximately 3.5 billion smartphone users worldwide, with penetration rates of 76 percent for advanced economies and 45 percent for emerging economies (Statista, 2020b, 2020c). In line with this, many public transport operators, at least in developed economies, have launched dedicated and branded mobile applications, many of which are now among the most popular travel applications in their respective countries. The example of Norway is shown in Table 25.1, where several public transport applications feature among the top 30 travel application downloads in 2019. This includes Vy, Ruter, Flytoget, Lime, Skyss, Entur, Kolumbus, AtB Mobillett, and Brakar Billett. Note that some have separate applications for travel planning (e.g. RuterResie, Skyss Reise, Kolumbus Reise) and ticket purchases (e.g. RuterBillett, Skyss Billett, Kolumbus Billett). 360

Service information for public transport Table 25.1 Top 30 iPhone travel applications in Norway 2019 Rank

App

Main sector

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30

Vy RuterBillett RuterReise SAS Norwegian Booking.com Airbnb Flytoget Skyss Billett Lime Google Earth TripAdvisor Entur Flightradar24 Uber Hotels.com Kolumbus Billett Norwegian Customs Ving Norge Skyss Reise Europeiske App Widerøe Google Street View AtB Mobillett UT.no Wizz Air Trivago Kolumbus Reise TUI Norge Brakar Billett

National rail and bus Public transport for Oslo and Viken Public transport for Oslo and Viken Airline Airline Lodging search engine Online marketplace for lodging Oslo Airport Express Train Public transport for Hordaland Electric scooter and bike sharing Mapping Reviews Public transport journey planner Flight tracker Ride sharing Lodging search engine Public transport for Rogaland Customs quota and payments Tour operator Public transport for Hordaland Travel insurance Airline Mapping Public transport for Trøndelag Norwegian Trekking Association Airline Travel search engine Public transport for Rogaland Tour operator Public transport for Buskerud

Source: Leading iPhone travel applications by downloads in 2019 on the AppStore in Norway

Mobile applications of public transport operators typically allow users to set up a profile where they can personalise settings. This also allows for customer relationship management (e.g. through loyalty programme activities) and gamification (e.g. with competitions and social activities) (Yen et al., 2019). The main function of mobile applications tends to be for travel planning (e.g. to access departure times and mode) (Jamal & Habib, 2019). It is also common to provide information on prices, trip duration, platform number, vehicle type, maps, and other wayfinding services. The integration of RTI has become a key feature on most applications with the provision of status updates and information, including whether there is seating or standing room only on the vehicle. The integration of RTI is important because mobile applications have been found to be the preferred option for receiving such information (see Harmony & Gayah, 2017). Furthermore, mobile-based RTI has been found to reduce perceived and actual wait time for public transport and improve the travel experience by making information available to users before they reach the stop (Watkins et al., 2011). Additional features allow users to personalise applications such as with saved or preferred journeys to get one-click travel suggestions, widgets 361

Nigel Halpern

to see updated RTI or saved journeys without opening the application, comparisons and filters for transport mode preferences (e.g. bus, walk, bike), and the option to set how fast you walk and to receive push notifications (e.g. regarding disruptions or delays). A key development in recent years is in ticketing, with opportunities for mobile-based purchases, ticketless travel, purchase history, and receipts. Coinciding with this is the trend for more payment options. For instance, RuterBillett (listed in Table 25.1) allows ticket payments using Visa or MasterCard (that can be saved in the application), Vipps (that connects card payments to phone numbers), Apple Pay (that uses payments from an e-wallet), PayPal (an online payments system), and AfterPay (a buy now and pay later service). Mobile applications may also offer e-commerce and merchandising opportunities and feedback and/or customer service functions (including contact details for help and information or live chats). However, a key to their success seems to be in keeping it simple and focusing primarily on travel planning and/or ticketing. The service provider’s website can then be used to provide more detailed information, so it is useful to provide a link to the website from the mobile application. Previous studies have investigated success factors for and outcomes of mobile applications in public transport. For instance, Schmitz et al. (2016) investigate how application usefulness and usability can influence perception of the service provider and quality of service. Their study analyses data from 197 public transport mobile application users in Germany. Results indicate that information fit to task, convenience value, and speed of transaction affect perceived usefulness. Moreover, ease of understanding, intuitive handling, and reliability were found to drive perceived ease of use. Their findings also identified perceptions of overall service quality, firm innovativeness, and subjective firm knowledge as outcomes of mobile application use, therefore emphasising the benefits of developing company owned mobile applications and encouraging customers to use them. There are also claims that mobile applications influence travel behaviour. For instance, Shaheen et al. (2016) identify 11 potential impacts: (1) cognitive impacts – because their powerful search capabilities can alleviate cognitive burdens; (2) improved actual and perceived control for users over their journey; (3) privacy safeguards to encourage and shape use; (4) improved trust in the service provider; (5) enhanced user experience and therefore greater use of public transport; (6) reframed norms and defaults about transportation choices, for instance, regarding the ease of mobile ticketing; (7) cheaper public transport and changed perceptions of value; (8) availability of information in a way that shapes behaviour; (9) social pressures that shape desired travel behaviour; (10) mitigated risk and therefore influence over travel choices and behaviour; (11) the offer of incentives in favour of one behaviour over another. Shaheen et al. (2016) also discuss current challenges associated with mobile applications, for instance, regarding privacy concerns, the provision of open data and interoperability among different services and transport modes, the authorisation of third-party applications, and accessibility (e.g. for those with special needs, non-users, users in rural and less urban areas, or those affected by the digital divide such as low-income groups or the elderly). Key trends in recent years have seen a growth in third-party developers of mobile applications for public transport, especially in the areas of open source, sharing, and gamification. One open source application that has been referred to several times in the literature is OneBusAway (e.g. see Camacho et al., 2013; Ferris et al., 2010). This is open-source software that offers RTI for public transport. It is maintained by a non-profit organisation called Open Transit Software Foundation. It uses GPS feeds and therefore does not require an automated vehicle location solution. The open-source nature of the application means that public transport operators or other partners can contribute to and update the application. It has traditionally been used to provide RTI for bus services. However, it can be used for multiple modes of transport. In 2019, 362

Service information for public transport

a feature was added to enable users to receive notifications (e.g. via vibrations, alerts, or audio) on how far they are from their destination, when they are approaching their stop, and where the vehicle is located in relation to streets and landmarks. This is particularly useful for people with disabilities. With regard to trends in sharing, Tiramisu Transit is a mobile application for bus users in Pittsburgh, United States. It crowdsources RTI for bus services. Information includes estimated arrival times, how full the bus is, and rider experience, and it is based on a combination of GPS data and information that is crowdsourced in real time from users of the mobile application. As a result, the application is able to provide RTI for parts of the service where users have the application active on their phone without requiring an automated vehicle location solution. When crowdsourced RTI is not available, the application shows the published schedule information or provides estimates based on historical data. Sarker et al. (2019) examine willingness to share travel information via a public transport mobile application. Their study is based on data from over 1300 people from Innsbruck and Copenhagen (cities that differ in size and level of social trust). Their findings show that the most important motivations for sharing are pro-sharing social norms and self-actualisation weighted against expected effort (e.g. of using the application). Trust in the information provided and social network engagement are secondary motivations, while perceived information quality and need of communication are less influential. There is then a positive relationship between motivation to share information and interest in service level and RTI and also in the use of public transport. Residents of Copenhagen in Denmark (with relatively high social trust) were found to have a higher level of motivation for information sharing compared to residents of Innsbruck in Austria. The likelihood of users to recommend a service is expected to be related to satisfaction with the service. For instance, Diab et al. (2017) find that satisfaction with the service increases the likelihood of recommending it to others. The most important service attributes in their study are satisfaction with waiting time, travel time, and onboard experience. Gaming is another key trend, and Di Dio et al. (2018) provide an example of the mobile application-based game called TrafficO2. It brings together local businesses as sponsors in the application. Commuters are informed of the options available to them when making mobility choices and are rewarded with “O2 points” when they opt for more sustainable choices (e.g. by foot, bike, public transportation, vehicle pooling, or car sharing). In line with Filsecker and Hickey (2014), the aim is to decrease congestion and pollution through an educational game that rewards sustainable choices. Initiatives like TrafficO2 and OneBusAway are not led by public transport operators. However, the information or incentives provided by them are likely to increase demand for public transport, so there are clearly mutual benefits to be gained from getting involved and making sure that the correct information (e.g. on routes and schedules) is shared with them. Operators may also decide to share data with other third parties such as Google Maps or to collaborate with the mobile applications of other local, regional, or national partners. A  good example of collaboration can be found in Norway, where a government-owned company called Entur has developed a mobile application for travel planning that aims to provide a national hub for public transport information in the country rather than being limited to individual modes or constrained by county boundaries as most public transport applications are. Entur’s mobile application also offers sales and ticketing services for a growing range of transport modes. Collaboration is vital to the success of Entur’s travel planner, because it operates the national registry for all public transport in Norway, collecting data from 60 public transport operators. The registry contains data from 4000 routing tables covering 60,000 stops and 110,000 boarding points (Entur, 2019). The travel planner, and the registry data that goes into it, allows the public to 363

Nigel Halpern

plan door-to-door journeys across Norway and seems to have had a great deal of success, with 160,000 application downloads, 24 million tickets conveyed, and 4 billion Norwegian Kroner (approximately 390 million Euros) in tickets sales in 2018 (Entur, 2019). However, as with many public transport operators or other developers of mobile applications, a challenge is to gain a significant market share and to maximise value from their application when there are so many other applications available on the market.

Conclusion This chapter shows how advancements in technology and changing needs and expectations of users have influenced the provision of service information for public transport. Three main approaches have been considered under the headings of static information, RTI, and user engagement. However, in reality, different users require different messages and media at different stages of their journey. There has been research undertaken on this (e.g. Harmony & Gayah, 2017; Mulley et al., 2017). However, the dynamics are not clearly understood, and a better knowledge and understanding of preferences is required in order to develop and prioritise future investments. This includes preferences regarding ticketing, which is increasingly mobile based, and also accessibility considerations (e.g. for people with special needs, low-income groups, the elderly, or those residing in rural areas). As shown in this chapter, the importance of accessible information is frequently mentioned in the literature. However, there are few studies that focus specifically on issues associated with accessibility. One exception is Papangelis et al. (2016), which explores the information needs of passengers in rural areas during public transport disruption. Given the range of media now available to public transport operators, it is also important to evaluate the costs and benefits associated with different options. The benefits of different media are frequently examined in the literature, as is shown in this chapter. However, cost-benefit analysis approaches (that subtract the costs associated with taking a decision from the benefits associated with taking it) are scarcely used. This chapter also shows that while traditional messages and media are still important, they need to evolve and coexist with new forms of communication that encourage closer relationships in more narrowly defined markets, or even with individuals. Published timetables provide a clear example of this because they not only exist in static printed format, but they also provide essential data for RTI (at stations, stops, or onboard vehicles) and online travel planners (on websites or mobile applications of the operator or any third parties). Messages and media therefore need to be fully integrated across a range of offline and online spaces. The importance of collaboration (e.g. with other providers of transport or travel planning services) is also mentioned in this chapter. However, issues associated with integration and collaboration are scarcely examined in the literature and therefore provide interesting lines of enquiry for future research, for instance regarding issues associated with open data and interoperability among different services and modes of transport. Related to this are issues of cyber security and personal privacy. These are also often mentioned but scarcely examined in the literature. Avoine et al. (2014) provide an exception to this with their study on personal privacy and the use of anonymous ticketing in public transport. However, more research is needed. This chapter has focused on the provision of service information for public transport, including the messages and media that are used and the trend of moving away from traditional approaches to the provision of static information to more interactive and personalised approaches to the provision of RTI. The current COVID-19 pandemic reinforces this trend and has several implications for the types of message and media that are used. First, messages need to be adapted to the current situation, as they would for other crises or disasters. In 364

Service information for public transport

particular, information is needed on changes to the service and any health measures that must be followed in order to minimise contact with others and avoid the potential spread of the virus. The situation for most operators is likely to be a dynamic one as a result of changing guidelines, for instance, from government or health authorities. Information therefore needs to be constantly updated and preferably available in real time. COVID-19 also has implications for the types of media to be used. In general, there needs to be a greater focus on providing information via contactless and touchless media so that information can be accessed online or via information screens rather than accessing it from staff, leaflets, published timetables, or touchscreens. Also, there needs to be less focus on providing experiential “in person” experiences, but otherwise, the developments mentioned in this chapter, for instance, with regard to the use of websites, social media, and mobile applications, are more important than ever due to the contactless and touchless solutions that they offer. This includes the use of mobile devices for ticketless travel.

References Avoine, G., Calderoni, L., Delvaux, J., Maio, D., & Palmieri, P. (2014). Passengers information in public transport and privacy: Can anonymous tickets prevent tracking? International Journal of Information Management, 34, 682–688. Brakewood, C., Barbeau, S., & Watkins, K. (2014). An experiment evaluating the impacts of real-time transit information on bus riders in Tampa, Florida. Transportation Research Part A: Policy and Practice, 69, 409–422. Brakewood, C., Macfarlane, G. S., & Watkins, K. (2015). The impact of real-time information on bus ridership in New York City. Transportation Research C: Emerging Technologies, 53, 59–75. Camacho, T. D., Foth, M.,  & Rakotonirainy, A. (2013). Pervasive technology and public transport: Opportunities beyond telematics. Pervasive Computing, 12, 18–25. Collins, C., Hasan, S., & Ukkusuri, S. V. (2013). A novel transit rider satisfaction metric: Rider sentiments measured from online social media data. Journal of Public Transportation, 16(2), 21–45. Communauté urbaine Caen la mer. (2017). La Maison du Tramway et des Grands Projets ouvre ses portes le 21 septembre 2017. Retrieved January  30, 2020, from www.caenlamer.fr/content/ la-maison-du-tramway-et-des-grands-projets-ouvre-ses-portes-le-21-septembre-2017-0 Cottrill, C., Gault, P., Yeboah, G., Nelson, J. D., Anable, J.,  & Budd, T. (2017). Tweeting transit: An examination of social media strategies for transport information management during a large event. Transportation Research Part C: Emerging Technologies, 77, 421–432. Di Dio, S., La Gennusa, M., Peri, G., Rizzo, G., & Vinci, I. (2018). Involving people in the building up of smart and sustainable cities: How to influence commuters’ behaviors through a mobile app game. Sustainable Cities and Society, 42, 325–336. Diab, E., van Lierop, D., & El-Geneidy, A. (2017). Recommending transit: Disentangling users’ willingness to recommend transit and their intended continued use. Travel Behaviour and Society, 6, 1–9. Dziekan, K., & Kottenhoff, K. (2007). Dynamic at-stop real-time information displays for public transport: Effects on customers. Transportation Research Part A, 41, 489–501. EMBARQ. (2011). From here to there: A creative guide to making public transport the way to go. EMBARQ. Entur. (2019). Annual and sustainability report 2018. Entur. Ferris, B., Watkins, K., & Borning, A. (2010). OneBusAway: A transit traveler information system. In T. Phan, R. Montanari, & P. Zerfos (Eds.), Mobile computing, applications, and services. MobiCASE 2009. Lecture Notes of the Institute for Computer Sciences, Social Informatics and Telecommunications Engineering, Vol. 35. Springer. Filsecker, M., & Hickey, D. T. (2014). A multilevel analysis of the effects of external rewards on elementary students’ motivation, engagement and learning in an educational game. Computers  & Education, 75, 136–148. Harmony, X. J., & Gayah, V. V. (2017). Evaluation of real-time transit information systems: An information demand and supply approach. International Journal of Transportation Science and Technology, 6(1), 86–98. Jamal, S., & Habib, M. A. (2019). Investigation of the use of smartphone applications for trip planning and travel outcomes. Transport Planning and Technology, 42(3), 227–243.

365

Nigel Halpern Kaplan, S., Monteiro, M. M., Anderson, M. K., Nielsen, O. A., & Dos Santos, E. M. (2017). The role of information systems in non-routine transit use of university students: Evidence from Brazil and Denmark. Transportation Research Part A: Policy and Practice, 95, 34–48. Liu, J. H., Shi, W., Elrahman, O. A., Ban, X., & Reilly, J. M. (2016). Understanding social media program usage in public transit agencies. International Journal of Transportation Science and Technology, 5, 83–92. Mulley, C., Clifton, G. T., Balbontin, C., & Ma, L. (2017). Information for travelling: Awareness and usage of the various sources of information available to public transport users in NSW. Transportation Research Part A: Policy and Practice, 101, 111–132. National Academies of Sciences, Engineering, and Medicine. (2012). Uses of social media in public transportation. The National Academies Press. Papangelis, K., Velaga, N. R., Ashomore, F., Sripada, S., Nelson, J. D., & Beecroft, M. (2016). Exploring the rural passenger experience, information needs and decision making during public transport disruption. Research in Transportation Business & Management, 18, 57–69. Politis, I., Papaioannou, P., Basbas, S., & Dimitriadis, N. (2010). Evaluation of a bus passenger information system from the users’ point of view in the city of Thessaloniki, Greece. Research in Transportation Economics, 29, 249–255. Ravenscroft, T. (2019). Station of Being is an interactive Arctic bus stop. Retrieved January 30, 2020, from www.dezeen.com/2019/12/11/rombout-frieling-lab-arctic-bus-stop-umea-sweden/ Sarker, R. I., Kaplan, S., Anderson, M. K., Haustein, S., Mailer, M., & Timmermans, H. J. P. (2019). Obtaining transit information from users of a collaborative transit app: Platform-based and individualrelated motivators. Transportation Research Part C: Emerging Technologies, 102, 173–188. Schmitz, C., Bartsch, S., & Meyer, A. (2016). Mobile app usage and its implications for service management – Empirical findings from German public transport. Procedia – Social and Behavioral Sciences, 224, 230–237. Shaheen, S., Cohen, A., Zohdy, I.,  & Kock, B. (2016). Smartphone applications to influence travel choices: Practices and policies. U.S. Department of Transportation. Statista. (2020a). Most popular social networks worldwide as of October  2019, ranked by number of active users. Retrieved January  30, 2020, from www.statista.com/statistics/272014/ global-social-networks-ranked-by-number-of-users/ Statista. (2020b). Number of smartphone users worldwide from 2016 to 2021. Retrieved January 30, 2020, from www.statista.com/statistics/330695/number-of-smartphone-users-worldwide/ Statista. (2020c). Smartphone ownership rate by country 2018. Retrieved January 30, 2020, from www.statista. com/statistics/539395/smartphone-penetration-worldwide-by-country/ Tang, L., & Thakuriah, P. V. (2012). Ridership effects of real-time bus information system: A case study in the city of Chicago. Transportation Research Part C: Emerging Technologies, 22, 146–161. Transport for London. (2020). Design standards. Retrieved January  30, 2020, from https://tfl.gov.uk/ info-for/suppliers-and-contractors/design-standards Transportation Research Board. (2002). Effective use of transit websites: A synthesis of transit practice. TCRP Synthesis 43. National Academy Press. UK Government. (2010). UK Equality Act 2010. Retrieved April 22, 2020, from www.legislation.gov.uk/ ukpga/2010/15/contents Watkins, K. E., Ferris, B., Borning, A., Rutherford, G. S., & Layton, D. (2011). Where is my bus? Impact of mobile real-time information on the perceived and actual wait time of transit riders. Transportation Research Part A: Policy and Practice, 45(8), 839–848. Yen, B., Mulley, C., & Burke, M. (2019). Gamification in transport interventions: Another way to improve travel behavioural change. Cities, 85, 140–149.

366

26 PUBLIC TRANSPORT AND SOCIAL INCLUSION John Stanley and Janet Stanley

Introduction Transport policy, like policy in most spheres of governmental endeavour, encompasses issues to do with both efficiency and equity. Trade-offs between the two are a frequent decision-making challenge. In public transport (PT), for example, should services aim to maximise patronage, an efficiency objective, or should they focus more on providing travel opportunities for people experiencing some form of transport disadvantage? This chapter seeks to provide an understanding of the equity perspective, an emerging field of research endeavour. Public transport service provision is sometimes described as having the characteristics of a merit good, which Stopher and Stanley (2014, p. 24) describe as: one which society, through its political processes, has decided should be provided on the basis of considerations of need rather than ability and willingness to pay. The good is provided in the private market place, but there is a social decision to ensure some base level is available, irrespective of individual preferences or circumstances. There are two aspects of the merit good argument. First, recognition that if provision of this good is left solely to the private marketplace, some people will consume or use less of it than is in their best long-term interest, perhaps because they cannot afford more. Second, this lower level of consumption/use by a significant number of people is then recognised as leading to lower levels of personal and wider societal wellbeing. Education is perhaps the best-known example of a merit good, where higher levels of education than might result if this was left solely to the private marketplace are seen as being good for both the individual and for the society. For societies that place a value on social inclusion, recognition that poor mobility opportunities may increase the risk that some people will be socially excluded often leads to subsidisation of public transport services as a form of social safety net. Targeted fare concessions are also often used to support mobility opportunities for particular groups, such as young or older persons, who may otherwise be at risk of mobility-related exclusion. Over the past two decades, understanding of social exclusion and its relationship with mobility has increased considerably, particularly stimulated by the work of the UK Social Exclusion Unit (SEU, 2003) and by multidisciplinary research in Europe (e.g. Mollenkopf et al., 2005) and 367

John Stanley and Janet Stanley

Australia (e.g. Currie, 2011; Stanley et al., 2011). In the latter research, factors such as social capital and psychological wellbeing were added to the analysis of mobility. A good understanding of social inclusion/exclusion is fundamental to thinking about how the value of public transport services can be enhanced, including in a future world in which the nature of (some) public transport services may change significantly (e.g. associated with evolution of autonomous vehicles). The next section of this chapter explores the concept of social inclusion and discusses its relationship to the ability to be mobile and to access opportunities. It identifies a number of other factors that also support inclusion, considers their connection to mobility/accessibility and illustrates how econometric analysis has been used to impute monetary values to (some of) the respective contributions. The chapter then examines the roles that public transport systems play in cities and regions and identifies the expected benefits from service provision, with a focus on low-density cities in developed economies. Mass transport (trunk) services are distinguished from social transport (local) services, the chapter arguing that social inclusion benefits are most likely to arise from the latter. In a time of scarce governmental funding, it is the less-patronised social transport (local) services that are often at most risk of being cut (‘rationalized’). Understanding the potential personal and societal consequences of such decisions is important to informed decision-making and to public transport policy and planning more broadly. The chapter then identifies ways in which public transport might effectively support social inclusion, in terms of availability, accessibility and affordability. This is followed by discussion of how the development of autonomous vehicles (AVs) might impact public transport services and whether this might put the role of PT in supporting social inclusion at risk. Social inclusion is potentially a significant benefit from penetration of AVs into the vehicle fleet. However, social transport is probably the most at-risk PT market segment, in terms of future competition from AVs as a means of providing shared mobility. The chapter finishes with a summary of its main conclusions in terms of shaping policy to protect those at most risk of mobility-related exclusion.

Social inclusion and the role of mobility To understand the potential role of public transport in supporting social inclusion, it is important to understand the way some fundamental concepts are defined and relationships explored. Social inclusion is defined here as the ability to participate in mainstream society; social exclusion is its obverse. The majority of people in an industrialised society have the capabilities and access to resources to facilitate their inclusion. However, particular groups of people remain at risk of social exclusion, such as those with limited education and on a low income, those in poor health and with a disability, people who are geographically isolated and some disadvantaged youth and older people. Social exclusion tends to become self-reinforcing when the only affordable living locations are those with the poorest infrastructure, services and job opportunities. Thus, social exclusion is in large part an issue of public policy and planning for the availability of the means for people to be included through the provision of infrastructure and services. It is an issue of social justice and improving society as a whole for all people. The most comprehensive study of relationships between mobility, social inclusion and wellbeing was a major research project undertaken around a decade ago in Victoria, Australia (Currie, 2011). The findings were based on linked travel surveys and in-depth interviews held in over 1,000 households in rural and urban locations. Mobility was viewed broadly. In terms of the statistical analysis discussed in this chapter, it was defined as number of trips undertaken, either by walking, cycling, public transport or motor vehicles. 368

Public transport and social inclusion

Finding a measure of social exclusion, and of key variables thought likely to affect exclusion, is critical to progressing research on this subject. The Australian study drew on, but slightly modified, the mainstream work of Burchardt et  al. (2002), who defined social inclusion as multidimensional. The Australian research measured social exclusion risk using the following five dimensions: income, employment status, political activity, social support and participation. Survey respondents who failed any of the criteria, or thresholds, were thought relatively more likely to be at greater risk of social exclusion, with those who failed several criteria at greatest risk. Of the sample, 13.6% met three criteria, suggesting that they are at high risk of social exclusion. This proportion approximated the percentage of Australians defined as being in poverty between 1999 to 2015, which ranged from 11.5% to 14.1% (Davidson et al., 2018). Participant wellbeing was measured using a personal assessment scale, the Personal Wellbeing Index, or PWI (International Wellbeing Group, 2013). It was also measured using an objective (not self-assessed) measure, the Psychological Wellbeing Scale, that measures eudaimonic wellbeing, which is the importance of life purpose and personal growth (Ryff, 1989). The number out of the five possible social exclusion criteria that a person ‘failed’ (seen as a measure of the strength of the risk of social exclusion) was expected to be negatively related to that person’s household income, social capital, if they had an extraverted personality, and the person’s level of realised mobility (number of daily trips undertaken). These relationships were all found to be statistically significant and with the expected directions (Stanley et al., 2011), as illustrated in Figure 26.1. The number of social exclusion criteria that a person met was then found to be significantly associated with self-assessed wellbeing (PWI). In other words, if you are at greater risk of social exclusion, you are also more likely to have lower wellbeing. A participant’s PWI rating was also found to be significantly associated with connection with community (stronger connection, lesser exclusion risk), three components of the objective (not self-assessed) wellbeing instrument (Psychological Wellbeing Scale) and, to a lesser extent, with other elements of that scale (Stanley et al., 2011; Vella-Brodrick & Stanley, 2013). In addition, those with lower PWI ratings were likely to have poorer levels of negative affect (stronger negative emotions, such as anger, fear, anxiety, sadness and depression) and lower levels of positive affect (joy, contentment, interest, engagement and pride). These relationships all confirmed and reinforced expectations about the likely impacts of social exclusion on at-risk people, which were key motivating factors in originally undertaking the research.

Figure 26.1 The drivers of social inclusion and wellbeing Source: Based on key variables analyzed in Stanley et al. (2011)

369

John Stanley and Janet Stanley

The key point for this chapter is the strong associations found between the ability to be mobile, social inclusion and personal wellbeing. In summary, for a low level of social exclusion risk, a person needs to: • • • • •

have mobility (a means to access activities/people) have sufficient income (education and work available and accessible) have supportive personal relationships (be able to gain social capital) and be connected to the community feel good about themselves (self-esteem, confidence) feel in control over their personal environment (capabilities to make choices, problem-solve)

Many of these drivers of social inclusion and wellbeing are associated with the ability to participate in activities, underlining the importance of being able to access services and people through some form of mobility, linked to the idea of mobility as a merit good, along with (for example) health services and education. As well as the personal impacts of social exclusion and low wellbeing, there is also a flow-on adverse impact on society, in terms of (for example) the impact of inequality and associated loss of productivity, additional health costs, a higher crime rate and so on. For example, the link between social exclusion and poor mental health found in this research is a general burden and financial cost to society. Pridmore et al. (2007) have shown a link between poor levels of social capital and poor physical and mental health and life expectancy after controlling for demographic and socioeconomic variables. Building social capital, especially bridging social capital that enables new important contacts with people, such as with employment opportunities, commonly requires mobility (Stanley et al., 2010).

The value of improved mobility for social inclusion The Australian research included econometric analysis intended to impute a value to improved mobility as it relates to reducing risks of mobility-related exclusion. Stanley et al. (2011) estimated separate models of these relationships for the Melbourne sample and for a regional Victorian sample. Because the trip making and household income variables were each significant in both Victorian models, the ratios of the coefficients on trip making and household income can be used in each case to impute the marginal value of an additional trip in terms of reducing risk of social exclusion. The implicit marginal trip value was $AUD24.40 in the Melbourne model and $AUD17.20 in the regional Victorian model, both in 2008 prices. To put these values in context, median sample weekly household incomes at survey time were $AUD237 and $AUD219 respectively. These trip values are sufficiently close to give some comfort of their validity. Because the demand for travel is a derived demand, derived from the demand to undertake the activity for which travel is required, these values essentially measure the value of the activity that a person would undertake if they made an additional (or marginal) trip. These values varied inversely with income in both cases, such that higher unit values are applicable to people with lower household incomes (halving household incomes doubles the marginal value of an additional trip/activity). Subsequent development of the regional model reduced the value of a marginal regional trip to $AUD12.80 (from $AUD17.20) but also suggested a high value for bridging social capital in reducing risk of social exclusion. If a person’s bridging social capital could be increased from a medium range to a high range, as discussed in Stanley, Stanley et al. (2019),

370

Public transport and social inclusion

this would be worth about $AUD100 a day to the person in question. Building bridging social capital is, in turn, assisted by the capacity to be mobile (Vella-Brodrick  & Stanley, 2013). Unit values such as these, adjusted upwards from 2008 prices, can be used to estimate in money terms the contribution that public transport services, or improved services, might make to reducing risk of social exclusion by facilitating trip making or increased trip making (depending on context) by those at risk of mobility-related exclusion. Application of these values is most justifiable in an Australian setting. Repeating the research in other countries would be of great value for confirming the significance of social inclusion benefits. This research justifies a new benefit category for transport cost-benefit analysis, and applications beyond public transport are possible, because the values are not modally based: they apply to increased trip making, irrespective of mode. The inverse relationship of the resulting values with household income levels means that it is important to identify the relevant expected beneficiaries and their household incomes. Also, the inclusion of bridging social capital values in the latest version of the modelling (Stanley, Stanley et al., 2019) indicates the importance of understanding the distribution of this variable among prospective beneficiaries and how improved public transport services can enhance bridging social capital. This is an important research opportunity.

Public transport and social inclusion Public transport service markets and their benefits As noted in the Introduction, and recognising a risk of oversimplification, public transport services are primarily of two kinds: mass transport (or trunk) services, where the intent is largely to take people to or from their neighbourhood, and social transport (or local) services, which mainly take people around their neighbourhood (and may connect to services that will take them to/from their neighbourhood). Some services seek to perform both tasks, often to the detriment of both. Public transport services sometimes operate on a fully commercial basis, where fares and other revenues fully recover costs of service provision, including a profit for the provider. However, the more usual financial context in a low-density developed country urban setting is one in which one or more levels of government provide some financial support to public transport to cover shortfalls in revenues. In Australia, the United States and Canada, for example, operating cost recovery rates for urban public transport services of between 20% and 60% are common (Stanley, 2014), whereas many PT services in the United Kingdom outside London, and informal transport services in developing countries, need to operate largely from the farebox. Fare revenues that are captured by PT service providers reflect benefits that service users expect to receive from travel.1 Governmental support for service provision should be predicated on wider benefits that government expects will flow from the service. These wider benefits are of two main types: • •

benefits that are expected to accrue to the wider society, known as external benefits (or costs, if negative), and benefits expected to flow to some system users through the merit good principle. This category of benefit accrues to the user but would not exist in the absence of governmental support for service provision.

371

John Stanley and Janet Stanley

External benefits are largely associated with public transport trunk services, which are usually provided for mass travel to relatively dense destinations, the main external benefits being agglomeration economies, congestion cost savings and lower environmental costs. Heavy rail, light rail/tram and bus rapid transit are major ways of providing trunk or mass transport. Merit good benefits are essentially about local (social) transport, because such services are primarily about enabling people to access the opportunities available in their community, which supports their social inclusion. The benefits that governments should expect from supporting local public transport services are thus essentially to do with merit good arguments about increased social inclusion. Relevant unit values, which are significant, were discussed in the preceding section of this chapter. Local or social transport services are unlikely to directly generate local agglomeration economies, congestion savings or environmental benefits of any significant magnitude, since local trips tend to be short and on uncongested roads at most times. However, to the extent that these services are used to access trunk mass transport services, there are likely to be some external benefits associated with use of the trunk services that are accessed. Importantly, increased social inclusion may be associated with external (flow-on) benefits, such as lower health system costs and reduced costs of crime, as the personal circumstances of some hitherto excluded people are improved. We are not aware of any studies that have systematically measured such flow-on benefits in a way that would permit their application in cost benefit analysis. However, research undertaken for the UK’s Urban Transport Group on the benefits of bus services recognises the probability of such flow-on benefits and elaborates on specific examples of public agencies (e.g., health, education) whose activities and budgets can be expected to benefit from well targeted bus services (see, for example, Fuller, 2019; Abrantes et al., 2013). More broadly, all PT services that attract people out of cars may realise safety benefits and health benefits, the latter flowing (for example) from increased incidental exercise. As an illustration, the journey to work by a typical Melbourne car user involves less than 10 minutes incidental daily exercise associated with the work trip, whereas a PT user gets around 40 minutes, which meets normal daily requirements for healthy living (National Heart Foundation of Australia, 2014). The societal contexts and associated expected benefits of trunk (mass) and local (social) transport services are thus different in many ways. The discussion about operating environments that follows, which explores the differences between these two types of service in more detail, is most applicable to conditions in low-density developed countries such as Australia, Canada and the United States, although much of the policy discussion has wider application.

Ridership versus coverage The key public transport market distinction is often described as being a choice between pursuing ridership via mass transport service versus coverage via local transport. Figure 26.2 illustrates some of the key characteristics of urban structure that have been shown to affect PT patronage and car use (Ewing & Cervero, 2010) and explores how these are likely to relate to mass and social transport services. In outer urban (and regional) areas, land use density and mix (diversity) are usually relatively low but distance from a city’s central business district (CBD) is relatively high (shown in reciprocal form in Figure 26.2 as 1/Distance from CBD being low). This is a relatively poor market for mass transport services, so PT connectivity will usually be relatively poor. PT service characteristics in this spatial setting are typically relatively low service frequency levels, short spans 372

Public transport and social inclusion

Figure 26.2 Aligning PT service with land use Source: Based on Stanley, Hensher et al. (2019)

of operating hours, less direct routes and relatively poor travel time compared to car (including access/egress/wait/transfer stages). This is where social transport needs are likely to be, in relative terms, at their highest, given the tendency for land prices in such locations to be relatively low and attractive to lower-income households, often young families. Accessibility is typically poorer in lower-density areas, in both the sense that PT stops may be less accessible than in better-served areas and in the broader sense that fewer access opportunities will be available by PT within any given travel time. As distance from the fringe reduces (i.e., the CBD becomes closer), densities increase, land use diversity (mix) typically increases and PT connectivity improves, because this is where mass transport services are most common, operating at higher frequency, over longer operating hours and with more direct routes. PT door-to-door travel time improves somewhat relative to that by car. Social transport needs in more inner locations will exist but are likely to be relatively less prevalent than in outer areas because of higher household incomes in more accessible locations and because many travel needs of those at risk of mobility-related social exclusion will be met by mass transport services in such locations. PT operating cost per passenger and per passenger kilometre tend to be relatively lower for mass transport services, where scale economies are most likely, and higher for the social transport service, recognising that different PT modes may perform some or all these respective services. Thus, for example, Victorian 2016–17 Budget Paper No. 3 (DTF, 2016) suggests that 2015–16 Melbourne public transport costs, mainly operating, were $AUD5.28 per passenger for bus (of which perhaps one fifth is capital cost), $AUD3.35/passenger for train (operating payments for metropolitan train services) and $AUD1.06/passenger for tram (operating payments for tram services). Buses mainly operate in outer areas with relatively more focus on service coverage, whereas train and tram are mass transport modes. Conversely, however, capital costs for mass transport, particularly rail and tram/light rail, are very high relative to social transport, for reasons such as the high cost of land acquisition and/or tunnelling (purchase/construction of dedicated right-of-way), fleet costs and signalling systems. For example, Melbourne’s Metro Rail Tunnel project alone has an estimated cost of $AUD11 billion, and total current committed rail capital works are around $AUD30 billion, none of which is reflected in the $AUD3.35/ passenger cost for rail cited previously. 373

John Stanley and Janet Stanley

Public transport service provision frequently confronts a trade-off choice between pursuing patronage and service coverage. Providing mass transport services, which tends to have relatively low operating costs per passenger kilometre, caters to large numbers of users and delivers external economic and environmental benefits, though the capital cost of new services is likely to be very high (see also Chapter 32). On the other hand, a focus on pursuing service coverage aims to support social inclusion and the benefits associated therewith, but boarding rates per service kilometre tend to be low and costs per kilometre, per passenger and per passenger kilometre tend to be high. Under funding pressures, governments often favour mass transport services, sometimes cutting local transport services, but high capital/borrowing costs might be involved in adding mass transport capacity (easier at times of low borrowing costs). Such choices need to be informed by analysis of the costs, or foregone benefits, associated with the changes that are contemplated and by opportunities that may be available to improve the performance of local transport services. The PT service characteristics box in Figure  26.2 lists several levers that can be changed in the short term. The extent to which they might improve boarding rates on local transport is a matter for detailed local examination. For example, some Melbourne case studies have shown how extending bus operating hours into the evening can increase boarding rates on earlier bus services, as people have a greater chance of travelling in both directions by PT (Loader & Stanley, 2009).

Target boarding rates for a regular local public transport (route bus) service Using the values of additional trip making as a factor supporting reduced risk of mobilityrelated social exclusion, as summarised earlier in this chapter, Stanley, Stanley et  al. (2019) estimated that a local bus service boarding rate of 6–7 persons per hour in regional Victoria would be sufficient for that service to break even on the monetary value of the expected social inclusion benefits, given that the large majority of users of those services would be in the atrisk category (primarily older persons, students and people without a car available). Stanley and Hensher (2011) estimated a boarding rate of 10–11 per service hour for Melbourne route bus services for break-even on social inclusion grounds, also including associated congestion cost savings benefits. They argued that about one third of local route bus users in Melbourne would be at risk of mobility-related social inclusion. It is interesting to look at patronage expectations that appear to be built into Vancouver, BC, bus services as a reflection on how policy makers in that jurisdiction may view low-patronage services. Examination of boarding rates across a large number of Vancouver bus services shows that the lowest average boarding rate per revenue hour is 3, with a number of other services averaging 10 or fewer boardings per revenue hour. Boarding rates of about 6–7 per revenue hour seem to be the lower bound that is commonly recognised as adequate for a service, with about half a dozen services averaging boarding rates around this level across Greater Vancouver (out of 217 services). It is notable that all the services with low average boarding rates use mini-buses for service provision, and all have average costs per boarding of around $CAN10 or higher, with the highest rate being $CAN22.35 (2018 prices) These numbers provide some insight into implicit minimum local transport boarding levels that are seen as needed to support social inclusion in Vancouver. In short, the value of additional trip making by people at risk of social exclusion is high to those persons. These high values should be recognised in service policy making and planning, particularly when service rationalisation of low-patronage services is contemplated and when planning new greenfield services. Flow-on external benefits from reduced social exclusion will 374

Public transport and social inclusion

add to the societal benefits of such services, suggesting that the break-even boarding rates cited previously might be lowered somewhat.

Planning for mobility-related social inclusion The Strategic land use influencers box in Figure 26.2 suggests some of the planning levers that might be used to improve the economic and financial performance of local transport. These are all essentially longer-term levers but can make a difference to patronage levels on local transport services over time or more quickly in new urban developments. While it is not possible to alter the distance of an existing neighbourhood from the CBD, development of a polycentric urban form may have a similar type of impact on improving service economics in larger cities. This is most likely to be associated with the other three elements in Figure 26.2. Population and job densities are usually both significantly associated with public transport boarding rates, though elasticity values are relatively low (Ewing & Cervero, 2010). However, most reported elasticities come from the United States. Higher values might be expected in settings with higher starting densities.

Some practicalities of effective public transport service delivery for inclusion The preceding sections have argued why public transport can play an important role in reducing risks of mobility-related social exclusion. The extent to which this opportunity might be realised depends substantially on how services are delivered, with matters of availability, accessibility and affordability being central.

Availability If public transport services are to support social inclusion, then some of the key service characteristics noted in Figure 26.2 need to be planned specifically to achieve this intent. Service availability, in terms of (for example) proximity to households, frequency of service and the span of service hours, is important. Walk distances to local public transport are usually regarded as being a maximum of 400–500 metres, with longer walk distances being acceptable for premium trunk services. Service frequencies and operating hours then need to be sufficiently attractive to reduce pressures on at-risk households that push their household budgets into a marginal car purchase, sometimes known as ‘forced car ownership’ (see Currie & Senbergs, 2007). Headways of 20–30 minutes, over 18 or so service hours a day, seem likely to be the minimum needed to provide a serious level of support for mobility-related social inclusion. In an Australian regional setting, the threshold boarding rate of 6–7/hour can be used as a guide to whether there is adequate patronage to support this service level by a regular PT service in a regional city, or 10–11/hour in a metropolitan setting. If this is not met for significant time segments, then alternative delivery methods to support inclusion may be necessary, which is likely to be less effective at supporting inclusion but have a lower total cost (but probably a higher cost per passenger).

Accessibility In the current context, accessibility means that the PT service that is provided to support social inclusion can be easily used by at-risk people, who may (for example) be persons with a disability or persons whose first language is not the local language. Thus, to illustrate for physical 375

John Stanley and Janet Stanley

accessibility, universal access measures such as safe wheelchair and child stroller spaces on board vehicles, access ramps, kneeling buses, stops with shelter, cues that assist waiting and boarding and the like should all be fundamental expectations. This extends to matters such as the quality of footpaths for accessing PT stops – where easy wheelchair accessibility should be available, recognising the challenges of terrain. The failure to meet universal access standards will exclude some people from the mobility support for inclusion that the service is intended to provide. There is a growing body of research on valuation of a range of public transport service characteristics that impact the customer experience. De Gruyter et al. (2018), for example, present a meta-analysis of studies on customer amenity valuations, most depending on stated preference analysis or simpler customer ratings. While this research is only in its infancy, it should be of assistance in helping service providers tailor services more finely to their customers’ expectations, including those at risk of mobility-related exclusion, and assisting governments in shaping their contractual expectations about service standards from contracted providers.

Affordability Affordability here has two meanings: affordability to the person at risk of mobility-related exclusion and affordability to the government authority tasked with financially supporting the PT service. We deal with them in this order. Low income is a common contributory factor towards social exclusion. If PT is to be used to reduce risks of mobility-related social exclusion, then fares need to be affordable. Fare discounts are commonly used to support some categories of user, such as older persons, students and persons with a disability, significant numbers of whom would be at increased risk of exclusion in the absence of this support. The affordability of fares that are levied on users will, in turn, be assisted by PT service delivery costs being kept at efficient levels, such that governmental subsidy requirements are not unnecessarily inflated. There are many ways this can be assisted. Operating costs for local PT services, which are usually local route bus services, are affected by many variables. Some of the key cost-influencing factors include: •

•

whether the service is provided by a government agency or a private provider, with private providers usually having lower operating costs, though the prospective cost difference may sometimes be only marginal (Wallis & Hensher, 2007), and if service provision is by the private sector with governmental financial support, the nature of the arrangement under which that support is provided will usually affect the costs of service provision (Merkert et al., 2018).

Service delivery contracts need to ensure performance pressure is applied to PT service providers to increase patronage and to operate an efficient and effective service. This usually means incentive and penalty clauses in the contract linked to both patronage and service quality. Hensher (2015a) sets out some useful service quality performance indicators in his Customer Service Quality Index. Competitive tendering for the rights to provide service is commonly used as a means of applying performance pressure on operators, but Hensher and Stanley (2010) and Hensher (2015b) have argued that negotiated performance-based contracts can be equally effective in terms of applying performance pressure to keep costs at an efficient level. With labour being the single largest operating cost for providing local PT services, ways to substantially reduce labour costs are matters that operators need to keep under constant review. The introduction of autonomous vehicles is a major wildcard here. 376

Public transport and social inclusion

Risks to local public transport from autonomous vehicles2 Changing societal values and emerging new technologies are increasing the probability of large shifts in transport service offerings in coming decades, with potentially huge benefits and/or costs. Autonomous vehicles and, more particularly in a warming world, autonomous electric vehicles (AEVs) have been the primary focus of this discussion internationally, with their potential to provide service to socially excluded people often being claimed as a major prospective benefit. In terms of public transport service provision, where labour costs can constitute half or more of total operating costs, the opportunity to remove labour costs with autonomous operation poses big questions for the future of public transport as we know it today. The high capital costs and associated high patronage of (radial) rail mass transport services to central cities and to other major activity clusters in the larger cities provides them with significant natural monopoly characteristics, which suggests that multiple sources of supply are unlikely.3 The external benefits associated with rail services, for example, are often very large and speak to the importance of strong governmental control over service provision. These natural monopoly characteristics and external benefits are such that, in coming years, the mass (trunk) transport market in most developed country urban settings is likely to remain as some form of mass public transport as we currently understand it. Local or social transport services can be provided by smaller units than mass transport services, which makes them more open to competition from a new shared mobility provider than is the case with mass transport. It is these local social transport services (mainly local fixedschedule route bus services) that are most likely to face intense competition from expanded personal travel opportunities offered through emerging and future shared autonomous mobility services, probably packaged under Mobility as a Service (MaaS) offerings (see also Chapter 3). Significantly, the valuable social inclusion benefits from local or social transport are also likely to be available from an alternative form of local mobility provision, at least to some extent; they are not unique to local bus services. It thus comes down to who can provide an adequate level of social (local) transport-like service most effectively, efficiently and sustainably. If social inclusion is seen as a societal priority, then, in a future that includes autonomous vehicles, provision of some base level of shared mobility service to support or underwrite social inclusion still seems warranted to ensure that minimum local mobility opportunities are available to ‘at risk’ people. Stanley, Hensher et al. (2019) suggest that local shared mobility contracts could be developed to support such service provision, which would be expected to vary by demographic/land use setting. For example, expectations should realistically be for a lesser service level in a rural area than in a town. Requisite minimum service levels need to be set out in service contracts and might be expressed, for example, in terms of: • •

seat kilometres to be supplied per time period/spatial setting, where time periods and spatial settings are specified, and/or the maximum wait time for a demand-responsive service, within specific locations and time periods.

Any such clauses would require mechanisms to be in place (e.g., bonuses, penalties) to help ensure compliance. In the interests of more efficient, integrated service delivery at minimum call on the public purse, shared mobility service contracts should strive to be as broad as possible, generally seeking to encompass route-type PT services, school services and community transport service offerings, together with taxis, other ride-sharing modes and bike-sourcing. This implies a need for 377

John Stanley and Janet Stanley

restructuring flows of public funds that currently support such varied services to ensure integrated service planning and provision on a spatial basis.

Risks to public transport from a pandemic The uncertainty of a pandemic, such as COVID-19, in terms of occurrence, impact and society’s ability to control the spread of the disease leads to uncertainty about the longer-term impact on public transport. There is likely to be an elevated risk for people experiencing mobility-related social exclusion, due to an increased probability of reduction in local bus services, associated with the relatively higher cost of this service per traveller. This risk is amplified to the extent that there is a policy failure to recognise the wider societal benefits of local transit, such as lower health costs, reduced need for welfare payments and improved personal wellbeing. Other shared mobility modes may also be at risk, being less appealing during a pandemic for those who do not wish to travel in close proximity to another. Exclusion risks will again arise.

Conclusions This chapter has emphasised the distinction in public transport service provision between mass (trunk) transport and social (local) transport and explained how the public policy basis for governmental funding support for each of these service types is different. Mass transport is essentially about direct user benefits plus external benefits, such as agglomeration economies, congestion cost savings and environmental benefits. In contrast, social transport is about supporting inclusion of those at risk of mobility-related exclusion, which the chapter has argued should be treated a merit good. Australian research has shown high values to the at-risk person. Similar research should be encouraged in other settings and would add much value to understanding of the social inclusion benefits of transport and public transport. Social transport services are also expected to generate external benefits from flow-on impacts of increased social inclusion, such as a lower crime rate/crime costs, reduced health care costs and lower unemployment benefits. Quantification of such social transport externalities has made little progress compared to valuation of external benefits of mass transport. This unequal state of understanding of service societal benefits is likely to bias resource allocation towards support for mass transport services and away from social transport. Research on the nature and magnitude of the external benefits of social transport should be a high priority. Social transport is likely to experience considerable pressure in coming years from the introduction of autonomous vehicles, While the timing and scale of impact are uncertain, the chapter suggests that policy measures should be put in place to prepare the ground for this evolution, with a focus on protecting those at risk of mobility-related exclusion. Shared mobility contracts are suggested as one way forward here, implying (among other things) a shake-up in existing fragmented funding flows that are often in place to support different forms of local public/community/school transport.

Notes 1 Fares are usually a low estimate of expected benefit due to the existence of consumer’s surplus, which is equal to the difference between the expected monetary value or benefit from the trip to the traveller and the fare paid. 2 This section draws heavily on Stanley, Hensher et al. (2019). 3 Although vertical separation of track and services can be used to reduce the degree of natural monopoly.

378

Public transport and social inclusion

References Abrantes, P., Fuller, S., & Bray, J. (2013). The case for the urban bus: The economic and social value of bus services in the metropolitan areas. Public Transport Executive Group (PTEG). Retrieved January 29, 2020, from www. urbantransportgroup.org/resources/types/reports/case-urban-bus-economic-and-social-benefits-inves ting-urban-bus Burchardt, T., Le Grand, J., & Piachaud, D. (2002). Degrees of exclusion: Developing a dynamic, multidimensional measure. In J. Hills, J. Le Grand,  & D. Piachaud (Eds.), Understanding social exclusion (pp. 30–43). Oxford University Press. Currie, G. (Ed.). (2011). New perspectives and methods in transport and social exclusion research. Emerald Publishing Limited. Currie, G., & Senbergs, Z. (2007). Exploring forced car ownership in metropolitan Melbourne. 30th Australasian Transport Research Forum. Davidson, P., Saunders, P., Bradbury, B., & Wong, M. (2018). Poverty in Australia, 2018, ACOSS/UNSW. Poverty and Inequality Partnership Report No. 2. ACOSS. De Gruyter, C., Currie, G., Truong, L.,  & Naznin, F. (2018). A  meta-analysis and synthesis of public transport customer amenity valuation research. Transport Reviews, 39, 261–283. Department of Treasury and Finance (DTF). (2016). Getting it done: Victorian budget 16/17, service delivery. Budget Paper No. 3. Retrieved August 25, 2020, from http://budgetfiles201617.budget.vic.gov. au/2016-17+State+Budget+-+BP3+Service+Delivery.pdf Ewing, R., & Cervero, R. (2010). Travel and the built environment. Journal of the American Planning Association, 76(3), 265–294. Fuller, R. (2019). The cross-sector benefits of backing the bus. Urban Transport Group. Retrieved January 29, 2020, from www.urbantransportgroup.org/resources/types/reports/cross-sector-benefits-backing-bus Hensher, D. (2015a). Customer service quality and benchmarking in public transport contracts. International Journal of Quality Innovation, 1(4). doi:10.1186/s40887-015-0003-9 Hensher, D. (2015b). Cost efficiency under negotiated performance based contracts and benchmarking. Journal of Transport Economics and Policy, 49(1), 133–148. Hensher, D. A.,  & Stanley, J. (2010). Contracting regimes for bus services. What have we learnt after 20 years? Research in Transportation Economics, 29(1), 140–144. International Wellbeing Group. (2013). Personal wellbeing index (5th ed.). Australian Centre on Quality of Life, Deakin University. Retrieved January 31, 2020, from www.acqol.com.au/instruments#measures Loader, C., & Stanley, J. (2009). Growing bus patronage and addressing transport disadvantage – The Melbourne experience. Transport Policy, 16, 106–114. Merkert, R., Preston, J., Melkersson, M., & Linke, H. (2018). Workshop 2 report. Competitive tendering and other forms of contracting out; Institutional design and performance measurement. Research in Transportation Economics, 69, 86–96. Mollenkopf, H., Marcellini, F., Ruoppila, I., Szeman, Z., & Tacken, M. (Eds.). (2005). Enhancing mobility in later life: Personal coping, environmental resources, and technical support. The out-of-home mobility of older adults in urban and rural regions of five European countries. IOS Press. National Heart Foundation of Australia. (2014). Blueprint for an active Australia. National Heart Foundation of Australia. Pridmore, P., Thomas, L., Havemann, K., Sapag, J., & Wood, L. (2007). Social capital and healthy urbanization in a globalized world. Journal of Urban Health: Bulletin of the N.Y. Academy of Medicine, 84(1), 1130–1143. Ryff, C. (1989). Happiness is everything, or is it? Exploration on the meaning of psychological wellbeing. Journal of Personality and Social Psychology, 57, 1069–1081. Social Exclusion Unit. (2003). Making the connections: Final report on transport and social exclusion. Retrieved August  25, 2020, from https://webarchive.nationalarchives.gov.uk/+/www.cabinetoffice.gov.uk/ media/cabinetoffice/social_exclusion_task_force/assets/publications_1997_to_2006/making_transport_2003.pdf Stanley, J. K., Hensher, D. A., Stanley, J. R., & Vella-Brodrick, D. (2011). Mobility, social exclusion and well-being: Exploring the links. Transportation Research A, 45, 789–801. Stanley, J., & Hensher, D. (2011). Economic modelling. In G. Currie (Ed.), New perspectives and methods in transport and social exclusion research (pp. 201–222). Emerald Publishing Limited. Stanley, J. (2014). Public transport: Funding growth in urban route services (Bus and Coach Industry Policy Paper 3). Bus Industry Confederation. Retrieved April 13, 2020, from www.movingpeople.com.au/ solutions-for-moving-people/bic-policies

379

John Stanley and Janet Stanley Stanley, J., Hensher, D., & Wong, Y. (2019). Moving people in the future: Land passenger transport and ‘new’ mobility technology (Bus and Coach Industry Policy Paper 11). Bus Industry Confederation. Retrieved April 13, 2020, from www.movingpeople.com.au/solutions-for-moving-people/bic-policies Stanley, J., Stanley, J., Balbontin, C., & Hensher, D. (2019). Social exclusion: The roles of mobility and bridging social capital in regional Australia. Transportation Research Part A, 125, 223–233. Stanley, J., Stanley, J. Vella-Brodrick, D., & Currie, G. (2010). The place of transport in facilitating social inclusion via the mediating influence of social capital. Research in Transportation Economics, 29, 280–286. Stopher, P., & Stanley, J. (2014). Introduction to transport policy: A public policy view. Edward Elgar Publishing. Vella-Brodrick, D., & Stanley, J. (2013). The significance of transport mobility in predicting wellbeing. Transport Policy, 29, 236–242. Wallis, I., & Hensher, D. A. (2007). Competitive tendering for urban bus services – Cost impacts: International experience and issues. In R. Macario, J. M. Viegas, & D. A. Hensher (Eds.), Competition and ownership in land passenger transport (pp. 453–488). Elsevier.

380

27 PUBLIC TRANSPORT AND TRAVEL WITH DOGS Jennifer L. Kent, Corinne Mulley, Laura Goh and Nick Stevens

Introduction Public transport policies are not made in a cultural vacuum. Instead, they are shaped by a complex mix of socially constructed dimensions, ranging from a collective appreciation of personal space to historical legacies of habit and custom. This chapter aims to demonstrate the link between one particular aspect of public transport policy and cultural attributes. Using the public transport policies regulating the carriage of dogs on public transport as a case study, the chapter applies a commonly accepted framework of national cultural dimensions to reveal associations between public transport policy and culture. It is intended that this chapter will both provide policy makers with suggestions for how to tailor public transport policies to allow the carriage of dogs and transport researchers and professionals an appreciation of the requirement for the public transport offer to respond to the changing needs of their diverse user base. The chapter first reviews existing research on companion animal ownership and travel, demonstrating the demand for travel with dogs. Public transport policies pertaining to dogs relating to cities in Europe, the United Kingdom and Australia are then examined. The analysis reveals substantial differences between countries and continents, ranging from policies that are entirely prohibitive to those that allow dogs on public transport under certain circumstances. Recognising that dog ownership is a culturally inculcated practice, the chapter then seeks to explore ways these differences are related to aspects of national culture as defined by the Hofstede model of cultural dimensions. The chapter concludes with discussion of future research requirements in this space, calling for transport research to place more emphasis on unusual, “messy” trips, as well as on cultural attributes.

Background What is culture? Social scientists studying practices such as travel generally prioritise the influence of either the pursuit of individual purposes, desires, intentions and motivations or the collective pursuit of a socially and structurally defined pattern (Steg, 2005; Scheiner & Holz-Rau, 2007; van Acker et al., 2010). A cultural outlook treads between these two views by seeking to articulate what 381

Kent, Mulley, Goh and Stevens

is desirable to the individual and what is normal to the collective. Culture, therefore, is an ‘implicit, tacit or unconscious layer of knowledge’ (Reckwitz, 2002, p. 246). It is the complex amalgam of attitudes, values and beliefs shared with those around us and exhibited in day-to-day practices such as working, eating, shopping and, of course, travelling. Research has long recognised variations in transport practices exhibited from city to city are underpinned by variations in culture, the study of which is helpfully defined as mobility cultures (Klinger et al., 2013; Aldred & Jungnickel, 2014). Mobility cultures mean the milieu of travel patterns, the built environment and mobility-related discourses that characterise transport in a particular city or nation (Haustein & Nielsen, 2016). They are defined by both material and the socially constructed dimensions of the transport system. The complexity and contextual dependency of the concept mean there is no one set way to analyse mobility cultures. Researchers instead employ a diverse offering of methods and methodologies, ranging from, for example, quantification of time series data (Mattioli et al., 2016) and combinations of built form characteristics and subjective factors (Klinger et al., 2013; Klinger & Lanzendorf, 2015) to qualitative methods involving life histories (Sattlegger & Rau, 2016), interviews (Hopkins & Stephenson, 2016; Nello-Deakin & Nikolaeva, 2020), qualitative mapping and photo elicitation (Dahl Wikstrøm & Böcker, 2020). While culture is not necessarily defined by nationality, nor is it uniform across nations, indices of national culture can also provide interesting insights into the mobility culture of a city. National culture is defined as ‘the collective programming of the mind that distinguishes the members of one national group from another’ (Hofstede et al., 2010, p. 6). The two most commonly used indices of national culture are those developed by Geert Hofstede (Hofstede, 1984, 2001; Hofstede et al., 2010) and Ronald Inglehart (Inglehart, 1990, 1997, 2018). Both are based on large-scale rolling surveys of the population of major OECD countries around the globe. Together, Hofstede and Inglehart have amassed over 200,000 academic citations, making them two of the world’s most frequently quoted social scientists (Beugelsdijk & Welzel, 2018). The concepts and tools associated with their frameworks, however, are rarely applied to the study of transport. Some preliminary conceptual work has recommended the inclusion of formal national cultural considerations to psychological aspects of individual transport decision making – particularly symbolism (Ashmore et al., 2017, 2020) and uncertainty avoidance (Syam, 2014). Scholars seeking to understand consumer behaviour have also used such indices to examine purchasing decisions, specifically for new alternative fuel vehicles (Oliver & Lee, 2010; Ashmore et al., 2018). More recently, the Hofstede indices have been used to analyse differences in actual travel patterns between nations (Dingil et al., 2019) and the travel behaviour and attitudes of people from different cultural backgrounds (Syam, 2014). Although it is widely recognised that the transport policies governing transport systems and influencing transport practices are also cultural constructs (Legacy et al., 2017), indices of national culture do not appear to have been applied to the analysis of differences in transport policy, including policies for public transport systems. While both the Hofstede and Inglehart indices have been critiqued widely  – equally for their simplicity and overallocation of complexity  – the breadth of data upon which they are based alone indicates some utility in efforts to understand transport policy decision making, particularly in preliminary considerations of the influence of national cultural attributes, as is the aspiration of this chapter. The chapter employs the Hofstede framework. While Inglehart’s work has an explicit focus on economic and political environments, it is most often applied in the study of politics and sociology and is particularly useful in examining shifts over time (Beugelsdijk & Welzel, 2018). The Hofstede framework, however, is more multidimensional in its consideration of psychological imprints on cultural phenomenon and is less adept at examination of shifts yet effective 382

Public transport and travel with dogs

at explorations of the status quo. The complexity implied by Hofstede’s multidimensional approach also suggests it provides a more useful base from which to analyse practices of mobility in national contexts. The Hofstede dimensions of national culture are based on an ongoing survey of value systems of residents in 100 countries. It was first conducted between 1967 and 1973 with IBM employees selected using a matched sampling procedure from 50 countries. The results of the survey are subject to factor analysis, resulting in scores from 0–100 allocated to each nation across six cultural dimensions. These dimensions are described as follows, making reference where appropriate to European countries and Australia, which are the focus of this chapter (adapted from Hofstede Insights, n.d.): Power distance: All societies suffer from inequalities, and this dimension expresses the national attitude towards this inequality. The power distance score reflects how the less powerful members of institutions and organisations within a country both expect and accept inequalities in the distribution of power and how social inequality is endorsed by society’s leaders. Low scores suggest a culture in which there is a desire or drive to equalise the distribution of power, while high values are exhibited in a culture where hierarchical order is accepted. In the sample of countries included in the analysis in this chapter, Austria has a particularly low score against the power distance dimension (11), indicating that hierarchy in Austria is for convenience only, with equal rights respected and unjustified control disliked. Slovakia, conversely, scores a maximum of 100 points on this dimension, suggesting that in Slovakia, it is accepted that some people have more power than others and that these people will use this power to create clarity and structure. Individualism versus collectivism: This dimension describes the extent to which individuals are willing to subvert their desires for maintenance of the wellbeing of the collective. Nations scoring highly on this dimension are individualistic cultures, where individuals look after themselves and are motivated by self-preservation rather than collective approval. Low scores on this dimension indicate a nation where people are bound by strong relationships into groups, which take care of individuals in return for loyalty. In the sample of countries included here, Australia scores highly on this dimension (90), translating to a loosely knit society in which the expectation is that people look after themselves and their immediate families. Portugal, however, scores 27 on this dimension, which is particularly low even in the context of other European countries. This suggests this culture is collectivist, valuing long-term commitment and loyalty to a “group”, be that extended family, relationships forged in a workplace or other collective such as a church or sporting team. The society fosters strong relationships where everyone takes responsibility for fellow members of their group. Masculinity: A high score on this dimension reflects a masculine culture, where competition is accepted and valued, and achievement, assertiveness and material rewards are highly prized. A low score – suggesting the feminine end of the dimension – identifies a culture where the dominant value is collective caring and quality of life. With a score of just 5 on this dimension, Sweden is a feminine culture, indicating an appreciation of balance and an appreciation of consensus over competition. Hungary, however, scores 88, which is relatively high and indicates a culture that resolves difference through competition and values decisiveness. Uncertainty avoidance: This dimension quantifies the way in which individuals feel comfortable or uncomfortable with uncertainty and ambiguity. High scores on this dimension suggest a culture intolerant of unorthodox ideas and ways of doing things and anxious 383

Kent, Mulley, Goh and Stevens

to control the future. A low score indicates a culture that is more comfortable with new and different ways of thinking and living and happier to let the future develop unfettered. Greece scores a maximum of 100 on this dimension, indicating a culture that is not comfortable with ambiguity and appreciates laws and rules that are clear and enforced. Denmark scores just 23, however, suggesting the Danes are more comfortable with a lack of structure and some unpredictability. Long-term orientation: This dimension describes the way a culture reflects on the past rather than looking to the future as a way to define itself. Low scores on this dimension suggest a society that is rooted in the past, with strong traditions and resistance to change. High scores, in contrast, suggest a pragmatic culture which seeks to prepare for the future in a more holistic and embracing way. Cultures with low scores on this dimension are not accustomed to the pursuit of results that take a long time to achieve. By sticking with what is known, they are habituated to the achievement of quick results and as such pursue the fast fix over investment in solutions that take time. With a score of 83, Germany is highly positioned on this dimension, signifying a culture that is practical and able to adapt tradition to accommodate the current situation, as well as being accustomed to the deployment of perseverance to achieve an outcome. Australia, conversely, scores just 21 on this dimension, suggesting Australians are less flexible when it comes to modifying habits and customs to accommodate change, as well as less interested in saving for the future. Indulgence: This dimension defines the extent to which a culture controls basic and natural human drives. An indulgent culture is comfortable enjoying life, with an overall tendency towards optimism and prioritisation of leisure time. In contrast, in a restrained culture, people try to control their desires and impulses, and such control is respected. With a relatively high score of 65, Ireland is a country where the culture accepts a willingness to realise individual impulses and desires, enjoy life and have fun. Latvia, however, scores just 13 on this dimension, indicating a society that does not value leisure time and has a tendency towards pessimism. The Hofstede scores are one way to show just how conspicuous variations between national cultures can be. Each dimension interacts with the other dimensions in different ways for different nations. While some are often correlated with others (for example, a high score in indulgence usually indicates a low score in long-term orientation), they are generally viewed as a set of complementary variables that together paint a colourful picture of the nuances that delineate one culture from another. This chapter seeks to examine expressions of national cultural constructs in transport with a particular focus on transport policy. Such an examination requires a case study, and for the purposes of this chapter, the case is the carriage of dogs on public transport.

Companion animal ownership – prevalence and benefits An underacknowledged accessory to the global trend towards urbanisation and ongoing economic growth (United Nations, 2018) has been an increase in pet ownership (Kestenbaum, 2018). Dogs are the world’s most popular pet, with over a third of the global population estimated to live with at least one dog (GfK, 2016). While the motivation to own a dog will vary between cultures, households and individuals, a compelling body of empirical research demonstrates objectively that dogs are good for human health (Wood et al., 2005) (see also Chapter 20). Dog ownership is associated with decreased 384

Public transport and travel with dogs

blood pressure (Wheeler & Faulkner, 2015), reduced risk of heart attacks (Arhant-Sudhir et al., 2011), improved survival rates (Friedmann & Thomas, 1995), increased physical activity (Cutt et al., 2008) and an increased sense of physical and psychological wellbeing (Wells, 2009). This work on the health benefits of dogs to their individual owners and human acquaintances is complemented by some research that suggests dogs also generate broader social sustainability (for example, social interactions, perceptions of neighbourhood friendliness and sense of community), which has benefits not only for individual dog owners but also extends to the wider community (Wood et al., 2007). Demonstration of the community-building benefits of dogs suggests that they are, increasingly, out and about with their owners in the public realm (see also Chapter 26). This propensity has also been well researched, with several studies exploring the way dogs inhabit and are accommodated within communities (Tissot, 2011; Urbanik & Morgan, 2013; Graham & Glover, 2014; Instone & Sweeney, 2014). Even conventionally, including a pet dog in one’s daily life involves the need to leave home with the dog from time to time, both for basic care (such as trips to the vet), for companionship (such as trips to a favourite dog park) and for exercise for the dog. If these trips cannot be accommodated on foot or by public transport modes, they will likely be accomplished by private car.

Pets travelling sustainably Despite the increasing popularity of dog ownership, and the emerging propensity for dogs to accompany their owners on trips outside of the home, very little research has been done on the way people travel with their dogs. Most research to date on the mobility practices of dogs has its focus on dog walking as a health-promoting activity for humans. Christian et al. (2013) for example, identified 29 peer-reviewed studies examining the relationship between dog ownership and physical activity through walking. This and similar reviews and empirical studies informed a scientific statement from the American Heart Association (Levine et al., 2013). This concludes that dog ownership is associated with reduced risk of cardiovascular disease. While most studies conclude that dog ownership does increase walking, very little attention is paid to the geography of dog walking beyond the home neighbourhood environment, including whether it is incorporated into the daily transport task or whether it becomes a trip-generating activity, for example, when the dog is taken to a favourite walking route or park.

Dogs and public transport: research to date Recognising this research gap, in 2015, the lead authors conducted a study of over 1,200 dog owners in Sydney, Australia. Over 95% of respondents reported travelling with their dogs at least once a week, generating over 9,500 trips per week. Almost half of these trips started in a private car, with each household surveyed making, on average 3.8 dog-related trips by private car per week (Kent & Mulley, 2017). This is reflective both of the dispersed geography and long distances which characterise travel in Sydney; however, it also reflects the fact that in Sydney, dogs are generally prohibited from accompanying their owners on public transport (Kent et al., 2020). Sydney’s public transport network consists primarily of heavy rail and bus services. Dogs, with the exception of service dogs, are entirely prohibited from travelling on the rail system. Dogs are permitted to use bus services if they are in a dog carrier; however, this is at the discretion of the bus driver, who can legally refuse access to a passenger with a dog. The need for a carrier and ability for the driver to refuse access ensures the restrictions of dogs on buses become a prohibition. It prevents transport of dogs too large for a dog carrier and erodes the certainty of 385

Kent, Mulley, Goh and Stevens

travel required for the service to be considered reliable from the perspective of the user. Complementing bus and heavy rail services in the Sydney network are ferry and light rail services: these allow dogs to ride on restricted routes and in certain areas. These services are restricted to very inner-Sydney travel and tend to require bus access to complete a journey. Thus, it is difficult for dog owners to plan and execute journeys which depend on public transport in Sydney. This chapter was motivated by the idea that something as (seemingly) obscure as the prohibition of dogs on public transport both represents and fuels the car dependency in Sydney and led to a review of dogs on public transport policy around the world being conducted. This review revealed a diverse array of policy approaches – some cities appear relatively open to the idea of dogs accompanying their minders on public transport, while others are entirely prohibitive. The variations in policy approaches are summarised to develop a dog-friendliness scoring system which is applied to each city, enabling analysis of the origins and geographical distribution of variation. Like most aspects of public transport policy, this variation is inevitably shaped by a complex amalgam of social, political, economic and built contexts, both past and present. This chapter recognises that the keeping of companion animals is a particularly culturally infused practice and concentrates on one specific aspect of context that, to date, has received very little attention in research on public transport policy: national culture.

Method As identified previously, this chapter aims to explore associations between one particular aspect of public transport policy – the carriage of dogs – and national cultural attributes. The review of dogs on public transport policies from a selection of cities was used to develop a scoring system to categorise the different policy approaches. This score was then supplemented by data on cultural attributes using the Hofstede model. Associations between these two variables were then analysed, the results of which are reported in the ‘Discussion’ section.

Public transport policy analysis For this study, data on policies governing the carriage of dogs on public transport came from the review of policies around the world first collected using a qualitative content analysis of posts from the website Travelnuity (Cleaver, n.d.). This site, maintained by an avid traveller and dog lover, has the aim of providing advice on international travel with a dog. The advice is primarily informed by the personal experiences of the site’s author, who has travelled extensively with her pet dachshund, Schnitzel. The information has a particular focus on Europe, Australia and the United States. For specific major cities within each country, the site provides advice on accommodation with dogs and dog-friendly outings and extends to the use of public transport systems with a dog. In January 2019, the lead author met with the author of the site to clarify its authenticity and confirm the processes used to collect the data contained on the site. Each post on Travelnuity was analysed for information on the use of public transport with dogs such as general rules, animal location, time restrictions and ticket rules. This information was then verified by direct examination of each public transport system’s website. Several cities not reviewed on Travelnuity were added, sourcing data from public transport websites in order to provide a more comprehensive dataset. To enable a deep policy analysis, however, the breadth of data analysed has been narrowed to travel within a contrasting set of European and Australian cities.1 In total, 130 public transport services operating in 48 different cities and 24 countries were reviewed. Of these, 81.5% of services reviewed operate in countries in Europe and the United 386

Public transport and travel with dogs

Kingdom, with the remaining 24% servicing Australian cities. Most operated in individual cities, and fewer than 20% were regional. Most services were either heavy (38%) or light rail (24%) services; however, bus (32%) and ferry (6%) services were also included. Cities ranged in population size from 149,000 (Darwin, Australia) to 8.2 million (London, UK), with an average population of 1.6 million. The policies analysed displayed several aspects of diversity which were surprisingly sophisticated, extending beyond simply prohibiting or permitting dogs to ride public transport. These degrees of difference are generally related to practical limits placed on dogs using public transport. The most extreme limit is the need to place the dog in an enclosed carrier. While this may not limit the carriage of very small dogs, it does restrict use of the service for dogs larger than can be physically carried. Less restrictive limits were those constraining the dog to one particular train carriage or area of the bus. Limits were also temporal, with dogs prohibited from riding public transport during peak hours. Finally, many policies also place financial limits by charging a fare for the dog. A 3-point ‘dog friendliness scale’ was developed to capture these nuances (Table 27.1), and this was applied to the policy governing each service. The Hofstede scores for national culture were then sourced for the host nation of each city, and it is this combination of the dog-friendliness scale and the Hofstede scores that forms the basis of our analysis, which is presented in the ‘Results’ section subsequently.

Results Almost two thirds of the transport services reviewed allowed dogs to travel with few restrictions (Level 3), as shown in Table 27.2. Only one of these services was in Australia, with the remainder located in the United Kingdom or Europe. All but three of the services entirely prohibiting dogs on public transport (Level 1) were in Australia. Of the 85 services permitting dogs on public transport, 57 enforced some kind of restriction while still allowing the dog to travel outside of a container. The most common limitation by far was the need to purchase a ticket for the dog, as shown by Table 27.3. Correlations between the Hofstede dimensions and the dog friendliness scores (as an ordinal scale) were examined using Spearman’s correlation coefficient. Table  27.4 shows that some cultural dimensions are correlated with the nature of public transport policies governing the

Table 27.1 The dog-friendliness scale Dog-friendliness scale

Description

Level 1: Prohibition

No pets allowed, or pets only allowed as checked luggage; includes pets not allowed on certain routes Travel restricted to pets that can fit in a carrier/container Minor restrictions included are: – fare charged (a ticket is required for dog) – location restrictions (pets may only travel in certain areas, such as the vestibule) – dangerous dogs restricted (restricted to pets with a pet passport or certain breeds restricted) – time restrictions (pets may only travel during specified nonpeak times)

Level 2: Restricted to container Level 3: Few or no restrictions

387

Kent, Mulley, Goh and Stevens Table 27.2 Distribution of dog friendliness scale Dog friendliness (3-point scale)

Number of cases

% of total

Level 1: Prohibition Level 2: Restricted to container Level 3: Little or no restriction

21 24 85

16.2 18.5 65.4

Dog travel restrictions

Number of cases

% of total

Time-based Location-based No dangerous dogs Pet fare charged

2 4 4 47

1.5 3.1 3.1 36.2

Table 27.3 Distribution of dog travel restrictions

Table 27.4 Correlations between pet friendliness scores and Hofstede dimensions Hofstede dimension

Spearman’s correlation coefficient

P

Power distance Individualism Masculinity Uncertainty avoidance Long-term orientation Indulgence

.216 –.488 –.139 .332 .608 –.444

.015* .000* .119 .000* .000* .000*

N = 127. * significant p ≤ 0.05

carriage of dogs, with five of the six dimensions having significantly strong correlations. Countries that currently welcome dogs on public transport are generally associated with higher scores on the power-distance, uncertainty avoidance and long-term orientation cultural dimensions. They have lower scores on the individualism and indulgence scales. The correlation statistic for the masculinity dimension was not significant.

Discussion Recalling the details for each cultural dimension outlined previously: • • • • •

A high score for power distance indicates a society where hierarchal order is accepted. A high score for uncertainty avoidance suggests a culture intolerant of unorthodox ideas and ways of doing things and anxious to control the future. A high score for long-term orientation suggests a pragmatic culture which seeks to prepare for the future in a more holistic and embracing way. A low score for individualism indicates a nation where people are bound by strong relationships into groups, which take care of individuals in return for loyalty. A low score for indulgence indicates a restrained culture, in which people try to control their desires and impulses and such control is respected. 388

Public transport and travel with dogs

The results suggest that countries where dogs are permitted to ride on public transport are countries where hierarchies are accepted, and restraint and a long-term view of things are both respected. Uncertainty is avoided, and people are cared for in groups rather than considered as individuals. The common thread emergent from this analysis is a sense of conservation and safeguarding of both the status quo and the collective. In many ways, it describes quite an unadventurous culture. These are not countries that routinely throw caution to the wind in the pursuit of short-term gratification or an enshrined sense of the protection of the rights of the individual. At first consideration, these are characteristics that suggest a situation more readily associated with public transport systems that prohibit, rather than welcome, dogs. To travel with a dog involves some objective risk (for example, even a well-trained dog can act impulsively when afraid). It is also a prioritisation of the rights of one passenger type (dog supporter) over another (dog opponent) and an indication of a society where public transport is equally accessible to all. Yet countries that allow dogs on public transport are willing and able to accept this risk and prioritisation and the resultant opening up of a service to all, even though such acceptance and action is not necessarily congruent with their approach to other issues. This finding has several implications; however, it first needs to be acknowledged that cultural dimensions provide the policy context rather than explaining it (as Hofstede notes, ‘National Culture cannot be changed, but you should understand and respect it’ [Hofstede Insights, n.d.]). The findings do not suggest that dogs are permitted on public transport because countries are conservative. The fact that countries are conservative yet still allow dogs on public transport, however, provides an insight into the way national culture shapes the environment in which public transport policy is made. The first implication is of relevance to countries who currently prohibit dogs on public transport. Policy makers in these nations can be reassured that the cultural context in which a policy change might occur need not be overtly speculative or experimental. Related to this is the possibility that dogs on public transport in countries where it is permitted are a common and everyday occurrence – it is an insignificant occasion and a triviality that need not be received with fuss or undue ceremony. The public transport system accommodates it, just like it accommodates the rest of everyday life. This suggests that, for these countries, public transport is a normalised way to travel and to accomplish the day-to-day tasks of modern life. In countries where dogs are prohibited, however, the system is less open and available and perhaps only suited to trips that are predictable, such as the journey to work.

Conclusion This chapter has used the case study of policies regulating dogs on public transport to explore the role of national culture in shaping public transport policy-making environments. It has found that allowing dogs on public transport is related to national cultures that are more conservative, with a long-term orientation and an acceptance of hierarchies. This finding has been used to suggest that dogs on public transport in these countries are a commonality – part of everyday life – just like the use of public transport is mainstream and uneventful. These findings demonstrate context at play in public transport policy and practice. While dogs on public transport is the example used here, there are a myriad of other seemingly inconsequential trips that make up modern life and that need to be accommodated by public transport services if they are to become a mainstream mode for travel. Trips with young children or the frail elderly, those to pick up bulky goods or those chained together using a bicycle are all examples of travel that for many simply construe daily life yet require a public transport system 389

Kent, Mulley, Goh and Stevens

that accommodates degrees of temporal and spatial flexibility. Exploring these less common trip generators, and seeking to understand the context in which they are regulated, can provide vital clues as to how public transport systems can become more useful. Future research is needed to reveal these less obvious mobility practices. This study has obvious limitations. First, its focus is geographically limited to two continents. Further research examining the public transport policy and national cultural contexts of other areas would provide deeper insights into the influence of culture on policy and vice versa. Second, understanding has been limited to using one index of culture and thereby has ignored the influence of other structural and social determinants of the policy environment. Third, the analyses do not take into account the historical political legacy of the policy-making process, the influence of which can override any attempt to understand how and why policy decisions are made. Finally, it should be noted that the analysis compares the policy environment for individual cities against averaged national scores. Cities, particularly major capital cities, are likely to have slightly different expressions of cultural indices when compared to the nation as a whole. As a postscript, this chapter was written when the COVID-19 pandemic was at its height. Public transport policy will need to adjust as societies transition to the “new normal”, coming to terms with the seismic decline in patronage and a future where passengers will be uncertain about the safety of public transport travel. For those cities where there are currently prohibitions, these new circumstances provide an opportunity to make the step to creating a more permissive policy environment for travel with dogs and allow such travel to become part of everyday life, as those cities which have a more conservative outlook have already done.

Note 1 Countries included in the analysis: Australia, Austria, Belgium, Bulgaria, Czech Republic, Denmark, Estonia, France, Germany, Greece, Hungary, Ireland, Italy, Latvia, Lithuania, Netherlands, Poland, Portugal, Romania, Slovakia, Slovenia, Spain, Sweden, United Kingdom.

References Aldred, R., & Jungnickel, K. (2014). Why culture matters for transport policy: The case of cycling in the UK. Journal of Transport Geography, 34, 78–87. doi:10.1016/j.jtrangeo.2013.11.004 Arhant-Sudhir, K., Arhant-Sudhir, R., & Sudhir, K. (2011). Pet ownership and cardiovascular risk reduction: Supporting evidence, conflicting data and underlying mechanisms. Clinical and Experimental Pharmacology and Physiology, 38(11), 734–738. doi:10.1111/j.1440-1681.2011.05583.x Ashmore, D. P., Christie, N., & Tyler, N. A. (2017). Symbolic transport choice across national cultures: Theoretical considerations for research design. Transportation Planning and Technology, 40(8), 875–900. doi:10.1080/03081060.2017.1355882 Ashmore, D. P., Pojani, D., Thoreau, R., Christie, N.,  & Tyler, N. A. (2018). The symbolism of ‘eco cars’ across national cultures: Potential implications for policy formulation and transfer. Transportation Research Part D: Transport and Environment, 63, 560–575. doi:10.1016/j.trd.2018.06.024 Ashmore, D. P., Thoreau, R., Kwami, C., Christie, N., & Tyler, N. A. (2020). Using thematic analysis to explore symbolism in transport choice across national cultures. Transportation, 47(2), 607–640. doi: 10.1007/s11116-018-9902-7 Beugelsdijk, S., & Welzel, C. (2018). Dimensions and dynamics of national culture: Synthesizing Hofstede with Inglehart. Journal of Cross-Cultural Psychology, 49(10), 1469–1505. doi:10.1177/0022022118798505 Christian, H. E., Westgarth, C., Bauman, A., Richards, E. A., Rhodes, R. E., Evenson, K. R.,  .  .  . Thorpe, R. J., Jr. (2013). Dog ownership and physical activity: A review of the evidence. Journal of Physical Activity and Health, 10(5), 750–759. doi:10.1123/jpah.10.5.750 Cleaver, S. (n.d.). Travelnuity: Dog friendly travel around the world. Retrieved February 2019–March 2020, from www.travelnuity.com

390

Public transport and travel with dogs Cutt, H., Giles-Corti, B., Knuiman, M., Timperio, A., & Bull, F. (2008). Understanding dog owners’ increased levels of physical activity: Results from RESIDE. American Journal of Public Health, 98(1), 66–69. doi:10.2105/AJPH.2006.103499 Dahl Wikstrøm, R., & Böcker, L. (2020). Changing suburban daily mobilities in response to a mobility intervention: A  qualitative investigation of an E-bike trial. Sustainability, 12(6), 2413. doi:10.3390/ su12062413 Dingil, A. E., Rupi, F., Schweizer, J., Stasiskiene, Z., & Aalipour, K. (2019). The role of culture in urban travel patterns: Quantitative analyses of urban areas based on Hofstede’s culture dimensions. Social Sciences, 8(8). doi:10.3390/socsci8080227 Friedmann, E., & Thomas, S. A. (1995). Pet ownership, social support, and one-year survival after acute myocardial infarction in the cardiac arrhythmia suppression trial (CAST). The American Journal of Cardiology, 76(17), 1213–1217. doi:10.1016/S0002-9149(99)80343-9 GfK. (2016). Pet ownership global GfK survey. Retrieved May  18, 2020, from https://geodata.gfk.com/ fileadmin/user_upload/website_content/Global_Study/Documents/Global-GfK-survey_Pet-Owner ship_2016.pdf Graham, T. M., & Glover, T. D. (2013). On the fence: Dog parks in the (un)leashing of community and social capital. Leisure Sciences, 36(3), 217–234. doi:10.1080/01490400.2014.888020 Haustein, S., & Nielsen, T. A. S. (2016). European mobility cultures: A survey-based cluster analysis across 28 European countries. Journal of Transport Geography, 54, 173–180. doi:10.1016/j.jtrangeo.2016.05.014 Hofstede, G. (1984). Culture’s consequences: International differences in work-related values. Sage Publications. Hofstede, G. (2001). Culture’s consequences: Comparing values, behaviors, institutions and organizations across nations. Sage publications. Hofstede, G., Hofstede, G. J.,  & Minkov, M. (2010). Cultures and organizations: Software of the mind. McGraw Hill. Hofstede Insights. (n.d.) National culture. Retrieved May 17, 2020, from https://hi.hofstede-insights.com/ national-culture Hopkins, D.,  & Stephenson, J. (2016). The replication and reduction of automobility: Findings from Aotearoa New Zealand. Journal of Transport Geography, 56, 92–101. doi:10.1016/j.jtrangeo.2016.09.005 Inglehart, R. F. (1990). Culture shift in advanced industrial society. Princeton University Press. Inglehart, R. F. (1997). Modernization and postmodernization: Cultural, economic, and political change in 43 societies. Princeton University Press. Inglehart, R. F. (2018). Cultural evolution: People’s motivations are changing, and reshaping the world. Cambridge University Press. Instone, L., & Sweeney, J. (2014). The trouble with dogs: ‘Animaling’ public space in the Australian city. Continuum Journal of Media & Cultural Studies, 28(6), 774–786. doi:10.1080/10304312.2014.966404 Kent, J. L., & Mulley, C. (2017). Riding with dogs in cars: What can it teach us about transport practices and policy? Transportation Research Part A, 106, 278–287. doi:10.1016/j.tra.2017.09.014 Kent, J. L., Mulley, C., & Stevens, N. (2020). Challenging policies that prohibit public transport use: Travelling with pets as a case study Journal of Transport Policy, 99, 86–94. doi:10.1016/j.tranpol.2020.08.024 Kestenbaum, R. (2018, November 27). The biggest trends in the pet industry. Forbes. www.forbes.com/sites/ richardkestenbaum/2018/11/27/the-biggest-trends-in-the-pet-industry/#2a532a44f099 Klinger, T., Kenworthy, J. R., & Lanzendorf, M. (2013). Dimensions of urban mobility cultures – A comparison of German cities. Journal of Transport Geography, 31, 18–29. doi:10.1016/j.jtrangeo.2013.05.002 Klinger, T., & Lanzendorf, M. (2015). Moving between mobility cultures: What affects the travel behavior of new residents? Transportation, 43, 1–29. doi:10.1007/s11116-014-9574-x Legacy, C., Curtis, C., & Scheurer, J. (2017). Planning transport infrastructure: Examining the politics of transport planning in Melbourne, Sydney and Perth. Urban Policy and Research, 35(1), 44–60. doi:10.1 080/08111146.2016.1272448 Levine, G. N., Allen, K., Braun, L. T., Christian, H. E., Friedmann, E., Taubert, K. A., . . . Lange, R. A. (2013). Pet ownership and cardiovascular risk: A scientific statement from the American heart association. Circulation, 127(23), 2353–2363. doi:10.1161/CIR.0b013e31829201e1 Mattioli, G., Anable, J., & Vrotsou, K. (2016). Car dependent practices: Findings from a sequence pattern mining study of UK time use data. Transportation Research Part A: Policy and Practice, 89, 56–72. doi:10.1016/j.tra.2016.04.010 Nello-Deakin, S., & Nikolaeva, A. (2020). The human infrastructure of a cycling city: Amsterdam through the eyes of international newcomers. Urban Geography. doi:10.1080/02723638.2019.1709757

391

Kent, Mulley, Goh and Stevens Oliver, J. D., & Lee, S. H. (2010). Hybrid car purchase intentions: A cross-cultural analysis. Journal of Consumer Marketing, 27(2), 96–103. doi:10.1108/07363761011027204 Reckwitz, A. (2002). Basic elements of a theory of social practices  – A  perspective in social theory. Zeitschrift Fur Soziologie, 32(4), 282–301. Sattlegger, L.,  & Rau, H. (2016). Carlessness in a car-centric world: A  reconstructive approach to qualitative mobility biographies research. Journal of Transport Geography, 53, 22–31. doi:10.1016/j. jtrangeo.2016.04.003 Scheiner, J., & Holz-Rau, C. (2007). Travel mode choice: Affected by objective or subjective determinants? Transportation, 34(4), 487–511. doi:10.1007/s11116-007-9112-1 Steg, L. (2005). Car use: Lust and must. Instrumental, symbolic and affective motives for car use. Transportation Research Part A-Policy and Practice, 39(2–3), 147–162. doi:10.1016/j.tra.2004.07.001 Syam, A. (2014). Cultural values: A new approach to explain people’s travel behaviour and attitudes toward transport mode [Unpublished doctoral dissertation, The University of Auckland]. https://researchspace.auckland. ac.nz/handle/2292/23559 Tissot, S. (2011). Of dogs and men: The making of spatial boundaries in a gentrifying neighborhood. City Community, 10(3), 265–284. doi:10.1111/j.1540-6040.2011.01377.x United Nations. (2018, May  16). 2018 revision of world urbanization prospects. United Nations Department of Economic and Social Affairs. www.un.org/development/desa/publications/2018-revision-ofworld-urbanization-prospects.html Urbanik, J., & Morgan, M. (2013). A tale of tails: The place of dog parks in the urban imaginary. Geoforum, 44, 292–302. doi:10.1016/j.geoforum.2012.08.001 van Acker, V., van Wee, B., & Witlox, F. (2010). When transport geography meets social psychology: Toward a conceptual model of travel behaviour. Transport Reviews, 30(2), 219–240. doi:10.1080/ 01441640902943453 Wells, D. L. (2009). The effects of animals on human health and well-being. Journal of Social Issues, 65(3), 523–543. doi:10.1111/j.1540-4560.2009.01612.x Wheeler, E. A., & Faulkner, M. E. (2015). The “pet effect”: Physiological calming in the presence of canines. Society and Animals, 23(5), 425–438. doi:10.1163/15685306-12341374 Wood, L. J., Giles-Corti, B., & Bulsara, M. K. (2005). The pet connection: Pets as a conduit for social capital. Social Science & Medicine, 61(6), 1159–1173. doi:10.1016/j.socscimed.2005.01.017 Wood, L. J., Giles-Corti, B., Bulsara, M. K., & Bosch, D. A. (2007). More than a furry companion: The ripple effect of companion animals on neighborhood interactions and sense of community. Society & Animals, 15(1), 43–56. doi:10.1163/156853007X169333

392

28 PUBLIC TRANSPORT USE IN LATER LIFE Charles Musselwhite and Maria Attard

Introduction The ability to travel and be mobile in later life is linked to a good quality of life (see HolleyMoore & Creighton, 2015). Furthermore, the importance of discretionary travel in later life has been highlighted as an important factor for the health and wellbeing of older people, particularly those who do not drive (Musselwhite, 2017). In view of the increase in travel and activities among this group, the role public transport plays in later life cannot be underestimated. More active lifestyles and car use throughout life have led to an increasing number of older drivers, in many western cultures (Mackett, 2018). But what about those older adults who cannot, or do not want to, drive? This chapter deals with the use of public transport in later life. It describes the main determinants of travel among this age group, briefly discusses some of the barriers to travel and focuses on the equity issues of public transport use among older adults. It focuses primarily on urban areas, even though problems and challenges are also experienced in non-urban and rural areas. The chapter uses evidence from United Kingdom and Europe primarily but also draws on evidence from further afield to attempt to better understand how public transport supports independent travel in later life. The concerns over the COVID-19 pandemic are discussed in the context of rising challenges which public transport users, particularly older adults, are and will continue to, face in the future. The aim here is to provide a comprehensive overview of the opportunities, barriers and challenges of public transport use in later life.

Mobility in later life Despite an overall reduction in out-of-home activity linked to age, there is an increase in and a need for discretionary travel among older adults in order for them to be part of society in a meaningful way. Older people travel mostly for shopping, leisure, medical care and religious activities. In the United Kingdom, Mackett (2018) identified shopping and social and leisure activities as the primary reasons for travel, and in the Netherlands, Boschmann and Brady (2013) found similar patterns of travel for the same age group. It is evident that as people age, leisure time and leisure travel increase. Average trip distances are shorter for older people but

393

Charles Musselwhite and Maria Attard

recreational travel distances are longer than other trips (Schmöcker et al., 2005). However, it is worth remembering that this will not be the case for all individuals everywhere. In Bangkok, for example, Srichuae et al. (2016) found medical care as the most common out-of-home destination above and beyond leisure activity. Despite the decrease in trips that occur in later life, walking remains the most important mode with older people, notably for leisure purposes and also often for the purpose of shopping (Musselwhite, 2017). Older people walk shorter distances, and they are also likely to own fewer cars per person. In the Netherlands, the decrease in the number of car trips (especially for compulsory activities) has been replaced by walking trips (Yang et al., 2013). Overall, however, walking, cycling and non-urban bus use have been in decline across all ages (Mackett, 2018). The decline in use of bus services has led to a reduction in the services offered, which in turn affects the availability of public transport as an option. This has significant consequences for older people when they have to give up driving and use other modes of transport, which they may not have used for many years (Musselwhite & Shergold, 2013). Older people are more likely than any other age group to suffer mobility deprivation, in that they cannot access the places they want because they cannot physically get to them (Mackett, 2018). Some of the research has shown that giving up driving because of older age is related to a decrease in wellbeing and an increase in depression and other related health problems, including feelings of stress and isolation and also increased mortality (Fonda et al., 2010; Musselwhite & Haddad, 2018; Musselwhite & Shergold, 2013). These macro-level changes in mobility due to ageing mask more complex behaviours, and, indeed, mobility of older people can be quite complex. Mifsud et al. (2017) adopted a multilevel conceptual ecological model to explain the determinants of travel behaviour which are affected by individual, sociocultural and environmental factors. The literature identifies age, gender, mobility tools, health, social issues, financial status, level of education and urban structure as main determinants of mobility in later life (see Mercado & Newbold, 2009). The focus on the use of public transport can also be seen through these factors. Schwanen et al. (2001) showed how transport mode choice in the Netherlands is determined by personal factors such as age, household composition and level of education, car ownership and the characteristics of the surrounding environment. Similarly, Kim and Ulfarsson (2004) found the same determinants for mode choice in Washington State. Furthermore, in this study, the proximity to public transport infrastructure was identified as a key factor increasing the use of public transport among older adults. This is also established by Beimborn et al. (2003). It is, however, contested in the study by Mifsud et al. (2017), which finds proximity to bus stop a non-significant determinant for public transport use in Malta, alongside other personal factors, including gender and level of education. Public buses play an important part in connectivity for older people, especially those who have given up driving. Bus use is especially high among older people where there are concessionary or free fares, as in the United Kingdom. Not only does the bus keep people connected, bus use is also correlated with health and wellbeing, being a protective factor in obesity for older people (Webb et al., 2011). Gender is another key determinant of mobility in later life, with women using public transport more than men (Siren & Hakamies-Blomqvist, 2004) either because they give up driving earlier or because they travelled more as passengers and used public transport throughout their life. Public transport remains a fundamental travel option for older people to remain mobile and reduce loneliness (Shrestha et al., 2016). Despite difficulties, there are circumstances, for example, low income or unsuitable alternatives, that restrict mobility to public transport modes (Beimborn et al., 2003), and whilst walking offers several benefits for older people’s physical, 394

Public transport use in later life

social and psychological wellbeing, there are several barriers which inhibit mobility and encourage older people to stay home (Musselwhite & Haddad, 2018; Mifsud et al., 2019).

Issues of equity in transport provision in later life Martens (2006) identified transport equity as a means of delivering social justice and reducing social exclusion. Social exclusion among older adults using public transport is due to limitations on mode choice, with age, income and access to private transport being main factors leading older people to rely solely on public transport (see also Chapter 26). System reliability, environmental impact and accessibility often affect the opportunities and capabilities of older people to use these services (Bocarejo & Oviedo, 2012). Within public transport systems, there are still a number of barriers that inhibit equitable provision of services among the different user groups. These groups are distinguished by age but also by gender, as it is well known that older women spend more time using public transport than men (Siren & Hakamies-Blomqvist, 2004). Older people suffer from difficulties and insecurities when using some public transport services, such as inaccessible infrastructure (distant stops, poor walking environments, high steps at bus stops, inaccessible buses or trains) and poorly designed and maintained interiors that do not fully support people with restricted or limited mobility (Wardman, 2001) (see also Chapter 31). New ‘intelligent’ mobility services being promoted as ‘disruptive’ rely heavily on a population that handles technology (e.g. smartphones) with ease. Real-time booking facilities like those available for Uber and ViaVan, but also many others, require ownership and skills which many older people do not possess. Intelligent mobility is using new technologies and approaches, such as connected and autonomous transport systems and new data-driven personalised on-demand transport, supported with open data platforms, to move people and goods around more easily, more efficiently and in a more environmentally friendly way (see also Chapter 40 and Chapter 41). These new integrated technologies include connected and autonomous transport, new mobility services and open data platforms. In an ageing society, the needs of older people are rarely considered in the development of intelligent mobility (see van Hoof et al., 2018), with a focus on supporting interurban business and commuting (Parkhurst et al., 2014). Although older people are more likely to commute compared to previous generations, their use of the mobility network is varied, and they utilise a variety of modes throughout the day for a wide range of purposes (Musselwhite & Curl, 2018). Hence, these solutions are less likely to be of value to an ageing population, especially as they have not involved older people in the design of such systems (Musselwhite, 2018).

Public buses There are still many barriers to using a bus that prevent or make it difficult for older people to use it. Having free journeys or reduced fares for older people increases the use of public transport, and as the density of bus stops increases, use amongst older people makes a difference as well. However, features in the surrounding environment may be equally important. Gilhooly et al. (2002) found the highest barrier to public transport use amongst older people was personal security in the evening and at night (79.8% of over-70s agreed), followed by public transport running late and having to wait. In addition, the journey is improved for older people if it is viewed as being “seamless” from door to door (Maynard, 2009). A report using accompanied journeys in London highlighted similar problems for older people, including crowds at the bus stop or on the bus, prams taking up the seats or area at the front of the bus, steps up to the bus 395

Charles Musselwhite and Maria Attard

being too high (or driver stopping too far from the kerb) and fear of falling over when the bus moves off (TfL, 2009). For example, Broome et al. (2010) in an Australian study found that for older people, driver friendliness, ease of entry/exit and information usability were prioritised barriers and facilitators for older people. Age UK London quantified this by surveying bus driving behaviour in 550 journeys in inner London and 541 journeys in outer London in 2011. In 42% of cases, passengers were not given enough time to sit down before the bus was driven away from the stop. In 25% of the cases, the bus did not pull up tight to the kerb at the bus stop (Age UK, 2011). Another barrier to use often overlooked is anxiety related to norms of use (Musselwhite, 2018; Mifsud et al., 2019). It maybe that on giving up driving, the older person has not used a bus in many years and is unsure how to use the service. They may be unaware of improvements in the system, such as real-time and en-route bus stop information. In addition, older people may be anxious about social norms, for example, the normal departure time (is it sooner than advertised?), what times of day are less busy, is there seat availability, are buses accessible, how much can be carried? (Musselwhite, 2018). Providing training and information for older people about how to use buses can help overcome these barriers. Travel training schemes may involve partnering new users with experienced users to learn from them, in so-called “buddy schemes” (Brown, 2010; Musselwhite, 2010; Stevens et al., 2013). Such schemes have proved successful for some novice users (see Ormerod et al., 2015; Stevens et al., 2013). Alternative approaches have also been suggested, such as more generic training in the use of public transport associated with a program of giving up driving, as suggested by Liddle et al. (2006) and Musselwhite (2010). Mifsud et al. (2019) go on to identify additional social norms, in particular those related to influence from family members, friends and health professionals, as major barriers to travel, particularly alone using buses. Another aspect is the relational nature of the bus, the use of the bus as a third space, to people-watch, a space for recreation and seeing the world pass by, rather than just a vehicle to get from A to B. The ability to interact with other passengers can be seen positively by older people (Mackett, 2018). Social support for using the bus, such as “Bus Buddy” schemes, can pair inexperienced or new bus users with an experienced user, which can help grow bus user confidence (Phillips & McGee, 2018). The social interaction between the individual and the driver is also vital and can make or break a journey if the driver is rude or discourteous (Musselwhite, 2018). A cheery or sympathetic driver attuned to older people’s needs, who, for example, says hello and asks how they are, allows the passenger to take time when boarding or may wait for the passenger to sit down before driving off is invaluable.

Dedicated public transport services As an alternative to conventional public bus services, there can be provision of specialist transport services, often operating door to door for people who cannot access public or private transport, known as specialist transport service, community transport or demand-responsive transport (see also Chapter 17). Such services usually run on demand rather than following a scheduled timetable. They are often run through a licence by a local authority by a third sector or charitable organisation. Such services are often viewed as a lifeline for older people who would otherwise be housebound (ECT, 2016). There are direct improvements on people’s health through affording greater access to GP and hospital services, for example. The ability of older people to maintain their regular appointments through dedicated services ensures a quicker diagnosis of illness or signs of loneliness (ECT, 2016). Other similar community services exist in many different forms across countries. They are offered at a very local level to the 396

Public transport use in later life

general public or specific groups, such as older adults or people with disabilities (Weckström et al., 2018). While community transport is prevalent in the United Kingdom and Australia, paratransit provides for similar needs in the United States. Overall, however, community transport remains somewhat sparse (Mulley et al., 2012). There are also barriers to its use. Because services are dependent on third sector and/or charity provision, they can be somewhat fragmented across larger areas. They might be short lived, often existing around a particular one-off or smallscale funding offer and key individuals. Indeed, success of these services across systems has varied, with many failing or requiring subsidy beyond the first few years of operation to survive (Mohamed et al., 2019). People who may well benefit from such a service can sometimes feel the service is not for people like them; there is sometimes the perception that it is for people with disabilities rather than for everyone with accessibility issues (Musselwhite, 2018). Some have even flagged issues of personal safety. Frequently, there is a lack of information and as a result much misunderstanding of the service (Parkhurst et al., 2014; Ward et al., 2013). Journeys typically are based around providing transport to shops, services and doctors and hospitals, but there need to be more “discretionary” journeys provided to places of leisure and fun (Musselwhite, 2017).

Taxi and shared services As an alternative to public transport, older people often turn to other forms of mobility where public transport is unavailable, inaccessible or too expensive; these include specialised transport services, taxis and other shared services. The use of taxis among older people is highly linked to income levels and transport used during the life course. Car ownership and the ability to afford a taxi are linked to higher levels of income. Driving cessation also leads many to resort to taxis even though these can be seen as extravagant by some, especially for discretionary trips (Davey, 2007). Taxi-like services, offered today by new shared mobility providers, have opened new opportunities for older people to extend their travel beyond the utilitarian at a more affordable price. Motorcycle taxi services are found in high numbers to transport people around congested and busy city centres in low- to middle-income countries (LMICs). This can still be expensive and unaffordable to the majority of older people or be inaccessible, with older people unable to physically get on a motorcycle or back of a pick-up truck (Porter et al., 2018).

Income, concessionary fares and free bus programmes The amount public transport is used and the use of buses, in particular, are linked to income levels among older people. Whilst high income levels are related to car ownership and high levels of mobility, lower income is related to public transport use and walking (Kim & Ulfarsson, 2004). Both Schmöcker et al. (2008) and Truong and Somenahalli (2011) showed how higher income negatively affected public transport use in London and Adelaide. Lucas et al. (2019) noted that lower-income groups of older people relied more on local neighbourhood activities which they could reach on foot. Various authors have also looked at the advantages linked to concessionary fare and free bus programmes for older people. Truong and Somenahalli (2011) analysed the effects of a concessionary fare introduced in July 2009 on public transport use in Adelaide and, similarly to Mackett (2013) in the United Kingdom, found that bus use increased. People aged over 65 years of age, or those of any age with a disability, have been entitled to travel free of charge on any 397

Charles Musselwhite and Maria Attard

off-peak local public bus service in England since 2007, thanks to the English National Concessionary Travel Scheme. Around 76% of all women and 79% of men take up the free bus pass in England, compared to 61% and 50% in 2005 – the year before free local travel. Humphrey and Scott (2012) suggest that ownership of a free bus pass is higher (around 80–82%) among those on lower income (less than £15,000). This group is also more likely to use the bus once a week than those on higher incomes, who use it less frequently. The most commonly reported activity older people cite as their destination on buses across all these surveys is shopping, followed by social and leisure, day trips and visiting friends, then medical, meaning people are socially connected and hence experience reduced isolation (Mackett, 2013). Mackett (2013) also notes how such bus journeys support the volunteering and caring work older people undertake, which would otherwise not take place. Webb et al. (2011) examined three waves of English Longitudinal Study of Ageing data (2004, 2006 and 2008) and found a link between eligibility for a free bus pass and increased use of buses. They also found those who used the bus more often had a reduced chance of becoming obese, suggesting that using the bus is associated with physical activity such as walking to and from bus stops and allowing people to engage with more activity. Dargay et al. (2010) modelled bus use against what would have happened if no free bus pass had been introduced. They found journeys made are more numerous and also often longer in duration and distance (Dargay et al., 2010). The findings suggest the number of bus stages (groups of bus stops) travelled by older people increased by 45.4% in rural areas and 26.5% in urban areas.

Railways In the United Kingdom, there has been unprecedented growth in rail travel over the past 20 years. The number of passengers on UK railways has grown significantly, both absolutely and in terms of percentage of overall distance travelled. In the United Kingdom, rail travel has increased almost 60% between 1995/97 and 2017 (DfT, 2019). Against a backdrop in an increase in the number of older people in the United Kingdom and an increase in the amount of travel per person for this age group, the number of older people using the railway has not increased at the same rate. In the last ten years, there has been a 23% increase in rail travel distance per person across all ages, and while for the over-70s, overall mobility has grown by 11%, rail travel per person for this age group has fallen by 10% (DfT, 2019). This warrants further investigation; why are older people less likely to use the railway for their journeys, and how can stations and rolling stock be made more age friendly? Interestingly, long-distance bus or coach travel could overcome some of the issues older people have when using railways, including having a designated seat and not having to transfer buses or stow luggage, which is the responsibility of the driver. However, there is little to no research on long-distance bus or coach use in later life. Very little prevailing literature exists on the subject of older rail travellers. Sundling et al. (2014) found for older people with high functional ability, the main barriers to travelling by train were travel costs and low punctuality, and for those with low functional ability, health was the main barrier. Musselwhite (2018) used Passenger Focus 2015 survey figures, which survey passenger transport satisfaction use for government and the industry in the United Kingdom and stratified them by age. The study showed that older people seem to have higher satisfaction with their train travel, including being positive about price and the overall journey experience. This may be because older people are making more recreational journeys than the average train user – leisure users are more satisfied than those using it for work and commuters across all ages (Ormerod et al., 2015). In addition, older people prioritise getting a seat on a train more 398

Public transport use in later life

highly than younger passengers do, and from 60 years onwards, it becomes more important than the cost of the ticket. This may in part be that older people are able to make more of cheaper off-peak and advanced tickets, as well as railcards reducing fares. Older rail passengers are more likely than rail passengers in general to want to be kept informed about the journey and any delays and, compared to younger and middle age rail passengers, less likely to be concerned about the availability of free Wi-Fi. Overall, for older passengers, there is more concern with the state and cleanliness of the carriage and of the toilet facilities, and they prioritise these over the length of journey and frequency of services, possibly showing their more intermittent and leisure use. Station design is also important for older people. Stations ideally should have indoor waiting areas and toilets wherever possible, and accessibility is a mandatory issue, with lifts preferred on all stations that require access by stairs (Musselwhite, 2018). Older people can feel more vulnerable on trains and in stations, and visibility is key to this (Gilhooly et al., 2002); older people feel more vulnerable are less likely to use the station where there is a lack of staff, lack of other passengers, lack of lighting and dark enclosed waiting areas (Cozens et al., 2004). This can be placated somewhat through better design. An excellent project was carried out on the valley lines in Wales which resulted in better designed stations. In particular, Dingle Road station (South Wales, UK) was redesigned from a station that contained two old, enclosed shelters to one that contained a see-through shelter, which improved feelings of safety for all age groups (Cozens et al., 2004). Similarly, the presence of staff at railway stations and on board can help create a feeling of security among older people (Musselwhite, 2018). Older people, more than other groups, value the importance of staff to help them at rail stations and on train services. They are more likely to trust information if it is given from authority figures, for example, railway staff, and like the staff to be friendly and approachable (Musselwhite, 2018). They use staff for timetable information, especially if trains get delayed or things go wrong, whereas other groups are more likely now to use mobile information, communication technologies, and apps (Gilhooly et al., 2002; Musselwhite, 2011). They use staff for backing up information they see on screens or hear over the announcements, which they trust less than younger groups. They also often want staff available should they need help carrying cases. The provision of information is vital, both on trains and at stations, especially for less frequent users and for when things do not go according to plan. Lamont et al. (2013) investigated the extra planning that dyslexic travellers needed for planning a journey by rail and how the intervention of staff could help remove concern and anxiety over the journey. Similar may be found for older people, who may have poorer eyesight; have issues with cognitive overload, memory, concentration and learning and could become overwhelmed by poorly designed signage. Around 30% of those aged over 70 have a mobility issue (DfT, 2019; Mackett, 2018). Older people with mobility issues are potentially more likely to have issues boarding and alighting trains, especially traversing a step up or down from the train or a gap between the train and the platform, and have more issues in crowded space and on a moving train. This is exacerbated when carrying luggage, as figures suggest around 28% of older people have issues with carrying items (Mackett, 2018). In on-train audits carried out with older people on a major network in the United Kingdom, Musselwhite (2019) found older people were overrepresented in passenger accidents, including boarding; slips, trips and falls on trains and slips, trips and falls at the station frontage, car park and concourse. Conclusions suggested better signage, lighting and places to sit and rest were needed for older people throughout the station, with more level boarding between platform and train needed as standard. 399

Charles Musselwhite and Maria Attard

COVID-19 and the future of public transport During the COVID-19 pandemic, starting in 2020, the impact on transport and health is highly evident (Musselwhite et al., 2020). There were plenty of discussions taking place at the time of writing this chapter regarding the impact of COVID-19 and public transport use. Tirachini and Cats (2020) show very clearly the decline in public transport use in many cities around the world as a reaction to the COVID-19 pandemic. The link between virus infection and public transport use is not always clear cut, but some studies suggest a link; for example, Troko et al. (2011) found those who got acute respiratory infections (ARIs) in winter were more likely to be using bus or tram use in the five days before symptom onset. Epidemiologists believe COVID-19 passes through the air in tiny droplets, which are easier to pass on in enclosed spaces and can live for hours or even days on hard surfaces. The greatest risk for infectious diseases in public transport is that people sit or stand in proximity in a closed environment (Edelson & Phypers, 2011). These vehicles can become a significant source of microorganisms when passengers do not close their mouths when coughing and sneezing. Handrails, ticket machines, smart-card machines, doors, handles, windows, panels, floors, elevators and seats are areas that can host infectious micro-organisms. To stop the spread of the disease, during the COVID-19 outbreak, epidemiologists are encouraging social distancing, meaning people should keep around 6 feet apart or more from others, a measure at odds with the use of public transport (Musselwhite et  al., 2020). Further to the restrictions on travel and the explicit discouragement of use by public authorities in a number of countries, including the United Kingdom (DfT, 2020) and the United States (Centres for Disease Control and Prevention, 2020), public transport has also been seen as a riskier means of transport for COVID-19 contagion (Tirachini & Cats, 2020). Budd and Ison (2020) report on a survey in the United Kingdom carried out in May 2020 which revealed a high percentage of people still unwilling to use public transport and only 18% using public transport after the lifting of restrictions. Following studies in epidemiology, one of the common measures provided by the authorities is internal cleaning and sanitation of public transport vehicles, used daily by thousands of people. They are disinfecting handrails, ticket machines, doors, handles, windows, panels, elevators and seats more frequently. They are also fumigating buses frequently both inside and outside, alongside main interchanges and bus stops. Another measure taken by some authorities is installing hand-sanitizing units inside public transport facilities. It is unclear whether these measures protect to the desired level. Also, it is questionable whether frequent cleaning and sanitation by staff is sustainable over time, as it demands much human resources and its logistics might be complicated. Although it was found that the use of crowded public transport vehicles can be associated with the acquisition of infectious diseases, it can be argued that these findings do not support the effectiveness of suspending mass urban transport systems as a pandemic countermeasure aimed at reducing or slowing population spread. It is evident that whatever the relevance of public transport to individual-level risk, household exposure most likely poses a greater threat (Cooley et al., 2011). All this is even more relevant for older people, considered most vulnerable in society in terms of contagion and severity of health outcome. The future of public transport is currently being debated in the context of new rules and restrictions imposed by various health authorities around the world. Whilst lockdown measures have decimated public transport use in many cities, leading to financial difficulties among operators and some closure of services, the restrictions in the post-pandemic months make public transport use difficult. Social distancing 400

Public transport use in later life

rules, requirements to wear masks, the switching off of air-conditioning units and the sanitisation required on buses are just a few examples of the challenges which lie ahead for public transport operators and users. And even after implementing new layouts, providing contactless door sensors, installing hand sanitizer dispensers and other measures to reduce contagion, the effectiveness of these and how they are perceived by the public remain largely unknown (Budd & Ison, 2020). So far the use of public transport has declined in most countries that had high levels of COVID-19 and were placed in lockdown by governments. Other than Japan, which saw at most 20% reduction at first, most countries saw at least a decline of 50% or more and, although this is recovering, rates of use still vary around a 20% (e.g. Hong Kong) to 60% (e.g. United Kingdom) reduction in use (Tirachini & Cats, 2020). The socioeconomic effect of the pandemic will be significant on public transport in a postCOVID-19 world. Patterns of inequality between those who cannot avoid using public transport and those who can are emerging. There is still too little research on how older people have been affected. It is certain that returning to normal use of public transport in the post-pandemic months will be difficult, and new procedures and maybe services will be required to support high-risk groups such as older people.

Conclusion With an increasingly mobile older generation, one that is increasingly wedded to the car, and as cars become increasingly automated, it remains to be seen what role public transport will play in the lives of older people in future societies. Public transport in the present day has a role as a great social leveller, especially where it is free or cheap to use, meaning those who cannot afford a car can remain mobile. Public transport keeps older people active, connected to things they want to do while reducing loneliness and isolation, and it can be seen to be a protective factor helping maintain health in later life, reducing obesity and potentially reducing heart disease and associated illness. The emergence of COVID-19 has resulted in a huge reduction in use of public transport, along with associated anxiety, even among older people who have been regular users. Recovery of public transport as mobility for older people will require huge public trust and confidence, along with reinstating services and provision. Hence, there is a real need for public policy to help maintain public transport for an ageing population. Many interventions aimed at improving public transport are at a utilitarian level, helping accessibility of older people, but the psychosocial aspects of public transport, including status, roles and norms, cannot be underestimated and should not be forgotten. Modern technologies such as intelligent mobility approaches, for example, could be used to improve mobility for older people yet rarely consider the needs of older people. Hence, there is a need to design these technologies with the needs of older people in mind, perhaps better still through co-design approaches with older people. There are evidently gaps in research addressing long-distance bus and coach use with older people and more research needed on older people’s use of railways, especially around why train travel use has not increased for older people as it has for other age groups. Research must take into account the wider social and psychological issues associated with public transport use and not simply identify accessibility barriers, and it should place public transport use within the wider social context of older people’s lives. Transport research is always highly contextualised and influenced hugely by geography and culture, and research examines generalisability of 401

Charles Musselwhite and Maria Attard

findings for other areas with care. So, finally, there is a plea for research on public transport and ageing in countries outside the United Kingdom and Europe, looking especially to research in the Global South.

References Age UK. (2011). Past campaigns  – On the buses. Retrieved August  4, 2020, from www.ageuk.org.uk/ london/projects-campaigns/our-campaigns/info/on-the-buses/ Beimborn, E. A., Greenwald, M. J., & Jin, X. (2003). Transit accessibility and connectivity impacts on transit choice and captivity. Transportation Research Record: Journal of the Transportation Research Board, 1835, 1–9. Bocarejo, S. J. P., & Oviedo, H., D. R. (2012). Transport accessibility and social inequities: A tool for identification of mobility needs and evaluation of transport investments. Journal of Transport Geography, 24, 142–154. Boschmann, E. E.,  & Brady, S. A. (2013). Travel behaviors, sustainable mobility, and transit-oriented developments: A travel counts analysis of older adults in the Denver, Colorado metropolitan area. Journal of Transport Geography, 33, 1–11. Broome, K., Nalder, E., Worrall, L., & Boldy, D. (2010). Age-friendly buses? A comparison of reported barriers and facilitators to bus use for younger and older adults. Australasian Journal of Ageing, 29(1), 33–38. Brown, R. (2010). Maintaining community mobility in Portsmouth after retirement from driving: How do people find out about and gain confidence in using alternative forms of transport? [MSc thesis, Accessibility and Inclusive Design, University of Salford]. Budd, L., & Ison, S. (2020). Responsible transport: A post-COVID agenda for transport policy and practice. Transportation Research Interdisciplinary Perspectives, 6, 100151. Centres for Disease Control and Prevention. (2020). Travel during the COVID-19 pandemic. Retrieved August 4, 2020, from www.cdc.gov/coronavirus/2019-ncov/travelers/travel-in-the-us.html Cooley, P., Brown, S., Cajka, J., Chasteen, B., Ganapathi, L., Grefenstette, J., . . . Wagener, D. K. (2011). The role of subway travel in an influenza epidemic: A  New York City simulation. Journal of Urban Health: Bulletin of the New York Academy of Medicine, 88(5), 982–995. Cozens, P., Neale, R., Hillier, D., & Whittaker, J. (2004). Tackling crime and fear of crime while waiting at Britain’s railway stations. Journal of Public Transportation, 7(3), 23–41. Dargay, J., Ronghui, L., & Toner, J. (2010). Concessionary bus fares in England: The impact on bus travel by the over-60s. Paper presentation. European Transport Conference. Davey, J. A. (2007). Older people and transport: Coping without a car. Ageing & Society, 27(1), 49–65. DfT (Department for Transport). (2019). National travel survey. Retrieved July 20, 2020, from www.gov. uk/government/collections/national-travel-survey-statistics DfT (Department for Transport). (2020). Guidance coronavirus (COVID-19): Safer travel guidance for passengers. Retrieved August  4, 2020, from www.gov.uk/guidance/ coronavirus-covid-19-safer-travel-guidance-for-passengers ECT (Ealing Community Transport). (2016). Why community transport matters? ECT Transport Report. ECT. Edelson, P. J., & Phypers, M. (2011). TB transmission on public transportation: A review of published studies and recommendations for contact tracing, Travel Medicine and Infectious Disease, 9(1), 27–31. Fonda, S. J., Wallace, R. B., & Herzog, A. R. (2010). Changes in driving patterns and worsening depressive symptoms among older adults. The Journal of Gerontology, Series B: Psychological Sciences and Social Sciences, 56(6), 343–351. Gilhooly, M., Hamilton, K., O’Neill, M., Gow, J., Webster, N., Pike, F., & Binbridge, C. (2002). Transport and ageing: Extending quality of life for older people via public and private transport (ESRC Award Reference Number L480 25 40 25). Retrieved July  27, 2020, from https://bura.brunel.ac.uk/bitstream/2438/1312/1/PDF%20ESRC%20Transport%20Final%20Report.pdf. Holley-Moore, G., & Creighton, H. (2015). The future of transport in an ageing society. Age UK. Humphrey and Scott. (2012). Older people’s use of concessionary travel. NatCen, Age. Kim, S., & Ulfarsson, G. (2004). Travel mode choice of the elderly: Effects of personal, household, neighbourhood, and trip characteristics. Transportation Research Record: Journal of the Transportation Research Board, 1894, 117–126.

402

Public transport use in later life Lamont, D., Kenyon, S., & Lyons, G. (2013). Dyslexia and mobility-related social exclusion: The role of travel information provision. Journal of Transport Geography, 26, 147–157. Liddle, J., McKenna, K.,  & Bartlett, H. (2006). Improving outcomes for older retired drivers: The UQDRIVE program Australian Occupational Therapy Journal, 53, 303–306. Lucas, K., Stokes, G., Bastiaanssen, J., & Burkinshaw, J. (2019). Inequalities in mobility and access in the UK transport system. Evidence Review Foresight Government Office for Science. Mackett, R. (2013). The impact of concessionary bus travel on the wellbeing of older and disabled people. Transportation Research Record: Journal of the Transportation Research Board, 2352, 114–119. Mackett, R. (2018). Older people’s travel and its relationship to their health and wellbeing. In C. Musselwhite (Ed.), Transport, travel and later life, Transport and Sustainability (Vol. 10, pp. 15–36). Emerald Publishing Limited. Martens, K. (2006). Basing transport planning on principles of social justice. Berkeley Planning Journal, 19, 1–17. Maynard, A. (2009). Can measuring the benefits of accessible transport enable a seamless journey? Journal of Transport and Land Use, 2(2), 21–30. Mercado, R. G., & Newbold, K. B. (2009). Car driving and public transit use in Canadian metropolitan areas: Focus on elderly and role of health and social network factors (No. 243). McMaster University. Mifsud, D., Attard, M., & Ison, S. (2017). To drive or to use the bus? An exploratory study of older people in Malta. Journal of Transport Geography, 64, 23–32. Mifsud, D., Attard, M.,  & Ison, S. (2019). An exploratory study of the psychological determinants of mobility of older people in Malta. Research in Transportation Business and Management, 30, 100373. Mohamed, M. J., Rye, T., & Fonzone, A. (2019). Operational and policy implications of ride sourcing services: A case of Uber in London, UK. Case Studies on Transport Policy, 7, 823–836. Mulley, C., Nelson, J., Teal, R., Wright, S., & Daniels, R. (2012). Barriers to implementing flexible transport services: An international comparison of the experiences in Australia, Europe and USA. Research in Transportation Business & Management, 3, 3–11. Musselwhite, C. B. A. (2010). The role of education and training in helping older people to travel after the cessation of driving. International Journal of Education and Ageing, 1(2), 197–212. Musselwhite, C. B. A. (2011). The importance of driving for older people and how the pain of driving cessation can be reduced. Journal of Dementia and Mental Health, 15(3), 22–26. Musselwhite, C. B. A. (2017). Exploring the importance of discretionary mobility in later life. Working with Older People, 21(1), 49–58. Musselwhite, C. B. A. (2018). Community connections and independence in later life. In E. Peel, C. Holland, & M. Murray (Eds.), Psychologies of ageing: Theory, research and practice (pp. 221–252). Palgrave Macmillan. Musselwhite, C. B. A. (2019). Towards an age friendly railway. Swansea University, UK. Unpublished report. Musselwhite, C. B. A., Avineri, E., & Susilo, Y. (2020). The coronavirus disease COVID-19 and implications for transport and health. Journal of Transport and Health, 16, 100853. Musselwhite, C. B. A.,  & Curl, A. (2018). Geographical perspectives on transport and ageing. In A. Curl & C. B. A. Musselwhite (Eds.), Geographies of transport and ageing (pp. 3–24). Palgrave Macmillan. Musselwhite, C. B. A., & Haddad, H. (2018). The travel needs of older people and what happens when people give up driving. In C. Musselwhite (Ed.), Transport, travel and later life, transport and sustainability (Vol. 10, pp. 93–115). Emerald Publishing Limited. Musselwhite, C. B. A.,  & Shergold, I. (2013). Examining the process of driving cessation in later life. European Journal of Ageing, 10(2), 89–100. Ormerod, M., Newton, R., Philips, J., Musselwhite, C., McGee, S., & Russell, R. (2015). How can transport provision and associated built environment infrastructure be enhanced and developed to support the mobility needs of individuals as they age? Future of an Ageing Population: Evidence Review Foresight Government Office for Science. Parkhurst, G., Galvin, K., Musselwhite, C., Phillips, J., Shergold, I., & Todres, L. (2014). Beyond transport: Understanding the role of mobilities in connecting rural elders in civic society. In C. Hennesey, R. Means, & V. Burholt (Eds.), Countryside connections: Older people, community and place in rural Britain (pp. 125–157). Policy Press. Phillips, J., & McGee, S. (2018). Future ageing populations and policy. In A. Curl & C. B. A. Musselwhite (Eds.), Geographies of transport and ageing (pp. 227–250). Palgrave Macmillan. Porter, G., Tewodros, A., & Gorman, M. (2018). Mobility, transport and older people’s well-being in subSaharan Africa: Review and prospect. In A. Curl & C. B. A. Musselwhite (Eds.), Geographies of transport and ageing (pp. 75–100). Palgrave Macmillan.

403

Charles Musselwhite and Maria Attard Schmöcker, J. D., Quddus, M. A., Noland, R. B., & Bell, M. G. H. (2005). Estimating trip generation of elderly and disabled people: Analysis of London data. Transportation Research Record: Journal of the Transportation Research Board, 1924, 9–18. Schmöcker, J. D., Quddus, M. A., Noland, R. B., & Bell, M. G. H. (2008). Mode choice of older and disabled people: A case study of shopping trips in London. Journal of Transport Geography, 16, 257–267. Schwanen, T., Dijst, M., & Dieleman, F. M. (2001). Leisure trips of senior citizens: Determinants of modal choice. Tijdschrift Voor Economische en Sociale Geografie, 92(3), 347–360. Shrestha, B. P., Millonig, A., Hounsell, N. B., & McDonald, M. (2016). Review of public transport needs of older people in European context. Population Ageing, 1–19. Siren, A., & Hakamies-Blomqvist, L. (2004). Private car as the grand equaliser? Demographic factors and mobility in Finnish men and women aged 65+. Transportation Research Part F: Traffic Psychology and Behaviour, 7, 107–118. Srichuae, S., Nitivattananon, V., & Pereram, R. (2016). Aging society in Bangkok and the factors affecting mobility of elderly in urban public spaces and transportation facilities. IATSS Research, 40(1), 26–34. Stevens, N., Battellino, H., & Pedler, K. (2013, October). Evaluating travel training in northern Sydney: Methodological and practical issues. In Australasian transport research forum (ATRF), 36th, 2013. ATRF. Sundling, C., Berglund, B., Nilsson, M. E., Emardson, R., & Pendrill, L. R. (2014). Overall accessibility to traveling by rail for the elderly with and without functional limitations: The whole-trip perspective. International. Journal of Environmental Research and Public Health, 11(12), 12938–12968. TfL (Transport for London). (2009, June). Older people’s experience of travelling in London. Transport for London. www.tfl.gov.uk/cdn/static/cms/documents/older-peoples-transport-experiences-report.pdf. Tirachini, A., & Cats, O. (2020). COVID-19 and public transportation: Current assessment, prospects, and research needs. Journal of Public Transportation, 22(1), 1–34. Truong, L. T., & Somenahalli, S. (2011, September). Exploring mobility of older people: A case study of Adelaide. In Australasian transport research forum (ATRF), 34th. ATRF. Troko, J., Myles, P., Gibson, J., Hashim, A., Enstone, J., Kingdon, S., . . . Nguyen Van-Tam, J. (2011). Is public transport a risk factor for acute respiratory infection? BMC Infectious Diseases, 11, 16. Van Hoof, J., Kazak, J. K., Perek-Białas, J. M., & Peek, S. T. M. (2018). The challenges of urban ageing: Making cities age-friendly in Europe. International Journal of Environmental Research and Public Health, 15, 2473. Ward, M., Somerville, P., & Bosworth, G. (2013). ‘Now without my car I don’t know what I’d do’: The transportation needs of older people in rural Lincolnshire. Local Economy, 28(6), 553–566. Wardman, M. (2001). A review of British evidence on time and service quality valuations. Transportation Research Part E: Logistics and Transportation Review, 37(2–3), 107–128. Webb, E., Netuveli, G., & Millett, C. (2011). Free bus passes, use of public transport and obesity among older people in England. Journal of Epidemiology and Community Health, 66(2), 176–180. Weckström, C., Mladenović, M. N., Ullah, W., Nelson, J. D., Givoni, M., & Bussman, S. (2018). User perspectives on emerging mobility services: Ex post analysis of Kutsuplus pilot. Research in Transportation Business and Management, 27, 84–97. Yang, D., Timmermans, H., & Grigolon, A. (2013). Exploring heterogeneity in travel time expenditure of aging populations in the Netherlands results of a CHAID analysis. Journal of Transport Geography, 33, 170–179.

404

29 PARKING PROVISION, PARKING DEMAND AND PUBLIC TRANSPORT ACCESSIBILITY Barbara T.H. Yen and Corinne Mulley Introduction The built environment plays a significant role in physical activities (Badland et al., 2010). Many recent studies have investigated the relationship between built environment and travel behaviour (e.g., mode choice), for example, Cao et al., 2006; Handy et al., 2006; Librett et al., 2007; Badland et al., 2010. Important components of the built environment are parking and land use, whether residential or business, and are considered in this chapter. Whilst parking provision could be seen as an encouragement to the use of private vehicles, it has in many places been the centrepiece of travel demand management (TDM) strategies used to reduce traffic congestion, to address traffic-related air pollution, to assist in funding selected infrastructure or to assist public transport access to facilities (Ison & Mulley, 2014). It is in this context, within broader TDM strategies, that parking facility provision can complement public transport accessibility when parking development strategy is integrated with public transport strategy to facilitate first-mile parking needs and improve the attractiveness of public transport services. A typical example here is a park-and-ride facility, defined as an intermodal journey that occurs when a private vehicle is used for the first mile of a trip to enable the use of public transport service for the ongoing part of the journey (Parkhurst & Meek, 2014) (see also Chapter 7). Examples include providing access to a major shopping centre (e.g., Westfield Garden City Shopping Centre in Brisbane, Australia), public transport nodes (e.g., Opatov Station in Prague Metro Station, Czech Republic) or a heritage city centre (e.g. York in the United Kingdom). Parking is a sensitive issue which has impacts on users (e.g., shoppers, commuters), stakeholders (e.g., retailers, employers) and policy makers (e.g., local authorities). Transport and land use planners typically work closely with the community of stakeholders to help shape the built environment, such as parking provision and transport systems (Yen et al., 2020). However, these processes tend to be dominated by those stakeholders wanting a higher level of involvement because decisions about parking affect their residential or business interests (Marshall et  al., 2012). An example here are those urban restaurants that often cluster together in a specific precinct for scale economics in order to attract more customers (Parsa et al., 2005). For these restaurants, accessibility will have an impact on customers’ activity patterns, travel behaviour and patronage of restaurants. In many such restaurant precincts, restaurateurs seek to maximise the 405

Barbara T.H. Yen and Corinne Mulley

accessibility to their business by private car, with demands to increase parking provision based on a belief that this is one of the most effective ways of enhancing customer satisfaction and encouraging return patronage. Parking as a TDM measure can be categorised into different types, including economic (e.g., parking charges), land use (e.g., car free development), substitution of communications for travel (e.g., Park and Ride) and regulation (e.g., parking control) (Ison & Rye, 2008). The relationship between parking provision and public transport can be competing (e.g., parking control to limit private car use and improve public transport patronage) or complementary (e.g., provide park-and-ride to increase the accessibility of public transport service). This chapter considers both. However, in the case where there is competition between parking provision and public transport, it is also important to understand the need for parking. This chapter uses a case study in Brisbane, Queensland, Australia, to explore parking needs in a business precinct (the restaurant precinct in this case) and to understand whether the business owners who argue for parking provision, potentially at the expense of public transport access, have the right perceptions of their customer parking needs. The chapter is structured as follows: the next section provides the literature context to understand how the built environment (in terms of parking provision and land use) can assist public transport accessibility and how urban parking provision in business precincts can be measured and assessed. This discussion is followed by an empirical study that considers parking and public transport accessibility and how it can be enhanced when need and stakeholder perceptions are better understood. The final section discusses the implications for policy with the evidence on parking need and concludes with policy implications and thoughts for further research areas.

Parking and public transport Parking provision and accessibility of public transport systems TDM measures are typically employed via a package of tools to address the issue under consideration. In this context, parking has long been used as one of the TDM tools. Ison and Rye (2008) provide a selection of TDM measures aimed at influencing travel behaviour. There are a number of parking-related options, including parking charges, park-and-ride facilities and parking controls. Parking charges can be used as a ‘stick’ to discourage access by private car. Indeed, the cost or availability of parking is often cited as the main reason the private car is not used for particular journeys. In contrast, subsidising public transport is a ‘carrot’ to encourage public transport use by influencing the relative costs of the different modes. Parking regulations can be used to discourage car use, and more recently, the imposition of a workplace parking levy in the United Kingdom has been used to drive travel behaviour change and to underpin funding for better public transport (Dale et al., 2014). Further, there are several objectives for parking concepts, including meeting the parking demand of local residents, commercial traffic and distributing limited parking spaces according to priorities, as well as using parking as a way to control traffic and reduce congestion. Alongside these, a land use approach to TDM might be to introduce park-and-ride facilities to expand the catchment areas of public transport and provide incentives to commuters and other users who drive private vehicles to change to more sustainable transport modes (e.g., heavy rail) as part of their journey (Hamer, 2010). However, Park and Ride has not been shown to be unambiguously useful at eliminating negative externalities (Parkhurst, 1995). 406

Parking provision and demand

Figure 29.1 The components of park-and-ride Source: Adapted from: Meek (2008)

Policymakers and transport planners emphasise the direct impacts of park-and-ride on reducing traffic and/or congestion, since this, as a TDM strategy, reduces the number of cars using the road network. Park-and-ride is also one of the main interfaces between parking management and public transport, since private car travellers transfer onto public transport. Meek (2008) has disaggregated the park-and-ride concept into three elements, including what is described as a ‘private transport mode terminal’, which is effectively an ordinary car park, public transport access and planned service (Figure 29.1). From this conceptual presentation, it is clear that a park-and-ride facility assists with the utilisation of multimodal transport journeys (Krygsman & Dijst, 2001). Park-and-ride facilities are usually located at the edge of urban areas or in suburban areas, especially for low population density areas where there is enough physical space for provision. Park-and-ride facilities provide access to various public transport modes. In Europe, park-andride is usually provided by light rail, bus and heavy rail. In Australia, park-and-ride facilities are usually heavy rail or bus based and are located in suburban areas. In particular, bus based parkand-rides are popular in Brisbane, Australia, where the bus service is a bus rapid transit (BRT) system (known as the Busway) with limited bus stops at higher service frequencies. A similar case can be found in United Kingdom, where bus-based park-and-ride services are generally operated by dedicated (often branded) buses serving only the park-and-ride sites at high-demand destinations (Meek, 2008; Meek et al., 2011; Mingardo et al., 2015). As a supply-side management policy, park-and-ride provision is probably the most important proactive implementation. However, in terms of public transport access and parking, to consider only the supply side ignores the way the supply needs to meet the needs of car drivers and other stakeholders using our cities.

Parking provision and parking needs The previous section reviewed parking from a supply-side perspective. This section considers the demand side and the relationship between parking provision and parking needs (see Bates, 2014). Parking provision is only considered an issue by users, in particular, if parking need is 407

Barbara T.H. Yen and Corinne Mulley

larger than parking provision. This frequently occurs in the more dense urban areas (e.g., central business district, CBD) and where other stakeholders, such as local businesses, often overestimate parking need because of an inflated view of the importance and economic contribution of private vehicle drivers on the one hand while on the other hand overlooking the contribution of shoppers travelling on foot, by bike and public transport (see, for example, Jones et al., 2007; Rye et al., 2008; Clean Air Partnership, 2010; Stantec, 2011). Alongside this, the peer-reviewed literature has been paying increasing attention to ‘subjective’ factors such as attitudes (but also perceptions, preferences, life stages, lifestyles) (Mokhtarian & Cao, 2008; Schwanen & Lucas, 2011; Van Acker et al., 2011; De Witte et al., 2013; Van Acker, 2015; Van Acker et al., 2016). In the context of parking, the grey literature has been more active in exploring the views of businesses towards parking and complements the peerreviewed literature. Looking internationally, the conclusion is that worldwide businesses, with the exception of those in Auckland, New Zealand, appear to overestimate the benefits to the business owner of parking and underestimate the role of more sustainable modes (Table 29.1). However, the reports forming the basis of Table  29.1 focus on general businesses whose customers might need or might perceive a need to carry a large number of goods or larger-sized goods. There are clearly limitations as to what a shopper is able to carry if walking, cycling or taking public transport, and this might be the reason the overestimation for car customers and underestimation for sustainable travel customers occurs. An Australian study found that amount spent did indeed vary by the shoppers’ mode of transport in the restaurant precinct of Lygon Street area in Melbourne (Lee & March, 2010). The intercept survey study compared expenditure per hour and found customers travelling by car spent most ($65/hr), followed by those on foot ($58/hr), bike ($47/hr) and public transport ($41/hr). However, car drivers made fewer trips during the week compared to the more sustainable or greener modes of travel. So parking need is more complex than just varying by mode, as the frequency of need also has to be taken into account. This chapter now explores the relationship between parking supply, need by mode and frequency in the empirical setting of restaurant precincts in different areas in Table 29.1 Comparison of actual mode share versus business expectations in various studies City

Actual

Business expectations

Reference

Graz, Austria

Foot 44%, Bike 8%, Bus 16%, Car 32%

Foot 25%, Bike 5%, Bus 12%, Car 58%

Bristol, England

Foot 55%, Bike 10%, Bus 13%, Car 22%

Foot 42%, Bike 6%, Bus 11%, Car 41%

Auckland, New Zealand

Car 66%, Foot 22%, Bike 2%, Bus 9%, Train 2%, Other 2% (including skateboards and scooters) Shopper surveys identified only 29% drove and parked in nearby street

Car 58%, Foot 29%, Bike 2%, Public transport 9%, Other 2% (including skateboards and scooters) 44% of business respondents thought 50% of customers drove and parked in nearby street

Sustrans, 2003 cited in Fleming (Allatt) et al. (2013) Sustrans, 2003 cited in Fleming (Allatt) et al. (2013) Fleming (Allatt) et al. (2013)

Edinburgh, Scotland

Source: Yen et al. (2015)

408

Rye et al. (2008)

Parking provision and demand

Brisbane, Queensland, Australia. The aim is to provide an evidence-based approach to assessing how parking need, parking supply and public transport accessibility interact with the aim of improving public transport when need and stakeholder perceptions are better understood.

Case study The empirical setting is a number of restaurant precincts in Brisbane, Queensland, Australia. These locations are typical of the way in which Australian cities are compartmentalised into suburbs, with the areas closer to the CBD rearranging themselves into precincts  – typically for food and eating. The case study relies on evidence from a parallel survey for both dining customers and restaurateurs to investigate if there is a gap between customers’ actual travel behaviour and restaurateurs’ perceptions. The following section outlines the case study area and is followed by details of the survey, data acquisition and the presentation of descriptive statistics. The modelling is then discussed, together with the results and commentary.

Study area As with many Australian cities, restaurants cluster into precincts within the inner-city area. In Brisbane, there are three streets where most businesses are restaurants, including Boundary Street, West End (650 m from the bridge leading to the CBD); Eagle Street, Brisbane and Caxton Street (450 m from CBD). These are shown in Figure 29.2.

Figure 29.2 Location of three restaurant precincts in inner-city Brisbane

409

Barbara T.H. Yen and Corinne Mulley Table 29.2 Restaurant types and definition in study area Restaurant type

Number

%

Definition

Fast food

6

6

Fast casual

22

21.5

Casual dining

28

27.5

Family style

23

22.5

Fine dining Total

23 102

22.5 100

Orders are made at the counter with fast service food partly prepared off-site. The cost is higher than the fast food, and the food is mostly prepared on-site. Table services are provided; cost is moderate with possible bar services. Provide full bar services with alcohol. Table services provided, cost moderate and may not provide bar services. Full services provided with waiters, fine décor and bar services.

Source: Counts in street undertaken in 2015

The restaurants within these precincts are divided into five separate types of restaurants, as reported in Table  29.2, which, in addition, provides some descriptive statistics of each area. Apart from fast food restaurants, all other restaurant types are present in roughly the same portion. Around and within these precincts, there are a number of different options for customers who arrive by private car, including on- and off-street parking (i.e., at-grade outdoor parking, at-grade covered parking, multiple-level parking and underground parking). The parking fees vary by parking type and location so that the parking fee in Eagle Street is the highest (around AUD$30–75), since it is located in the heart of Brisbane’s CBD. In contrast, Caxton Street has the lowest parking fee (AUD$11), and it is the only street with off-street parking. All streets are served by train, ferry, bus and active travel opportunities, including Brisbane’s CityCycle public bicycle hire system that encourages more people to cycle around the inner city. Further, Brisbane also provides a high-frequency bus service, the CityGlider buses. The CityGlider service links to the CityCycle bike hire scheme. CityCycle provides bike hire stations near CityGlider stops. Eagle Street benefits in addition from the free City Loop bus service and the free ferry service (City Hopper) provided by Brisbane City Council. Figure 29.3 shows the route maps for both City Loop bus and City Hopper ferry services, showing that customers wanting to go to Eagle Street can have easy access to both services.

Survey and data statistics This study uses an intercept survey method in the three restaurant precincts, starting on March 1, 2015, and lasting for a week. A survey of restaurateurs was undertaken in parallel. In total, there were 300 valid customer responses (394 responses with 76% completed) and 40 valid restaurateur responses (44 response with 91% completed). Table 29.3 provides descriptive statistics from the customer survey that included four sections (i.e., sociodemographics, trip characteristics, perceptions and dining habits). It is interesting to note that 64% of the survey respondents had access to a car, but only 18% of them actually travelled to the restaurant precincts by car. For car drivers, access was described by the majority as convenient (46%), with a much smaller proportion of car drivers claiming a lack of public transport options (23%). For most customers, the travel time was less than 20 minutes, with travel cost of less than $10 AUD. A question about how more visits could be encouraged had 410

Parking provision and demand

Figure 29.3 Free City Centre Free Loop bus services routes and City Hopper ferry services route around Eagle Street Source: Brisbane City Council and Google Map

Table 29.3 Descriptive statistics of restaurant customers (N = 300) Social demographics Occupation Manager Professional Technician Administration Sales Labour Student Stay-at-home parent Retired Other Gender Male Car owner Yes Education level Primary school High school Trade/technical College graduate/certificate Under/postgraduate degree or higher

Value

Trip characteristics 15% 24% 10% 18% 10% 14% 5% 1% 2% 1% 51% 64% 1% 19% 18% 21% 36%

Value

Travel time Average travel time (mins) Travel cost Average travel cost (AUD dollar) Travel mode Car Walk Cycle Bus Train Ferry Taxi Scooter/motorcycle Dining habit Visiting time Average visit times (mins) Visiting spend Average visit spend (AUD dollar) Restaurant type Fast food Fast casual

23.8 7.3 18 23 7 19 17 5 6 5

96.5 32.9 8 13

(Continued)

411

Barbara T.H. Yen and Corinne Mulley Table 29.3 (Continued) Social demographics

Value

Trip characteristics

Value

Not completed anything Nil response Average no. of family members in household Average no. of children (< 18) Average age of respondent Average annual income before tax (AUD$) Perception Travelled by car Lack of public transport Convenient Don’t like public transport Dropping and/or picking up items Travelling with kids Passenger mobility impairment Own mobility impairment Ways to encourage visits* Wider footpath Improved bicycle accessibility Pedestrian-only space Reduced car traffic Improved car parking Improved bus service Improved train service

3% 2% 2.23 0.45 37.73 $60,264

Casual dining Family style Fine dining Nil response

42 22 14 1

23% 46% 10% 2% 11% 6% 2% 16% 20% 16% 41% 18% 41% 29

responses heavily dependent on better public transport access by bus (41%) and train (29%). Respondents also indicated that reduced car traffic would also encourage more trips. Car parking featured with a much lower response in favour of improved car parking (18%). In terms of dining habit, only a few customers (4%) had a visit duration of less than 30 minutes, and most customers spent around $16–50 dollars per visit. Respondents preferred the casual dining–style restaurant more (42%), followed by a family-style one (22%).

Customers’ mode choice and restaurateurs’ perceptions Restaurateurs in Brisbane are constantly requesting more parking space to serve their customers, since they perceive that most of their customers are coming by private cars. However, the survey in this study tells a different story: only 18% of customers actually access the restaurant precincts by private car, even though 64% of respondents were car owners. Figure 29.4 shows the customers’ actual mode choice and restaurateurs’ perceptions of how their customers access the precincts. There is an obvious gap between them, especially for the private car mode and public transport mode. A similar situation happens to revenue share or spending at restaurants (Figure 29.5). Restaurateurs expect customers who drive private cars to spend more than those who use other modes. However, from the survey, customers who walk and take public transport modes are spending similar amounts to car-driving customers.

412

Parking provision and demand

Figure 29.4 Mode share for restaurateur perceptions and revealed customer travel

Figure 29.5 Revenue share for restaurateur perceptions and revealed customer travel

Modelling process and results From a customer point of view, parking need is driven by their travel behaviour and, in particular, by the choice of mode. The evidence here is derived from a series of nested logit (NL) models. The choice modelling approach allows an understanding of how variables influence a customers’ mode choice in order to understand parking need. In the modelling process, a customer’s travel behaviour is determined by the trip characteristics, customer’s perceptions of transport facilities, dining habits, location and social demographic variables. Table 29.3 shows

413

Barbara T.H. Yen and Corinne Mulley

a low mode share for some transport modes (e.g., 5% for ferry), so the travel modes have been grouped into four different model selections (alternatives), including private vehicle (including both car and scooter/motorcycle), public transport (including bus, train and ferry), active travel (including walk and cycle) and taxi. A nested model approach is taken here, and, after investigating a number of options, the analysis proceeds with a model that has one nest with the cost travel of travel modes including public transport, car and taxi, while active travel is included as a single alternative. This model structure suggests that travellers would decide whether to travel because this costs money and then decide which mode to use. Table 29.4 shows model results of NL models. The first model is the basic model which explains the mode of travel only by reference to trip characteristics. However, as the literature context explains previously, the choice of mode to access facilities can be influenced by why the access is necessary (dining habit and location) as well as the perceptions and sociodemographics of the respondent. The second model includes variables relating to perceptions, dining habits, locations and sociodemographics. Both models have an acceptable fit, with the dissimilarity Table 29.4 Estimation results for standard NL models Variables Constants for access mode Private vehicle Active Public Trip characteristic Average travel costa Mode speedb Perception Convenient_ private vehicle Improve public service_ public Dining habit Average spendc_ private vehicle Average spend_ active Average spend_ taxi Visiting frequency_ active Restaurant type (family)_private vehicle Restaurant type (family)_public Location Eagle St._ public Social demographic Age_ active Family_ private vehicle Dissimilarity (t-value vs. 1) Final log-likelihood Likelihood ratio Adjusted likelihood ratio

Basic NL model 3.15(1.11) 5.78(1.25) 7.33(1.82)** –0.84(–7.41)*** 1.43(1.21)

NL model 5.35(0.39) 7.40(0.95) 9.27(1.48)* –0.62(–6.35)*** 1.31(2.54)*** 6.68(4.47)*** 0.88(1.95)** 2.58(1.46)* 1.16(1.72)** 4.10(3.69)*** 0.91(1.72)** 1.97(2.06)*** –1.15(–1.84)** 0.88(1.82)**

5.71(3.09)*** –278.88 0.31 0.30

–0.26(–1.97)*** 2.61(3.44)*** 1.36(3.67)*** –230.27 0.42 0.41

Note: t-value in parentheses; *** indicates 0.05 level of significance; ** indicates 0.10 level of significance; * indicates 0.15 level of significance. a: Average travel cost = travel cost (AUD dollar)/travel time (minutes). b: Mode speed = travel distance (km)/travel time (minutes). c: Average spend = visiting spend (AUD dollar)/visiting time (minutes).

414

Parking provision and demand

estimate significantly different from unity at a 99% level of significance and the model having all variable sets in showing a better fit. The reference mode is taxi due to its low market share (6%). In the basic model, there are two generic variables, with the same coefficients for all alternatives: these are average travel cost and travel speed. As expected, average travel cost (travel cost/travel time [minutes]) has a negative impact, and mode speed (travel distance/travel time [minutes]) has a positive effect on the utility of travel mode alternatives. The result indicates that customers would prefer to choose their travel mode with lower cost and higher speed. In the second model, these two trip characteristic variables have the same impact on utility but with smaller absolute values, as might be expected since other explanatory variables are significant. The other four sets of explanatory variables are treated as alternative-specific variables, so there is a different coefficient for each mode alternative. A negative coefficient indicates that an increase in the value of this variable decreases the utility for the travel mode and thus decreases the probability of that mode being chosen, provided all else remains unchanged. The reverse is true for a positive coefficient. The perception variable can be viewed as a proxy for how customers value the built environment in terms of the land use in the precincts, including the presence of parking. Not surprisingly, if private vehicle users consider convenience their main reason to drive, they would then drive more (positive parameter coefficients indicate the utility of car travellers is increased). For public transport travellers, there is a positive coefficient for public transport services being improved suggesting they will visit more if public transport service is improved. As to dining habits, the results show that customers with higher spend would prefer to access by taxi, private car or active transport as their travel mode, especially for taxi users, which has a much larger coefficient. Further, although there are five type restaurants, only the family dining restaurant has significant impact on utility. Thus, a dummy variable of “Restaurant type (family)” is used, taking the value of 1 if the respondent is dining in a family dining restaurant but 0 otherwise. For customers dining in family dining restaurants, they prefer to drive (positive coefficient) rather than taking public transport (negative coefficient). This is likely because dining in family restaurants would mean that the travelling was as a family group, where if there is access to a car, it is likely to be most convenient. Despite all restaurant precincts being in inner Brisbane, the results indicate the importance of location on customers’ travel mode choice. In particular, customers dining at Eagle Street prefer to choose public transport modes which have particularly good public transport access. In terms of social demographics, customers with more dependents prefer to access the restaurant precincts by private car, mirroring the result of family dining customers’ mode choice which is driving private car, too. In terms of age, younger customers would prefer to travel via active transport modes, such as walking and cycling.

Discussion and conclusions Parking can be used as a key strategy within TDM packages to influence the mode choice of travellers. However, taking this land use response and using parking as a strategy to reduce car use often creates tensions between policy makers and businesses. From the policy maker’s point of view, restrictions on parking can reduce the traffic level, and, indeed, park-and-ride strategies can both reduce traffic levels and encourage more sustainable travel for part of the journey. Understanding parking issues requires a better understanding of parking demand and business perceptions rather than simply concentrating on influencing parking supply. The empirical study in this chapter is the result of a parallel survey between restaurateurs and customers to assess how parking and public transport accessibility can be enhanced when need and stakeholder 415

Barbara T.H. Yen and Corinne Mulley

perceptions are better understood. The evidence from this case study is that reducing car traffic in the precincts was the number-one way (along with improving bus services) of encouraging more visits. However, from the business owners’ perspective, parking is seen as necessary to the economic vitality of their businesses. The case study provides sufficient evidence to show that, in line with the evidence provided in the literature section, businesses do overestimate the number of people accessing their businesses by car and underestimate the number of customers arriving by more sustainable modes. Perhaps more importantly, businesses do not appear to take the frequency of visit into account when claiming that customers arriving by car spend more. The authority responsible for parking – the local government in this case – needs to spend more effort on the difficult task of providing evidence to businesses so as to help change stakeholder perceptions. The case study also sheds some light on the relationship between the built environment (expressed as the different attributes of the restaurant precincts), parking and public transport access. The modelling results show that of the three precincts, it is only Eagle Street that is a significant location and this only to diners arriving by a public transport mode. All precincts in the case study areas have good public transport access, with a wide range of public transport services, including train, ferry and bus (e.g., CityGlider). Further, all areas have taxi ranks, bicycle racks, motorcycle parking spaces and access to the CityCycle routes. However, Eagle Street in particular benefits from the two free public transport services of the City Centre Free Loop (bus) and City Hopper (Ferry), making it the precinct with the best public transport access. Eagle Street does, however, have the highest parking charges. Overall, this suggests that parking need is inversely related to public transport access. Indeed, as city centre locations, the survey suggests that implementing a park-and-ride strategy that would simultaneously reduce traffic flow and increase the supply of public transport could lead to a greater number of visits to the precincts. Further, the modelling results also show that younger customers prefer to travel via active transport modes, such as walking and cycling. This can be explained by the other built environment characteristic of Boundary Street in West End. Currently, it has eight hostels (backpacker dormitories) and four hotels/ motels within 200 metres, accommodating more than 1,600 people (16% of West End [local suburb]’s population). We would expect Boundary Street could attract a certain amount of customers who choose to walk or cycle, and so parking need is lower and less parking needs to be provided. As indicated earlier, parking provision can influence public transport in two ways, providing a complementary or competing situation. The literature demonstrates that being used as part of a TDM strategy, the supply side tends to be considered in isolation without linking parking need to parking supply. The policy challenge is the way in which parking and driving of private cars create a competing situation with public transport. Parking can be used as a way of ‘pushing’ travellers to use more public transport, but as a policy, this can cause conflict with the business communities, who believe it is important to provide for car-based customers and, because local business tends to overestimate the importance of driving customers, they argue for more parking provision. Further, parking only becomes an issue if parking demand is larger than supply, and this usually happens in urban or suburban areas. Parking policy as a TDM tool needs to proactively respond to parking needs to maximise land use efficiency, as these urban or suburban areas have limited space, instead of reactively providing parking facilities with passive parking policies (e.g., parking regulation and control). The case study in this chapter provides an evidence base for policy makers to make more plans for parking and public transport to have a complementary relationship. For example, if or when there is a new city or redevelopment plan being prepared, the results of this chapter suggest that policy should conduct user segmentation 416

Parking provision and demand

analysis for both business and final users within the planning area so as to understand the structure of demand structure for parking and public transport to provide a more realistic plan. Future research could be informed by case studies which focus on the whole journey or trip chain leading to the access of facilities. This way, parking demand and supply and the interaction of the built environment can be more holistically understood in their role to promote public transport accessibility. As a final note, as this chapter was written before the announcement of the COVID-19 pandemic, the main conclusion from this chapter that private transport modes mode may not be as important as is thought might need qualifying. The COVID-19 outbreak has had a significant impact on public transport, identifying it as a high-risk transport mode. If this becomes a more permanent feature, future research on mode choice in the parking context will need to take this into account.

Acknowledgements An earlier version of this study was presented at the Australian Transport Research Forum in 2015.

References Badland, H. M., Garrett, N., & Schofield, G. M. (2010). How does car parking availability and public transport accessibility influence work-related travel behaviors? Sustainability, 2(2), 576–590. Bates, J. (2014). Parking demand. Parking issues and policies. In S. G. Ison & C. Mulley (Eds.), Parking: Issues and policies. Emerald Publishing Limited. Cao, X., Handy, S. L., & Mokhtarian, P. L. (2006). The influences of the built environment and residential self-selection on pedestrian behavior: Evidence from Austin, TX. Transportation, 33(1), 1–20. Clean Air Partnership. (2010). Bike lanes, on-street parking and business. Year 2 report: A study of Bloor street in Toronto’s Bloor West Village. Clean Air Partnership. Dale, S., Frost, M., Ison, S., & Warren, P. (2014). Workplace parking levies: The answer to funding large scale local transport improvements in the UK? Research in Transportation Economics, 48, 410–421. De Witte, A., Hollevoet, J., Dobruszkes, F., Hubert, M., & Macharis, C. (2013). Linking modal choice to motility: A comprehensive review. Transportation Research Part A: Policy and Practice, 49, 329–341. Fleming (Allatt), T., Turner, S., & Tarjomi, L. (2013). Reallocation of road space (Research Report 530). NZ Transport Agency. Hamer, P. (2010). Analysing the effectiveness of park and ride as a generator of public transport mode shift. Road & Transport Research, 19(1), 51–61. Handy, S., Cao, X. Y., & Mokhtarian, P. L. (2006). Self-selection in the relationship between the built environment and walking – Empirical evidence from northern California. Journal of the American Planning Association, 72(1), 55–74. doi:10.1080/01944360608976724 Ison, S. G., & Mulley, C. (2014). Parking: Issues and policies. Emerald Publishing Limited. Ison, S. G.,  & Rye, T. (2008). TDM measures and their implementation. In S. Ison  & T. Rye (Eds.), The implementation and effectiveness of transport demand management measures: An international perspective (pp. 1–130). Ashgate Publishing Limited. Jones, P., Roberts, M., & Morris, L. (2007). Rediscovering mixed-use streets: The contribution of local high streets to sustainable communities. Joseph Rowntree Foundation. Retrieved April 16, 2020, from www.jrf.org. uk/publications/contribution-local-high-streets-sustainable-communities Krygsman, S., & Dijst, M. (2001). Multimodal trips in the Netherlands – Microlevel individual attributes and residential context. Transit Planning, Intermodal Facilities, and Marketing, 1753, 11–19. Lee, A., & March, A. (2010). Recognising the economic role of bikes: Sharing parking in Lygon Street, Carlton. Australian Planner, 47(2), 85–93. Librett, J. J., Yore, M. M., Schmid, T. L., & Kohl, H. W. (2007). Are self-reported physical activity levels associated with perceived desirability of activity-friendly communities? Health Place, 13(3), 767–773. doi:10.1016/j.healthplace.2006.07.003

417

Barbara T.H. Yen and Corinne Mulley Marshall, N., Steinmetc, C., & Zehner, R. (2012). Community participation in planning. In S. Thompson & P. Maginn (Eds.), Planning Australia: An overview of urban regional planning (2nd ed., pp. 276–293). Cambridge University Press. Meek, S. (2008). Park and ride. In S. G. Ison & T. Rye (Eds.), The implementation and effectiveness of transport demand management measures: An international perspective (pp. 165–184). Ashgate Publishing Limited. Meek, S., Ison, S., & Enoch, M. (2011). Evaluating alternative concepts of bus-based park and ride. Transport Policy, 18(2), 456–467. Mingardo, G., van Wee, B., & Rye, T. (2015). Urban parking policy in Europe: A conceptualization of past and possible future trends. Transportation Research Part A: Policy and Practice, 74, 268–281. Mokhtarian, P. L., & Cao, X. (2008). Examining the impacts of residential self-selection on travel behavior: A focus on methodologies. Transportation Research Part B: Methodological, 42(3), 204–228. Parkhurst, G. (1995). Park and ride: Could it lead to an increase in car traffic? Transport Policy, 2(1), 15–23. Parkhurst, G., & Meek, S. (2014). The effectiveness of park-and-ride as a policy measure for more sustainable mobility. In S. G. Ison & C. Mulley (Eds.), Parking: Issues and policies (pp. 185–212). Emerald Publishing Limited. Parsa, H., Self, J. T., Njite, D., & King, T. (2005). Why restaurants fail. Cornell Hotel Restaurant Administration Quarterly, 46(3), 304–322. Rye, T., Hunton, K., Ison, S., & Kocak, N. (2008). The role of market research and consultation in developing parking policy. Transport Policy, 15(6), 387–394. doi:10.1016/j.tranpol.2008.12.005 Schwanen, T., & Lucas, K. (2011). Understanding auto motives. In K. Lucas, E. Blumenberg, & R. Weinberger (Eds.), Auto motives: Understanding car use behaviours (pp. 3–38). Emerald Publishing Limited. Stantec. (2011). Vancouver separated bike lane business impact study. Stantec. Retrieved April 16, 2020, from https://council.vancouver.ca/20110728/documents/penv3-BusinessImpactStudyReportDowntown SeparatedBicycleLanes-StantecReport.pdf Sustrans. (2003). Traffic restraint and retail vitality. Information note FF39. Retrieved January 31, 2013, from www.polisnetwork.eu/uploads/Modules/PublicDocuments/sustrans_ff39.pdf Van Acker, V. (2015). Defining, measuring, and using the lifestyle concept in modal choice research. Transportation Research Record, 2495(1), 74–82. Van Acker, V., Goodwin, P., & Witlox, F. (2016). Key research themes on travel behavior, lifestyle, and sustainable urban mobility. International Journal of Sustainable Transportation, 10(1), 25–32. Van Acker, V., Mokhtarian, P. L., & Witlox, F. (2011). Going soft: On how subjective variables explain modal choices for leisure travel. European Journal of Transport Infrastructure Research, 11(2). Yen, B. T. H., Burke, M., Tseng, W.-C., Ghafoor, M., Mulley, C., & Moutou, C. (2015). Do restaurant precincts need more parking? Differences in business perceptions and customer travel behaviour in Brisbane, Queensland, Australia. Paper presentation. Australasian Transport Research Forum (ATRF). Yen, B. T. H., Mulley, C., Burke, M., & Tseng, W. C. (2020). Parking and restaurant business: Different in business perceptions and customer travel behaviour in Brisbane, Queensland, Australia. Land Use Policy, 92, Article. 103818. doi:10.1016/j.landusepol.2019.01.021

418

30 INTERMODAL STRATEGIES COMBINING CYCLING AND PUBLIC TRANSPORT TO IMPROVE SERVICE AND ACCEPTABILITY Lake Sagaris Introduction: more than incidental, strategic for public transport A central challenge for public transport today is that while many need it, rely on it and receive crucial benefits from the better systems, it is not well loved by users, and surveys tend to reveal a preference for travel by train or private car. Planners, moreover, continue to focus on infrastructure and operations, behaving as if transport were “gender neutral” and thereby failing to respond to key social groups, particularly women. To these unsatisfied social needs, public health experts add demands that urban transport improve health, reducing impacts on the global climate crisis or noise, water and air pollution and favouring active transport to address obesity and related cardiovascular and other diseases (Mindell, 2018; Rydin et al., 2012). In 2020, the COVID-19 pandemic raised further challenges, with public transport increasingly stigmatized as a possible source of contagion. This has reinforced the appeal of private cars in contexts where these are affordable for the general population, although this is a situation less common in countries of the Global South, particularly amidst severe economic downturns. To date, the improvements recommended for public transport involve boosting frequencies to offset reduced capacity (due to distancing) and improvements to hygiene measures, which will be credible depending on the sociopolitical context of each place. These challenges make intermodal integration even more attractive within strategies to retain existing and capture new users. Replacing some feeder services with improved bike-bus or bike-Metro combinations can reduce overcrowding and the time passengers share breathing space in enclosed buses or Metro cars, thereby reducing potential for accumulating the viral loads associated with serious contagion. An upsurge in cycling worldwide (Buehler & Pucher, 2021) contrasts with largely ineffective efforts to position public transport, including bus rapid transit (BRT), as central to sustainable transport. Despite the early success of Curitiba, Bogotá and elsewhere, people still prefer streetcars, trains and Metros rather than buses if they can’t have their own car. Public transport alone cannot compete with the door-to-door comfort of a personal car or, increasingly, taxis or carshare. Sustainable public transport systems must overcome these limitations and thus need strategies that better integrate public transport and feeder services, particularly walking and cycling. 419

Lake Sagaris

Since the early 1990s (Replogle, 1992a, 1992b), the advantages of integrating public transport with diverse modes of cycling have received modest attention. Asian countries, particularly India and China, where cycling, cycle rickshaws and three- and four-wheelers are an integral part of both passenger and logistics transport chains, have served as significant sources of information (Cervero, 1997, 2000; Jain & Tiwari, 2011; Mohen & Tiwari, 1999). Transport innovations, particularly BRT in Latin America, have included modest efforts to integrate cycling by associating hubs with segregated cycle paths and cycle parking. In North America, bike-on-bus programs have become commonplace in both Canada and the United States: a recent internet search for bike-n-ride programs found over 238 million. From a sustainable development perspective (UN, 2015), strategies that treat “sustainable” transport as an ecology of modes, centering on intermodal walk-bike-bus/Metro combinations can increase catchment areas and resolve the first/last kilometre dilemma. They can also address social aspects of sustainability, particularly road safety and urban security, health, social inclusion, gender, age and other issues (Sagaris & Arora, 2016, 2018; Tiwari et al., 2008). This chapter reviews trends in intermodal planning for sustainable transport, considering strategies and tools that could speed transitions by improving public transport. The next section analyses the current state of intermodal transportation, defining its main components as part of an ecology of modes/users and looking at development to date. The following section considers how these measures can help public transport to respond more effectively to social justice, inclusion and other related issues. The next section presents some tools useful to implementation, particularly modal shift targets based on current origin-destination patterns, and the final section offers reflections and areas for further research.

Public transport: governance, practice and (un)sustainable transport A major weaknesses of attempts to position public transport as central to sustainability initiatives – and people’s hearts – is the lack of clear definitions and a general narrative of sustainable transport. “Sustainable” transport is often defined by a single characteristic, such as energy efficiency or emissions, rather than a clear analysis of what is essentially a sociotechnical system linked to such diverse spheres of action as politics and policy, software and social movements (see also Chapter 9). While researchers, mainly from the social sciences, articulate and analyse “unsustainable” transport, these definitions may be unfamiliar to those working from an engineering or purely technical perspective. “Unsustainable” transport is usually defined primarily as the use of the private car as the main mode for personal mobility (Banister, 2005), influencing city form, energy consumption, pollution, noise and road safety. “Automobility” (Beckmann, 2001; Sheller & Urry, 2000; Urry, 2004) expands this perspective, tracing the psychological, sociocultural, economic and ideological components of a complex bundle of values that drive the planning of car-centred cities. They also note, however, the relevance of movements that push back against automobility and inspire increasingly influential pro-walking advocacy. A variety of concerns motivate efforts by public transport agencies (Sagaris, 2006) to improve bike public transport (PT) integration. Urban and social planning considerations combine health, safety, energy and equity concerns (Table 30.1), while transit-related considerations often focus on operational improvement, particularly related to access, a growing concern, given new attention being paid to transport justice (Cook & Butz, 2019; Martens, 2017). Bike-bus integration also provides significant economic benefits (Campbell  & Wittgens, 2004), as summarized in Table 30.2. 420

Combining cycling and public transport Table 30.1 Motivations driving modal integration Urban and social planning considerations “Active” transport policies to counteract obesity-related illnesses and reduce the burden on health programs; Energy efficiency programs to reduce dependency on fossil fuels; Air quality programs to reduce the emissions associated with cars holding large modal shares; Programs to reduce emissions that contribute to global warming; Improvements to quality of life, social equity and public spaces; Responses to demands from cyclists; Concerted efforts to retain or expand transit’s modal share. Transit-related considerations Bicycling extends the catchment area for transit services and provides greater mobility to customers at the beginning and end of their transit trips. Bike-on-bus programs can attract new riders to the bus system, thereby boosting revenues. Bicycle-on-transit services give cyclists backup when it gets too dark, weather changes, illness strikes or a major highway or hill blocks daily commutes, bringing them onto transit. Bicycle and transit integration usually forms part of plans to decrease automobile traffic congestion, reduce air pollution (by reducing motor vehicle trips) and improve the public image of transit. It is particularly effective for reducing air pollution, since the worst pollution occurs during the first 11 km driven, when the motor is just warming up. It offers more commute options for workers, giving firms more flexibility on where to locate. Source: Sagaris, 2006, based on Transportation Research Board (TRB), 2005. © Laboratorio de Cambio Social Table 30.2 Economic benefits of bike-bus integration 1 Reduction in road construction, repair and maintenance costs 2 Reduction in costs due to greenhouse gas emissions 3 Reduction in health care costs due to increased physical activity and reduced respiratory and cardiac disease 4 Reduction in fuel, repair and maintenance cost to user 5 Reduction of costs due to increased road safety 6 Reduction in external costs due to traffic congestion 7 Reduction in parking subsidies 8 Reduction of costs due to air pollution 9 Reduction of costs due to water pollution 10 The positive economic impact of bicycle tourism 11 The positive economic impact of bicycle sales and manufacturing 12 Increased property values along greenways and trails 13 Increased productivity and a reduction of sick days and injuries at the workplace 14 Increased retail sales in pedestrian-friendly areas Source: Own elaboration based on Campbell and Wittgens (2004), p. 4. © Laboratorio de Cambio Social

Understanding the role of governance in transitions and transformations From an institutional or macro perspective, Geels (2012) examines governance challenges limiting sustainability transitions using a sociotechnical approach that considers diverse disciplinary perspectives, identifying three main levels relevant to transitions. These are: niches, “the 421

Lake Sagaris Table 30.3 Emerging “niches” of walk-bike-public transport integration around the world Format

Where Bogotá, Munich, Amsterdam, Santiago

2

Cycle parking (medium- and long-term at hubs, key stops and all stations) Bike racks on buses and taxis

3 4 5

Bikes on trains, usually in special cars or times Cycle rentals Public bikeshare

6

Public bikeshare with fare integration

7 9 10

Cycle routes connecting to stations Shared bike-bus lanes Cycle taxis on fixed or flexible routes

1

Canada, United States and elsewhere; taxis in Copenhagen are required to have racks for at least two bicycles Europe, Canada, United States Netherlands, most cities associated with tourism New York, Barcelona, Mexico, Washington, Montreal, etc. Montreal, Washington, New Orleans, Paris, Seville, China, elsewhere Netherlands, Denmark, Germany France, Belgium, Berlin India, New York, London

Source: Own elaboration based on Godefrooij et al., 2009; Pucher & Buehler, 2012; observations in diverse cities; presentations Velo-City conferences (2012 Vancouver, 2015 Nantes). © Laboratorio de Cambio Social

locus for radical innovations”; sociotechnical regimes, consisting of established practices, associated rules and an exogenous sociotechnical landscape. Together these form a “nested hierarchy with regimes being embedded within landscapes and niches existing inside or outside regimes” (Geels, 2012, p. 472). Using these categories, it is possible to identify emerging niches relevant to walk-bike-PT integration (Table 30.3). These concepts will be revisited to analyse intermodal strategies as part of more effective transitions to sustainability.

Enriching “behavioural change” approaches with a more complete theory of daily practice From a more micro perspective, Shove et al. (2012) emphasize the “constitutive role of things and materials in everyday life”, exploring how practices emerge, exist and die; the elements on which diverse practices depend and how practices “recruit” practitioners (p. 14). Their work explores how materials, such as the objects that constitute transport modes; the competencies necessary for individuals and groups of people to use them and meanings, the way these elements are woven together as collective symbols, ideas and aspirations, generate practices that can block or open the way to sustainability. This perspective is useful for developing actions to drive behavioural and system change through intermodal mobility strategies.

Ecologies of modes and users: a powerful way to understand “sustainable” transport The author’s own work with academics and practitioners in Europe, Chile, India and North America (ECF, 2016; Karner  & Sagaris, 2016; Sagaris  & Arora, 2016) started from defining

422

Combining cycling and public transport

social sustainability as driven by human agency through sociopolitical structures that may favour or repress collective action and social movements, simplifying or blocking transitions. The ability to exercise human agency depends on the right to act with relative freedom and security within a framework that guarantees human rights (social, cultural and economic) and therefore requires ongoing democratization for full development. Other aspects of social sustainability with regard to transport involve health effects and road safety, along with cultural values, norms and habits, as they influence everyday behaviour and must therefore change to improve greater sustainability. The next sections explore “sustainable transport” as an ecology of modes and users, drawing on these three sources to consider how intermodal, walk-bike-bus/Metro combinations require shifts in governance, daily practice and urban transitions through advocacy and other catalysts and can pressure for systemic change in all three of these components. Thus, rather than treating intermodal arrangements as accidental or incidental to general strategies to position public transport within a new, sustainable transport system, they are placed at the centre of efforts to transition toward sustainable transport.

“Public” transport as an ecology of modes and users: cycling as missing link “Public” transport has long been considered almost synonymous with bus or train passenger services. Originally, the term was associated with “ownership”, but after more than a decade of diverse public-private partnership arrangements, “public” usually refers to a system that moves “the public”, a generic term for many and diverse users, with a similarly diverse set of (dis)abilities, travel purposes and needs. Faced with unsustainable transport, or automobility, it is tempting to assume that any mode that requires less space, generates less pollution and consumes less energy per passenger is “sustainable”. In fact, however, a bus or Metro system alone does not resolve multiple problems of specific groups of users, so while these may be sustainable from an energy or emissions standpoint, they may not be sustainable from a social perspective. Moreover, train systems that encourage sprawl may prove unsustainable in specific contexts, just as electric vehicles are no panacea for ensuring cleaner travel, given that unsustainable fuel sources may rise substantially (Bahamonde-Birke, 2020). In contrast, focusing on walk-bikebus/Metro combinations can reduce energy consumption and emissions and improve social sustainability aspects of mobility. The importance of social sustainability has become particularly clear in recent years, as a plethora of studies has revealed widespread sexual harassment of women on public transport worldwide (Allen, 2016; Allen et al., 2017). Far from being “neutral”, public transport interacts with discrimination, gender and other forms of violence, social and spatial factors, worsening their impacts (Allen, 2016; Allen et al., 2017). Thinking about public transport as a combination, or ecology, of modes rather than a single standalone mode can mobilize governance and social practice more effectively in favour of sustainability transitions. (see Table 30.4). Walking is preferable for short distances (0–2 km), while cycling is best for intermediate (2–8 km) and bus-Metro in relatively dense cities for longer distances (over 5 km; Table 4, Karner & Sagaris, 2016). These distances are relative and vary according to custom, culture, infrastructure and other aspects of the built environment in each place, but they offer a guide to achieving public transport systems that can provide better service than cars, private or otherwise.

423

Lake Sagaris Table 30.4 Walking and cycling times/distances for different kinds of users m

400 800 1200 2000 3000 4000 5000 8000 10,000

Walk (minutes)

Cycle (minutes)

Easy

Moderate

Fast

Easy

Moderate

Fast

(4.5 km/h)

(5 km/h)

(5.5 km/h)

(15 km/h)

(19 km/h)

(24 km/hr)

5.3 11 16 27 40 53 67 107 133

4.8 9.6 14 24 36 48 60 96 120

4.4 8.7 13 22 33 44 55 87 109

1.6 3.2 4.8 8.0 12 16 20 32 40

1.3 2.5 3.8 6.3 9.5 13 16 25 32

1.0 2.0 3.0 5.0 7.5 10 13 20 25

Note: Dark grey = reasonable time for travel to access public transport; grey = standalone single-mode travel or cycle service trip (bike taxi, bike share, etc.); light grey = best served by combination with motorized modes Sources: From Karner  & Sagaris, 2016; moderate walking speed taken as the average in TCRP (2003, pp.  3–9) and range from (Knoblauch et  al., 1996). Cycling speeds from Gould and Karner (2009). © Laboratorio de Cambio Social

From this perspective, cycling becomes a meso or intermediate mode, a missing link that can significantly increase the time/distance ratio for human-powered travel, the healthiest for human bodies and the environment. Cycling, moreover, comes in diverse formats, which can be adjusted to specific populations and local needs and are best applied together in combinations adjusted to specific contexts (Figure 30.1). Ignoring cycling’s potential to significantly increase people’s ease of travel seriously weakens the sustainable transport ecology and significantly undermines the usefulness of public transport to most people. Personal bicycles lend themselves to an enormous variety of vehicle designs, including cargo uses (Figure 30.1) and accessories that can serve diverse users, needs and purposes. Individual bicycles come in racing, sporting and working models, particularly the step-through frame, a central adaptation to clothing worn mainly by women (skirts and dresses but also men’s Scottish kilts, Burmese paschous, Bhutan ghos and Fijian sulus). The step-through model is essential for carrying cargo and children. Increasingly, people with mobility disabilities are also turning to variants of bicycles with hand pedals or other adaptations, which allow them to participate in city life on a more equal basis (Figure 30.2). Nongovernmental organizations, such as Cycling Without Age, bring bicycle mobility services to the elderly in cities on virtually every continent worldwide (see website, Cycling Without Age, 2021). Cycling also offers diverse vehicles for personal and commercial transportation: tricycles and four-wheelers widely used in Asian and Latin American countries (Behrens et al., 2015; Cervero, 2000; Fernandes Ferreira et al., 2019); two-wheel cargo bikes more common in Europe, particularly Denmark and the Netherlands and increasingly the United States (Cox & Rzewnicki, 2015; Riggs, 2016; Schwartz, 2016). Thus, cycling comes in many forms: individual bikes but also bikeshare and cycle taxis, with growing interest in cargo bikes for personal use and within logistics chains.

424

Combining cycling and public transport

Figure 30.1a Bikes on trains as part of intermodal integration, Copenhagen 2017 Source: Laboratorio de Cambio Social

Figure 30.1b Recycler navigates a heavy cargo using her tricycle, Santiago 2017 Source: Laboratorio de Cambio Social

425

Lake Sagaris

Figure 30.1c “Crespa”, a nickname and a brand for cycling accessories sold by cargo, Santiago, 2018 Source: Laboratorio de Cambio Social

Bikeshare, which first appeared in the 1960s, took off in the late 1990s (Fishman, 2015). Usage varies but commonly follows peak travel times, particularly where bikeshare is co-ordinated with Metro stations and/or bus stops. Bikeshare users typically report convenience as their main reason. In cities like Montreal, where bikeshare is located in residential as well as commercial areas, having a bikeshare close to home is important (Fishman, 2015), whereas in cities such as Santiago, bikeshare is used mostly for travel from a public transport station to destination. Fortaleza, Brazil, overcame the challenge of locating bikeshare stations in residential areas, where they may be subject to vandalism or theft, by creating four business models, one of which allows low-income users to keep the bikes overnight, then cycle back into town or to the nearest Metro station (field visit and personal communication, 2019). Cycle taxi services have emerged, often associated with tourism (Copenhagen city centre, London’s theatre district). In developing countries, they reinforce the attraction of bus or Metro transport by filling in the first-/last-km gap. Cycle rickshaws in India have done this for over a century, providing employment while offering an essential service to women and other caregivers (Shanbaug, 2012). In Santiago, an experiment with a cycle taxi circuit in the city centre became an underground success, but was eliminated when sponsors suspended their support and neither the local government or transport authorities stepped in to maintain the system. The usefulness of these services has become particularly relevant given new research on “care-related travel” (Sánchez de Madariaga & Zucchini, 2019), done mainly by women. Figures for Santiago, Chile, for example, indicate that women make two to three times more care-related trips, which include shopping, accompanying children, older adults or people with mobility or medical needs. Indeed, these constitute the largest category of daily trips 426

Combining cycling and public transport

Figure 30.1d Bike messengers teach children their craft, Santiago 2017 Source: Laboratorio de Cambio Social

(47%), more than work (38%) or education (10%) (Herrmann-Lunecke et al., 2020; Sagaris & Tiznado-Aitken, 2020a). Sexual harassment is a significant factor inhibiting women’s use of public transport, particularly when they travel alone or after dark, a reality that affects their participation in work, education, culture and recreation. The journey from home to public transport access point and from egress to destination is particularly fraught, and even very low-income women avoid public transport, combining trips so they can afford a taxi rather than taking a bus or Metro (Sagaris et al., 2018). Having access to a bike-bus combination that eliminates a long wait at a feeder station or a lonely walk can provide vulnerable people, particularly women, with better options than relying on bus or Metro alone (Sagaris & Tiznado-Aitken, 2020b; Sagaris et al., 2017). Indeed, in the 427

Lake Sagaris

Figure 30.1e Carrying human cargo, bike taxis in Manhattan, 2015 Source: Laboratorio de Cambio Social

Figure 30.1f Bikes as cargo and bikes for cargo, increasingly key to sustainable logistics chains Source: Laboratorio de Cambio Social

428

Combining cycling and public transport

Figure 30.2a Strollers, bike seats, protected walking spaces make life easier for caregivers, Seville, 2010 Source: Laboratorio de Cambio Social

Figure 30.2b Tandems all decked out for a cycling parade integrate blind and seeing users seamlessly, VeloCity Taipei, 2016 Source: Laboratorio de Cambio Social

429

Lake Sagaris

Figure 30.2c Bike lanes provide inclusive mobility for cyclists and wheelchair users alike, Seville, 2010 Source: Laboratorio de Cambio Social

Figure 30.2d Crowded sidewalks when facilities don’t provide sufficient space for cyclists, wheelchair users and pedestrians, Seville, 2010 Source: Laboratorio de Cambio Social Emerging patterns of cycle- and other forms of mobility inclusion.

430

Combining cycling and public transport

Netherlands, one of the world’s most egalitarian societies, women make up half of those cycling, where cycling as part of a general bike-bus or (more commonly) bike-train commute accounts for a larger share of cycling than standalone cycle trips.

Practical approaches to intermodal integration Planning for an ecology of walk-bike-bus-public transport combinations opens the way to substantial health, environmental, efficiency, social equity and other benefits but is often hampered by conflict-driven debates and siloed governance arrangements that separate planning by transport mode.

Target modal shifts Innovative planning for more intermodal transport benefits from innovation in citizen participation, bringing in advocates, neighbourhood associations and other organized citizens’ groups to adequately adjust plans and systems and build a constituency of informed supporters able to communicate horizontally across diverse social groups (Sagaris, 2014, and see Chapters 9 and 36). Co-ordinating governmental, institutional and some private actors (Godefrooij et al., 2009; McClintock, 2002; Parkin, 2012; Pucher & Buehler, 2021) are also central. To address these challenges, many cities are applying target modal shifts to establish specific objectives regarding cycling (ECF, 2016). They often simply seek to increase a current share, for example 5%, to a larger share, within a specific time framework. This perspective can be rather arbitrary and seems to depend on assumptions about promotion and infrastructure. It can be more realistic to use travel data, particularly origin-destination surveys, which include information about travel purpose and trip length, to estimate how many trips by car or other motorized mode would be better served by cycling or walking (Karner & Sagaris, 2016). This can generate a relatively simple table of current and preferable modal shares to guide planning activities seeking to build new consensuses. Modal share targets, calculated for specific cities and contexts, should guide decisions rather than generating rigid rules. One such analysis considered car trips by length for the heavily car-dependent city of San Francisco (US) and the rapidly transitioning city of Santiago (Chile), finding that a remarkably similar number of trips by car (around 25%) were under 2 km, making them more suited to walking or a walk-bike service combination. Similarly, from half to over 60% of car trips were from 2–8 km, distances better served by cycling, bikeshare or other formats. In both cases, with some zonal exceptions, car travel for longer distances, more appropriate to public transport or private car use, accounted for just 20–30% of current trips (Karner & Sagaris, 2016). Estimating desirable modal shifts (Table 30.5) can help to visualize savings in space and resources that can be better invested in improving walkability, cycle inclusion and the general liveability of shared urban spaces, public and private. This approach brings out the importance of land use and improving the quality of carerelated trips by ensuring primary schools are within walking-cycling and secondary schools within cycling distance of residential areas. Similarly, the traditional pattern of local corner shops encourages walking for the most frequent shopping trips. From this perspective, the street fairs and vendors who circulate in many Latin American, Asian and African cities contribute to sustainability and, often, social equity, particularly if this leads to their treatment as a valuable, rather than an “illegal” or “informal”, part of urban landscapes. Similarly, moving more people by walking and cycling frees up space in residential and commercial areas for crucial eco-system services: shade trees, kitchen gardens, composting, parks and corridors for birds and other forms of life. 431

Lake Sagaris Table 30.5 An example of modal shift targets for San Francisco and Santiago Current share of trips by distance covered

San Francisco Car Walk Cycle Public transport Santiago Car Walk Cycle Public transport

Current mode share %

Under 2 km

2–8 km

Over 8 km

Target modal share

71.5 19.3 2.1 6

23.4 96.3 47.7 16

44.8 3.7 46.8 39.8

31.7 0.03 5.5 44.2

12.9 37.7 36.7 12.9

25.6 34.4 4 29.4

21.6 95.8 62.8 8.9

42.0 3.8 32.4 39.3

36.2 0.39 4.8 51.7

13.3 46.6 26.7 13.3

Source: Table 30.4, p. 15, Karner & Sagaris, 2017, using data from California Department of Transportation (2013) and SECTRA and Universidad Alberto Hurtado (2014). Target modal share redistributes motorized (car and public transport) trips by mode and distance; that is, very short trips are assigned to walking, intermediate trips to cycling and long trips remain motorized

Planning and governance Using cycling to connect human and public transportation may also require additional effort, depending on the experience of specific planning departments and professionals involved. University-government-advocacy collaborations can help to develop appropriate combinations of measures for specific contexts. These need to be evaluated to ensure they are genuinely meeting their objectives and typically require institutional and governance improvements (Table 30.6). Based on studies of innovation relating to transplants of ideas or strategies from one country to another, Table 30.7 indicates that it may be easiest to start with innovations in the field of informal practices (right-hand column), particularly daily operations activities (bottom row), before moving upward to formalize the most effective practices in regulations, laws and, where necessary, constitutional modifications. Transitions and transformations have much to learn from the literature on institutional transplantation. From Rose’s Lesson-Drawing in Public Policy (1993), a table of key questions can be extracted to evaluate the strengths and potential barriers to measures for transitioning toward more intermodal transport (Table 30.8). Measures to foster cycling and intermodality (Table 30.9) have been well studied and offer a broad range of possible applications, according to local context. Adjusting on-road networks to the variety of vehicles that can be mobilized for service is important and can be a challenge. Nonetheless, as this chapter indicates, the rapid advance of walkability and cycling-inclusive measures in very diverse countries and cities suggests that barriers can sometimes be overcome relatively quickly.

Specific policies and measures One crucial element in the success of cities as diverse as Bogotá, Portland or Seville is the development of a plan that combines a full network of cycling and walking amenities associated with transport hubs, preferably using strategies that include deliberative participation (see also Chapter 9) to build broad consensuses regarding the necessary behavioural shifts. A master 432

Combining cycling and public transport Table 30.6 Planning governance and institutional considerations 1

2 3 4

Organize roundtables, co-ordination meetings, public hearings, working groups which can bring together planners, designers, advocacy and other relevant actors, including innovations in internal and external participatory and co-ordination methods Apply design approaches and standards that simultaneously improve “walkability” and “cycleinclusion” in general, particularly around bus stops, Metro stations and other transport hubs Build innovative, deliberative approaches to participation into planning, education and marketing strategies Consider adjustments or deep innovations to current planning institutions to ensure that projects are designed in a harmonious collaborative way that gets the most out of each mode

Source: Own elaboration based on field experience 2006–2019. © Laboratorio de Cambio Social Table 30.7 Domains of institutional transformation

3 2 1

Level of action

Formal relations

Informal practices

Constitutional level (ground rules) Policy area (relations between government bodies) Operation level (daily activities

Legal system Formal regulations, programs, policies Procedures

Value orientation Informal codes Roles

Note: Dark grey = potential for initial interventions to achieve modal integration. Source: From Sagaris (2006), based on De Jong (1999, p. 23). © Laboratorio de Cambio Social Table 30.8 Key questions to evaluate policy transplants Factor

Relevance

Similarity of the problem Is the problem addressed by the method or tool similar to the problem for which it was designed? Institutional requirements Does the method or tool require certain institutions? Resource requirements Does the method or tool assume certain resources, such as skills, knowledge, computers, software, and monitoring systems, which may or may not be available? Complexity Is the method or tool based on simple or complex cause-and-effect relationships? The more complex the relationships, the more difficult it is to implement in an organization. Scale of change Does the use of the method or tool result in small, incremental or large-scale changes? Total quality management systems, for example, usually require large-scale changes throughout the entire organization. Interdependencies Does the method or tool assume other methods and tools on which it is dependent? Values of the managers Are the methods and tools consistent with the values of the managers? Source: From Rose, cited in Van Bueren et al.’s case study of multilateral learning for housing agencies, De Jong (1999, p. 268). © Laboratorio de Cambio Social

plan establishes a network of relevant facilities (points 1  & 6, Table  30.9), focusing on links, connections (points 2 & 3) and services (point 4), such as cycle parking, replacing the current lane-by-lane approach, which leads to many efficiency and safety deficits. It can also shed light on necessary operational innovations, which can be phased in as part of an overall program (point 5). 433

Lake Sagaris Table 30.9 Key policies and measures to foster cycling for intermodality 1

1.1 1.2 1.3 1.4 2 2.1 2.2 2.3 2.4 3 3.1 3.2 3.3 4 4.1 4.2 4.3 5 5.1 5.2 5.3 5.4 5.5 5.6 6 6.1 6.2 6.3 6.4 6.5

Extensive systems of separate cycling facilities with sufficient widths and turning ratios for cargo bikes, cycle taxis, passenger bikes, adapted bikes (people with disabilities) and others • Well-maintained, fully integrated paths, lanes and special bicycle streets connected to key transport hubs, stops and stations • Adaptation of existing lanes and roads to reduce car volumes and speeds and improve safety, mobility and interconnections for diverse cycles • Fully co-ordinated system of colour-coded directional signs for bicyclists, with transport hubs clearly signed and adequate information on forms of cycling integration • Off-street short-cuts, such as mid-block connections and passages through dead-ends for cars adequately designed to avoid conflicts with pedestrians Intersection modifications; priority traffic signals and ease of access to transport hubs, Metro stations and other relevant intermodal exchanges • Clear signage and clean access (no curbs) to facilitate access • Advance green lights for cyclists and waiting positions ahead of cars, fed by bike lanes • Bike paths brightly coloured when crossing intersections, accessing transport hubs, stops and stations • Traffic signals synchronized at cyclist speeds ensuring consecutive green lights for cyclists (green waves) and to facilitate access to hubs, stops, stations Traffic calming • Traffic calming of areas around stations, stops and hubs via context-dependent speed limit (7–30 km/hr) and physical infrastructure deterrents for cars • Bicycle streets, narrow roads where bikes have absolute priority over cars • Home zones with 7 km/hr speed limit, where cars must yield to pedestrians and cyclists using the road associated with stops, stations, hubs Cycle parking, taxis and bikeshare • Large supply of good bike parking throughout the city and adequately supervised (mediumand long-term parking) at hubs, stations, stops • Improved lighting and security of bike parking facilities often featuring guards, video surveillance and priority parking for women • Bikeshare, rental, repair, information and other services at major hubs, stations, stops Co-ordination with public transport • Fare integration (smart card) for all bike services at stops, stations and hubs • Extensive bike parking at all metro, suburban and regional train stations • Bikes on buses (external racks or reserved spaces within) and trains (specific times, cars or other facilities), at least during non-peak hours, preferably at no extra cost • Bike rental or bike share at stops, stations, hubs • Cycle taxis available by app at major Metro, bus and train stations • Deluxe bike parking garages at some train stations, with video surveillance, special lighting, music, repair services and bike rentals Road safety measures • Comprehensive cycling training courses for all school children with test by traffic police • Stringent training of motorists to respect pedestrians and cyclists • Special legal protection for children and elderly cyclists • Motorists assumed by law to be responsible for almost all crashes with cyclists • Strict enforcement of cyclist rights, including cargo, passenger, private, cycle taxi and bikeshare uses, by police and courts

Source: Own elaboration, based on field work and measures reported in table 1, p. 512, Pucher & Buehler, 2008. © Laboratorio de Cambio Social

434

Combining cycling and public transport

Making complexity work for intermodal planning As with all significant innovations in human activities, whether individual behavioural or governmental planning and policy, shifting from unsustainable to more sustainable transport systems presents significant challenges. This chapter has examined current knowledge about intermodal transport systems, focusing on walk-bike-public combinations with their respective ecologies of users, who are often neglected in the current fixation with new technologies, big data and “smart” cities. The news from the front of real-world planning is, notwithstanding, remarkably encouraging. From Geels’ (2012) perspective, significant innovation niches are emerging in many cities around the world, which are experimenting with public bikeshare, cycle master plans, walkability or modal shift goals as part of their planning environment. These are successfully introducing innovations into existing sociotechnical regimes and landscapes but often in isolated ways that do not yet realize the potential of full walk-bike-PT integration. From this perspective, advanced intermodal integration as practiced in the city of Copenhagen or throughout the Netherlands can be understood as examples of how niche innovation has successfully converged to generate system-wide change in the specific social and technical components of mobility landscapes, becoming an integral part of both urban and general policies that involve health, happiness, education and land use. These countrywide successes and the rising modal shares for cycling in many previously cardominated cities underline the importance of combining measures in the policy sphere with strategies to influence values, norms and behaviour, often through collaborations with advocacy groups, and economic components, including products such as the step-through bikes so useful to women and care-givers or services such as cycle parking appropriate to the specific use (Spapé & Godefrooij, 2009). Thus, intermodal strategies also foster combinations that can alter everyday practice, ensuring the availability of material goods, such as bikeshare, without requiring full commitment to purchasing, parking, maintaining and owning a bicycle or cargo bike. They require encouragement of the necessary competencies, particularly driving at safe speeds, which are common to fully developed environments (the Netherlands) and tend to emerge within school and other programs (for example, Safe Routes to School programs) in transitioning environments.

Final reflections Intermodality also helps to generate the meanings central to shifting everyday practice, providing a rich array of highly visible symbols – cycle ways, advance lights and preference for walking and cycling at intersections, signage indicating safe routes, cycle parking and other amenities – that intrinsically encourage new meanings, ideas and aspirations. By allowing bikes on metro cars and buses, transport agencies make visible a travel option that can better serve specific groups of users (children, carers) while encouraging other passengers to “give it a try”, thereby improving their public transport experience. During COVID-19 and post-COVID-19 planning, they also offer alternatives that can reduce overcrowding, by taking passengers travelling short distances (under 5 km) off public transport vehicles through encouraging cycle taxi, public bike share and other options. Thus, by combining often familiar objects and skills into new meanings and values, intermodal approaches to public transport show significant potential to transform daily practice, effectively mobilizing strategic factors identified by Shove et  al. (2012). This is particularly

435

Lake Sagaris

important, given that a growing body of research demonstrates that traditional social marketing and advertising campaigns, which often inform strategies to shift behaviour in the transport and other spheres, are not effective (Whitmarsh et al., 2011), at least when climate change and sustainability objectives are central. Future research will have to consider changing knowledge regarding COVID-19 and other possible public health threats associated with the exhaustion of the earth’s natural resources. It should also focus on emerging practices in cities ranging from Fortaleza (Brazil), through New Orleans (US), to London (UK), exploring both social practice and governance requirements central to innovation and then consolidating new ways of connecting these diverse, complementary transport modes.

References Allen, H. (2016). Safe and sound, international research on women’s personal safety on public transport (FIA Foundation Research Series, Paper 6). FIA Foundation, Retrieved April 18, 2020, from www.fiafoundation. org/media/224027/safe-and-sound-report.pdf Allen, H., Sagaris, L., Pereyra, L.,  & Cárdenas, G. (2017). Ella se mueve segura  – Ella se mueve segura  – A study on Women’s Personal Security and Public Transport in Three Latin American Cities (FIA Foundation Research Series, Paper 10). FIA Foundation. Retrieved April 18, 2020, from www.fiafoundation.org/ media/461162/ella-se-mueve-segura-she-moves-safely.pdf Bahamonde-Birke, F. J. (2020). Who will bell the cat? On the environmental and sustainability risks of electric vehicles. Transportation Research Part A: Policy and Practice, 133, 79–81. Banister, D. (2005). Unsustainable transport: City transport in the new century. Routledge. Beckmann, J. (2001). Automobility a social problem and theoretical concept. Environment and Planning D: Society and Space, 19, 593–607. Behrens, R., McCormick, D., & Mfinanga, D. (2015). Paratransit in African cities: Operations, regulation and reform. Routledge. Campbell, R., & Wittgens, M. (2004). The business case for active transportation, the economic benefits of walking and cycling. Better Environmentally Sound Transportation. Retrieved April 20, 2020, from https:// nacto.org/docs/usdg/business_case_for_active_transportation_campbell.pdf Cervero, R. (1997). Paratransit in America: Redefining mass transportation. Praeger. Cervero, R. (2000). Informal transport in the developing world. UN-HABITAT. Cook, N., & Butz, D. (2019). Mobilities, mobility justice and social justice. Routledge. Cox, P., & Rzewnicki, R. (2015). Cargo bikes: Distributing consumer goods. Cycling cultures. University of Chester Press. Cycling Without Age website. (2021). Cycling Without Age. Retrieved from https://cyclingwithoutage.org De Jong, W. M. (1999). Institutional transplantation: How to adopt good transport infrastructure decisionmaking ideas from other countries? Eburon. ECF. (2016). Cycling delivers on global goals. ECF & European Cyclists’ Federation. Fernandes Ferreira, A., SK Jason, C., & de Almeida d’Agosto, M. (2019). Urban multimodal sustainable transport: An environmental assessment of cargo bikes in Rio de Janeiro city. Journal of the Eastern Asia Society for Transportation Studies, 13, 1045–1061. Fishman, E. (2015). Bikeshare: A review of recent literature. Transport Reviews, 36(1), 92–113. Geels, F. W. (2012). A socio-technical analysis of low-carbon transitions: Introducing the multi-level perspective into transport studies. Journal of Transport Geography, 24, 471–482. Godefrooij, T., Pardo, C., & Sagaris, L. (2009). Cycling-inclusive policy development: A handbook. In Interface for cycling expertise, GTZ, Federal Ministry for Economic Cooperation and Development. Berlin: Interface for Cycling Expertise: Utrecht and GTZ. Gould, G., & Karner, A. (2009). Modeling bicycle facility operation: Cellular automaton approach. Transportation Research Record, 2140(1), 157–164. doi:10.3141/2140-17 Herrmann-Lunecke, M. G., Mora, R., & Sagaris, L. (2020). Persistence of walking in Chile: Lessons for urban sustainability. Transport Reviews, 40(2), 135–159. Jain, D., & Tiwari, G. (2011). Impact of strategies changing the infrastructure for NMV and buses on accessibility of urban residents. Urban Transport Research Journal, Institute of Urban Transport, 8–14.

436

Combining cycling and public transport Karner, A., & Sagaris, L. (2016). Testing a new approach to planning sustainable transport using data from metropolitan Santiago de Chile and the San Francisco Bay Area. Transportation Research Board. Karner, A., & Sagaris, L. (2017). Testing a new approach to planning sustainable transport using data from Metropolitan Santiago de Chile and the San Francisco Bay Area. Paper presented at the Transportation Research Board, Washington, DC. Knoblauch, R., Pietrucha, M.,  & Nitzburg, M. (1996). Field studies of pedestrian walking speed and start-up time. Transportation Research Record: Journal of the Transportation Research Board, 1538, 27–38. doi:10.3141/1538-04 Martens, K. (2017). Transport justice. Routledge. McClintock, H. (2002). Planning for cycling principles, practice and solutions for urban planners. Woodhead Publishing Limited, CRC Press. Mindell, J. (2018). Editorial transport and health expertise can help meet the world health organization’s goals for global climate change and noncommunicable disease prevention. Transport  & Health, 11, A3–A4. Mohen, D., & Tiwari, G. (1999). Sustainable transport systems: Linkages between environmental issues, public transport, non-motorised transport and safety. Economic & Political Weekly, 34(25). Parkin, J. (2012). Cycling and sustainability. Emerald Publishing Limited. Pucher, J., & Buehler, R. (2008). Making cycling irresistible: Lessons from The Netherlands, Denmark and Germany. Transport Reviews, 28(4), 495–528. Buehler, R. & Pucher, J. (2021). The future of city cycling (2nd ed.). MIT Press. Replogle, M. (1992a). Bicycle access to public transportation: Learning from abroad. ITE JOURNAL, 62, 15–21 Replogle, M. (1992b). Bicycles and cycle-rickshaws in Asian cities: Issues and strategies. Transportation Research Record, 1372, 76–84. Riggs, W. (2016). Cargo bikes as a growth area for bicycle vs. auto trips: Exploring the potential for mode substitution behavior. Transportation Research F: Traffic Psychology and Behaviour, 43, 48–55. Rose, R. (1993). Lesson-drawing in public policy: A guide to learning across time and space. Chatham House. Rydin, Y., Bleahu, A., Davies, M., Dávila, J. D., Friel, S., De Grandis, G., . . . Wilson, J. (2012). Shaping cities for health: Complexity and the planning of urban environments in the 21st century. The Lancet, 379, 2079–2108. Sagaris, L. (2006). Integrating bicycle travel into transport networks in Santiago, Chile [Unpublished master’s thesis, University of Toronto]. Sagaris, L. (2014). Citizen participation for sustainable transport: The case of “living city in Santiago, Chile (1997–2012). Journal of Transport Geography, 41, 74–83. Sagaris, L., & Arora, A. (2016). Evaluating how cycle-bus integration could contribute to “sustainable” transport. Research in Transportation Economics, 59, 218–227. Sagaris, L., & Arora, A. (2018). Cycling for social justice in democratizing contexts: Rethinking “sustainable” mobilities. In T. Priya Uteng & K. Lucas (Eds.), Urban mobilities in the global South (pp. 19–40). Routledge. Sagaris, L., & Tiznado-Aitken, I. (2020a). Sustainable transport and gender equity: Insights from Santiago Chile. In D. Oviedo Hernandez, N. Villamizar-Duarte, & A. M. Ardila Pinto (Eds.), Urban mobility and social equity in Latin America: Evidence, concepts, methods, volume 12. Emerald Publishing Limited. Sagaris, L., & Tiznado-Aitken, I. (2020b). Walking and gender equity: Insights from Santiago Chile. In A. Pinto  & D. Oviedo Hernandez (Eds.), Urban mobility and social equity in Latin America. Emerald Publishing Limited. Sagaris, L., Tiznado-Aitken, I., & Rivera, M. (2018). Ella se mueve segura – Santiago Case Study. Ella se mueve segura, Mujeres Seguridad & Transporte Público (Unpublished Report). FIA Foundation. Sagaris, L., Tiznado-Aitken, I.,  & Steiniger, S. (2017). Exploring the social and spatial potential of an intermodal approach to transport planning. International Journal of Sustainable Transportation, 11(10), 721–736. Sánchez de Madariaga, I., & Zucchini, E. (2019). Measuring mobilities of care, a challenge for transport agendas. Integrating gender into transport planning, from one to many tracks (C. Lindkvist Scholten and T. Joelsson, Eds., pp. 145–173). Palgrave Macmillan. Schwartz, J. (2016). The impact of cargo bikes on the travel patterns of women [Unpublished master’s thesis, California Polytechnic State University]. Shanbaug, A. (2012). ‘Rickshaw Bank’ concept changes lives of thousands of pullers. The Economic Times.

437

Lake Sagaris Sheller, M., & Urry, J. (2000). The city and the car. International Journal of Urban and Regional Research, 24(4). Shove, E., Pantzar, M., & Watson, M. (2012). The dynamics of social practice. Sage. Spapé, I., & Godefrooij, T. (2009). Bicycle parking: Tools for success. In T. Godefrooij, P. Carlosfelipe, & L. Sagaris (Eds.), Cycling-inclusive policy development: A handbook. Eschborne. Interface for cycling expertise, 2009 (pp. 110–123). Deutsche Gesellschaft für Technische Zusammenarbeit. TCRP. (2003, January). TCRP REPORT 88: A guidebook for developing a transit performance measurement system. Transit Cooperative Research Program. Tiwari, G., Arora, A., Jain, H., & Godefrooij, T. (2008). Bicycling in Asia. Interface for Cycling Expertise. UN. (2015). Sustainable development goals. Retrieved August  25, 2020, from www.undp.org/content/ undp/en/home/sustainable-development-goals.html Urry, J. (2004). The system of automobility. Theory, Culture & Society, 21(4–5), 25–39. Whitmarsh, L., O’Neill, S. J., & Lorenzoni, I. (2011). Engaging the public with climate change, behaviour change and communication. Earthscan, Taylor & Francis.

438

31 ACCESSIBILITY AND DESIGN FOR ALL Nick Tyler

Introduction The starting point for any account of accessibility and design for all has to be that we are talking about people. Places are people. Whether the ‘place’ being discussed is a city, town, village or remote rural area, it is the people who give it life and who, in return, receive the protection and provision of opportunity that the ‘place’ provides as its basic function. If the ‘place’ is not accessible to everyone, it fails to honour that basic function. Homo sapiens, as a species, survived the change in climate 40,000  years ago alone of the hominids, largely because, as the food-rich jungles receded and left just a relatively barren savanna, people were able to collaborate to search for and capture food. They had the vestiges of language and communication and the propensity to collaborate, rather than compete, with each other in order to capture the rarer but often much larger food to satisfy the energy needs of their large brain. Over time, the other hominid species were less good at this, and they died out. Humans are the progeny of these collaborative social beings: we – Homo sapiens – are essentially a social species. Urban living is a very recent process. Modern cities as we know them only started with the Industrial Revolution in the eighteenth century, with Manchester often cited as the first ‘modern city’. The characteristic of this city was not that it was based on a source of water or a communication path, or that it provided a defensive protection against attack but because energy, first in the form of water and later as steam, could be centralised to provide the large power supply for a centralised factory. Such a factory required workers, workers required living accommodation, and so the ‘modern’ city was born. Genetically, we are indistinct from those long-ago inhabitants of the savanna, but our style of living accommodation is only around 250 years old. The increasing size and scale of cities have resulted in the need to develop mobility assistance – and public transport is a good example of the ingenuity of Homo sapiens in creating a means of locomotion that enables cities to be large and yet still enables the people to live well within them. As with any system created for the general use, public transport is not universally capable of providing movement to everyone: some people are unable to use it for a variety of reasons, and this chapter explores how ingenuity can be brought into the task enabling public transport to ensure that the fundamental advantages of a city are genuinely available to everyone. That is not yet the case in any city in the world. Indeed, it is true to say that without an 439

Nick Tyler

accessible public transport system, a city is inequitable, unjust, exclusive and basically failing as a place for people (see also Chapter 26). Tyler (2006) sets out a basic approach to conceiving cities in a way that puts the people at the centre of decision-making, from visionary ideas through policy-making to systems design – including the public transport system – for all. This chapter is not simply about how to make public transport accessible, it is about how to make cities fit for everyone and the role that public transport plays in order to bring that about. Accessibility to all is a primary requirement of a just city, and as the public transport system is how we offset the spatial and temporal displacements between the people and the activities they wish to pursue, it is fundamental that accessibility to public transport be ensured for everyone. Inaccessible public transport is an oxymoron: it cannot be ‘public’ if it is not accessible to the public. It is not possible to ensure that everything is available in the same place, so it is necessary to distribute these characteristics around the city. Therefore, in order to provide equity, a system is required in order to enable people to be able to travel to access cities. If that means of travel is not accessible – for economic, physical or other reasons – the city cannot provide equity to everyone, and it fails. Public transport – in the sense of the provision of a means of travel for everyone – is the equitable answer to this challenge but only if it is actually accessible to all. However, people do not just use public transport for its own sake: they need it in order to be able to activate their life’s necessities.

Capabilities People ‘do’ activities. These could be activities such as eating or sleeping but also working, playing, learning, enjoying and many more. They are fundamental to survival as well as to the wellbeing of society. These activities happen in places, so to ‘do’ an activity, and thus contribute to survival and wellbeing, it is necessary to reach the place where it will take place. What would prevent that from happening? The answer depends on capabilities. Sen (1992) introduced the idea of capabilities and functionings as issues of human rights. In dealing with accessibility, they are also a matter of human rights. Put simply, in this context, functionings are things that one can have a right to be able to do, and capabilities are the subset of those things that one is capable of achieving. The gap between them is an example of the inability of society to overcome the restrictions it creates that reduce functionality. In the context of accessibility, this can be considered an issue of a difference between the capabilities a person has and the capabilities that the environment requires in order for them to ‘do’ their chosen activity. This was conceptualised in the ‘capability model’ (Tyler, 2006): Accessibility = Provided Capabilities – Required Capabilities which shows that accessibility is positive if the person’s capabilities are greater than those required by the environment. The importance of the capability Model is that it frees up consideration of accessibility from particular ‘disabilities’. This places a different emphasis from that of the medical model of disability, in which the emphasis was placed on the condition which yielded the disability, and the social model of disability (Oliver, 1996), in which the emphasis was placed on the social inability to create an accessible environment as the cause of the disability. The capability model places the emphasis on what the person can do, however they manage to do it – including the use of any assistive device – and contrasts this against what the environment requires, however it imposes that requirement. This openness is very important when it is necessary to take into 440

Accessibility and design for all

account the combination of capabilities that might be involved in a particular interaction with the environment, for example, the combination of physical, sensorial and cognitive capabilities that can be brought to bear to enable someone to overcome an obstacle. It also places a much heavier burden on the environment to be able to respond to the full range of, and differences between, people’s capabilities. Thus a person walking along a footway and wanting to cross a road to reach a bus stop needs to be able to maintain their balance, even when turning their head and upper body to look for traffic, requiring their cognitive system to know from which direction the vehicles will be approaching, vision and hearing systems to be adequate to see and hear traffic at that place, the processing ability to estimate when the vehicles might appear at the crossing point . . . and so on. Failure of any of these would render impossible the activity – not just of crossing the road, and not even of catching the bus, but the desired activity at the end of the journey. This capability test is continuous and is exercised whenever an interaction with the environment takes place. Figure 31.1 shows this in a simple form and represents the test that needs to be applied when considering every single aspect of the process of accessing public transport (or anything else). In Figure 31.1, the objective is to meet the need (Step 0). In order to achieve this, it is necessary to achieve physical access – it is necessary to reach the place where the need can be met (Step 4). The need could be anything – it could be to buy a loaf of bread, or it could be to find information about whether it is possible to cross the road – and in every case, it will be necessary to obtain physical access of some sort or other (even if the interaction is virtual, the person will still need to be in a suitable place in order to achieve it). In order to achieve physical access, it is necessary for it to be physically possible. For example, if the need is to obtain information from a sign, it needs to be the case that the sign is findable, visible, readable and intelligible. Each of these requirements needs to be addressed and tested against the person’s capabilities before the sign can be considered accessible. So, Step 1 is to know the provided capabilities. This will suggest a possible route, which will require knowledge about the required capabilities in the environment (e.g. the state of lighting, positioning of the sign) (Step 2). Step 3 then compares

Figure 31.1 The process of determining if a given interaction with the environment is accessible Source: Author

441

Nick Tyler

the required capabilities against the person’s provided capabilities: if the provided capabilities are greater than the required capabilities (“Yes”), then we can proceed to Step 4. If the required capabilities are greater than the provided capabilities (“No”), it is necessary to return to Step 2, and another route needs to be considered. This procedure applies to every stage in a person’s journey – the procedure for making the journey, or for determining what needs to be done for it to be accessible to everyone, is exactly the same in both cases. The only difference is that in the latter case, every eventuality needs to be considered for individual people across the population as a whole rather than just for a single person. There is a strong perception that capability reduces with age. The World Health Organization, for example, shows a figure of capabilities increasing through younger ages, stabilising in middle age and declining in older age. This is a common but unhelpful view. However, it is a very ageist view. First, it is debateable whether capabilities should be set against calendar age – people who are younger may well have reduced capabilities in some regards, and older people might have greater ones. Secondly, even though capabilities undoubtedly change during a lifetime, and some activities become more difficult as a result, not all become worse. For example, ‘experience’ increases, and older people are adept at finding ways to do things differently if necessary in order to maintain their independence. The capability might be retained, although the method might change. That increase in, and utilisation of, experience could make a big difference to the capabilities of individuals, and transport designers should be looking to make the most of this when considering how people make the journeys they wish to make. This means making options open: there is more than one way to board a bus, more than one way to know it is the right bus, and it is essential that public transport designers ensure that these possibilities are available as a routine part of system design – in infrastructure, vehicles and operations. It is therefore essential to ensure that quality of life is maintained and enhanced for people and that they are able to enjoy living in their local community, interacting with local people. The sense of friendliness extending to unknown others is an important aspect and a major element of quality of life, in particular for people who find it difficult to move around as a result of differences between their own capabilities and those required of the environment. This can be referred to as ‘sociality’.

Sociality Sociality is the propensity for a person to interact freely with unknown others. Joffe and Smith (2016) originated the awareness of this when they found that “the desire for a strong sense of community was expressed by approximately two thirds of dwellers. . . . For many, this was associated with friendly neighbours fostering a close-knit and interconnected environment”. This is marked by the brief inconsequential greetings between individuals as they pass each other in the street and is important when considering accessibility because it refers to the importance of connectedness between people. If the city is inaccessible, that connectedness cannot happen, and as a result, people become isolated. For those with problems in relation to accessibility, this is a really difficult problem. In the 1960s, Hall (1966) studied people in New York to see how they interacted with each other. He coined the term ‘proxemics’ – the study of proximity – to allow him to consider different spacing between people under different social conditions. Hall’s observations came up with a set of ranges of distance which set the social interactions of people in the city (see Table 31.1). Hall’s ‘distances’ delineate sociality. It is possible to wave and greet a person (known or unknown) at around 8 m (this is, for example, the distance between the midpoint of a footway 442

Accessibility and design for all Table 31.1 Distances in urban space (summarised by the author from Hall, 1966) Distance

Description

>8 m 3–8 m 1.2–3 m 0.4–1.2 m 0–0.4 m

Far Public distance Social distance Personal distance Intimate distance

on one side of a UK residential street to the equivalent point on the other side), but to have a more complete conversation, it would be necessary to be on the same footway. The essential nature of sociality has been demonstrated clearly in the 2020 COVID-19 event – people have been required to distance themselves from others, such that the minimum distance is supposed to be 2 m – too far for conversations at a personal distance. However, people have demonstrated a need for sociality and have found ways of enabling it at a greater distance – typically through communal singing, clapping and so on. Physical distance can be changed, but sociality is rooted in the social need to interact with people, and that has to happen somehow, even if the permitted physical distance makes it more difficult. Mahoffey et al. (2020) suggest that a neighbourhood needs to be considered an area within which every residence is within 200 m of a public space and emphasises the need for infrastructure such as footways to be wide enough (3 m) to allow such interactions to occur without causing problems to others. Of course, many of the community’s needs cannot be provided within such a small space, and it is necessary to link these neighbourhoods with a transport system. This is where public transport comes into play. Transport for London, for example, aims to have public transport accessible to every resident within a 5-minute walk (about 400 m). However, sociality is a human construct and remains with the person wherever they are, and the way in which the public realm is designed needs to ensure that sociality is always facilitated. This tends to prioritise pedestrian space over vehicular space, as it is much easier for a person to meet, greet and converse with someone else when both are in a pedestrian space than if, for example, both are in a car. Thus, the public transport system, if it is to serve the public, needs to deliver sociality, and it needs to deliver it to everyone. The question is how to deliver sociality in a way that is accessible to all.

Accessibility and urban design The first consideration for urban design is the relationship between the present and the future: “How will I be able to do my chosen activity given the present state of the urban environment?” This is Step 0 in Figure 31.1 and is based on experience from the past (“Could I do it last time I tried?”), mitigated by a reconsideration of the person’s provided capabilities (Step 1). That reconsideration is affected by the requirements of the environment, as it is not the provided capabilities that determine accessibility but the relationship between these and the required capabilities. The public transport designer, therefore, has to ensure that the requirements are minimised. Table 31.2 captures these questions in summary form. First, it is necessary to remember that what is being looked for is to enable people to come together to create a better society through sociality. Given that desire, it is important to understand the present situation – how things are now, at the moment of wanting to make the journey. The task for the designer is to anticipate 443

Nick Tyler Table 31.2 High-level requirements for accessible design Issue

Present

Future

Aims of the design process

Sociality

Propensity for interaction with unknown others Based on composition of memory (past experience), present sensorial information, intersubjective interactions Present situation “Where I am” “Can I move?” “Can I board the bus?”

Wellbeing and future of society Anticipation based on past presents

To increase sociality

Understanding of the present

Physicality

Capability

Required: “What I need to be able to do? Provided: “What I can do?”

“Can I go further?” depends on information about future ‘presents’ (“Will I be able to board the bus?”) Information about future required capabilities

To make a clear and true indication of the present situation

To ensure physicality presents no barriers

To ensure that required capabilities < provided capabilities

these questions for all people in the community, whatever their capabilities. Then it is necessary to think about the physical situation of where the person is now, what their capabilities are and how this combination affects the future opportunity to achieve their chosen activity. The aim of the design process is to ensure that this physicality does not impinge on the person’s ability to carry out their chosen activity. The fourth consideration is all the required capabilities represented by the physicality of both the present and future situation that the person will encounter as they seek to carry out their activity. It is essential for the person to know what they are likely to face in the course of their journey so that they can prepare in advance and not be shocked by an unexpected impossibility in the course of their journey. This means that required capabilities need to be signified. This can be by a sign (e.g. a warning of a narrow gap) or through other information modes, such as maps, which could give such information in advance. Finding out in real time is not an option in an accessible system. The designer needs to have foreseen all such possibilities and enabled people to find out about them in advance. The last issue is often forgotten yet is essential. This is how someone knows what might or might not be possible. It enables them to look for alternative ways of achieving their activity – it also highlights to the designer where problems need to be addressed. To understand how to conceive this in the public transport system, the example of a bus stop can serve as an illustration. Breaking down all the activities required at a bus stop illustrates the level of detail necessary to consider the requirements being placed on the passenger in order to be able to use the bus system. This is illustrated in Table 31.3, where the concept of transfer between being a pedestrian and a passenger via a ‘pedenger’ (defined as the state of adjusting from the needs of one to the needs of the other) is used to highlight the different needs of each phase of the process (Tyler, 2015). Table  31.4 explores this information provision in a little more detail. This segments the information provision into three issues: what form the information takes, when the information is needed in order to ensure that the person can act on it in time and what the information means. 444

Accessibility and design for all Table 31.3 Different passenger functions at a bus stop Phase transition

Stage

Pedestrian to pedenger

1 2 3 4 5 6 7 8 9 10 11 12 13

Pedenger to passenger

Passenger to pedenger

Pedenger to pedestrian

14 15 16 17

Confirm correct bus stop using information at the bus stop Wait Rearrange bags Find money or validation ticket Identify the correct bus Locate the correct door Step onto the bus Initiate payment/validation Move inside the vehicle (free up space for the next pedenger) Locate the correct bus stop Gather bags/belongings, etc. Move to the correct door Initiate payment/validation (this is rare but is the case in some countries, e.g. Japan) Step off the bus Rearrange bags Orient themselves to the new environment, using information at the bus stop Move away from the bus stop (free up space for the next pedenger)

Source: Summarised from Tyler, 2015 Table 31.4 Factors to take into account when thinking about information provision for accessible design Issue

Questions

How delivered

What

Is there something there? What is it? What does it say? Is it timed/placed so that I can act on it in time?

Contrast, differentiation foreground/background

When Meaning

Can I understand it? Can I use it?

Timing/location How does it attract attention? Memory/learning, processing cognition

‘What’ form the information takes thus relates not only to whether it is a map or a signpost but also how the person knows it is there, what to expect, and what kind of information it is trying to convey. Information that cannot be found is useless. The designer should ask sharp questions: How would a blind person know where the Braille map is located? Can all blind people read Braille (information delivered in Braille is an example of a required capability)? How does a person find out what information is available, and where can it be found? This requires a consistent approach to the provision of information so that expectations can be met  – no information suggests that there is nothing to inform people about and so could give an incorrect impression that there are no required capabilities to be concerned about. ‘When’ is how the information needed relates to the time – and therefore the location – that the information is required and depends on how much time is needed for the person to act on it. Sometimes this is several hours or even days, if a lot of preparation would be required, but often it is a matter of warning people of an upcoming obstacle in their path. Cheng (2014) 445

Nick Tyler

found that older people preconsciously adjusted their gait when they first perceived a change in footway level about 3 m before they reached it. This suggests that information about such a change in footway surface should be indicated around 3 m in advance – and that there is little to be gained by placing that information further or nearer to the change. How the information attracts attention is a really important issue. Normally, it could be expected that this could be done through vision – a warning sign of some sort, but that requires the capability to see it – and also to know where to look for it. Human evolution for the savanna does not help here. The savanna has little in it, and vision is evolved well for determining potentially dangerous objects on a relatively empty horizon. However, it is not so good when the horizon is more complex: a person can only actually ‘see’ a small image about three times per second – the rest of what is seen is essentially made up from combination of memory, past experience and guesswork (this is how magicians manage to fool audiences and why eyewitnesses of events are so poor). So, it is necessary to ensure that the sign (whatever it is) is detectable by the person’s preconscious brain in order to realise that it is there and pay conscious attention to the information and implications it contains. For this reason, it is better to try to remove such obstacles if at all possible and to ensure that the person is warned about them beforehand if that is not possible. Information can be delivered in other ways, of course – sound or touch are both useful ways of conveying information. ‘Meaning’ of the information means that it will be necessary to be sure that the person understands what it means. Although sounding tautologous, a sign that conveys no meaning is not informative and so is useless. Images, icons and language can all work well, but they require the capabilities needed to understand them. Information needs to stand out, so it needs to contrast with the background – whether it is visual, auditory or haptic. Table 31.5 summarises some sensorial issues arising from the need to convey information. If the lighting level is too low, it will be difficult to distinguish visual cues, but if the lighting is too bright, it can be equally difficult to see them, because the principal way of delivering information is through contrast, including shadows: bright lighting tends to wash out shadows,

Table 31.5 Sensorial issues for accessible design Too little

Too much

Small

Large

Lighting

Difficult to see obstacles, information

Glare makes it difficult to see obstacles/ information Shadows disappear

Difficult to detect/see information

Sound

Difficult to hear information, approaching hazards

Difficult to Difficult to continue to do detect/hear tasks/speak/ information hear

Lose connectivity Difficult to as words/ make out images too information difficult to see against the in one go background Difficulty if fixations are not distributed enough Can be Difficult to distressing distinguish if adding information to already from loud noises background (including noise tinnitus)

446

Contrast

Accessibility and design for all

Too little

Too much

Small

Large

Difficult to Difficult to know how know where to navigate you are in around it the overall context Becomes an Touch Non-informative Non-informative Difficult to detect obstacle Difficult to through feet/ distinguish as fingers an identifiable object Difficult to see/ Could be Too difficult Information Too difficult hear confusing to tell what to know about what it is useful what it is refers to information saying, orient, and what is determine data/noise location, time, etc. Space

Crowded, cannot Can become manoeuvre, disoriented/ claustrophobic lost, agoraphobic

Contrast Important to distinguish between different parts of the space Difficult to make out against other haptic surfaces Difficult to tell what is information and what is data/noise

so obstacles such as steps cannot be seen. A common misperception is that ‘bigger is better’, but the designer needs to remember that the key is contrast, not size. It can be difficult to visualise a whole word if it is presented in a way that makes it difficult to see as a whole in one go. The high-resolution acute colour-sensitive part of the eye only sees a range of about 2 degrees, and to see more, the eye needs to move, thus making it harder to contain the whole information provision in one glance. Some people can only see a limited range within a scene in any case. The equivalent arises with sound or haptics – loud or rough only works as an information source if it contrasts with a background. A  voice within an environment of other voices is hardly distinctive. On the other hand, people seek cues from what there is, so the combination of vision, sound and touch can deliver information in a way that any of them in isolation would fail to do. It also opens up the possibility for people with different capabilities to be able to access the information, at least in part, so it is important to think in terms of combinations of different modes of delivery and not to think that only one would be enough. For example, a crucial issue for many people is how to know where to alight from a bus or train, where to board it and whether the vehicle in front of them is the correct one. This can be indicated by visible sign (on or off the vehicle as required) and by audible voice announcements, but it could also be indicated by simple signs, musical jingles and other methods to widen the availability of the information – and thus the bus system – to many more people. When moving around the environment, it is necessary to recognise that this is often the repository of accessibility problems – people are trying to reach and do their chosen activity and, as noted before, this usually requires them to move. So, locomotion is often the indicator of a failure to be accessible, but it is not by any means the sole root cause of inaccessibility. When considering locomotion, there is a really important distinction that has to be made: Homo sapiens evolved feet for bipedal locomotion (Table 31.5). All the human body systems fit with this model. However, if for whatever reason feet are replaced with wheels to create motion, the whole situation changes. Feet and wheels need two very different sets of environments, and they are often in conflict with each other. Perhaps the best example of this is a change in vertical height. Feet generally prefer the walking surface to be horizontal 447

Nick Tyler

because it suits the way in which they create propulsion. When standing on two feet, the body weight passes through the feet in an evenly distributed way. Basically, walking requires the body weight to be put a bit out of equilibrium by allowing the body weight to align itself in front of the feet; the response is to move one foot forward in an attempt to regain equilibrium, thus pushing forward and creating motion. This is repeated with the other foot. It makes sense, therefore, that the foot, with a complex skeletal and muscular system, maintains this capability best when the surface on which it is positioned is flat and horizontal – quite a lot of variation is possible as a result of the metatarsals and ankles, but there is no doubt that flat is best. So, to change vertical height, although slopes are possible up to a point, steps are the best way for feet to achieve this. A lot of older people find slopes difficult because of the pressure placed on the ankle and knees. On the other hand, steps are simply impossible for wheels, unless the wheel is large enough that the vertical height change is significantly less than their radius. Wheels are also poor at crossing gaps, something that feet, in combination with legs, can manage, up to a point. So, feet can cope with a small degree of roughness and unevenness, but wheels need a continuous surface with no steps. The main lesson from this is that ramps cannot replace steps: to provide an accessible means of changing height, it is necessary to have both steps and ramps. A person changing vertical height has to overcome gravity, whether they are on foot or in a wheelchair. What is often a surprise is just how much effort this takes. To be accessible, slopes should be as short as possible, and there are limits to the gradient that is actually feasible. Slopes should have stretches of horizontal surface to allow for rest. The same is true for steps. Both should be equipped with handrails on both sides so that these can support and provide assistance for propulsion if necessary. It is also essential to remember that what goes up must come down – downward motion on either steps or slopes is often more challenging than upward motion. In the case of walking, this is because shifting weight forwards on a downward trajectory means putting the body even further out of balance, and the fact that the ‘next step’ means putting the foot not only further forward but also further down increases the disequilibrium and so needs to be resisted. In the case of wheels, the effect is to increase speed, which then needs to be resisted in order to remain in control. Those horizontal rest zones are arguably more important for the downward direction than they are for the upward one. Table 31.6 contains a summary of matters to take into consideration when thinking about locomotion.

Table 31.6 Comparison of locomotion using feet or wheels Locomotion – feet

Locomotion – wheels

Vertical – upwards Steps identifiable from ~3 m, handrails both sides, starting before the stairs and ending afterwards, consistent vertical risers, non-slip.

Steps impossible; ramps