A Study of a Community Franchise Business Model of Public Transport Service Delivery
A Dissertation Submitted to Fulfil Requirements for the Degree of Doctor of Philosophy
The Institute of Transport and Logistic Studies The Business School University of Sydney Sydney, Australia
Candidate: David Emerson MPhil (UNSW) BArch (SydU) FAIA Student Number: 196505953 Supervisors: Professor Corinne Mulley Professor Michiel Bliemer
July 2020
ii
Originality Statement “This is to certify that, to the best of my knowledge, the content of this thesis is my own work. This thesis has not been submitted for any degree at Sydney University or any other educational institution or other purposes.
I certify that the intellectual content of this thesis is the product of my own work and that all the assistance received in preparing this thesis and sources have been acknowledged, except to the extent that assistance from others in the project’s design and conception or in style and linguistic expression is acknowledged.
Signed…………………………………………………………….
Date…………………………………………………………………
iii
iv
Abstract
An unusual and unreported business model of public transport, known locally as a Community Franchise, was identified operating in a large Italian ski field. It is investigated by case study and compared to other more conventional ways of providing ski lifting services. The similarities between ski lifting and public transport are used to construct a hypothetical application of the regime to an urban setting. The study establishes the ways that the regime functions and how it differs from more conventional models of urban public transport in ownership, funding, and incentives. There is no explanation in the literature for the way that this regime undertakes network planning or the effects of franchisees owning and operating the individual lines of transport. A definition of design, broader than the dictionary definition, is suggested to bridge between the normal and the observed resource allocation processes. A computer game is used to experimentally investigate aspects of the ways a centrally directed network functions compared to one where individual lines of the network are owned by competing, profit-orientated franchisees. This experiment establishes that there is no significant difference to a measure of social welfare between the centrally planned version and one where competing franchisee providers maximize their profit. The principal study objective of evaluating the possible application of the regime to urban public transport is approached by canvassing a group of experts who provided anonymous responses to questions about its hypothetical application to the city of Sydney. There is little consensus in the expert view of the urban application of this regime, but an overall indication of its possible application to this role.
v
vi
Author Attribution Statement
I am a supervisor of the candidate, David Neale Emerson, of the thesis entitled: “A Study of a Community Franchise Business Model of Public Transport Service Delivery”. This dissertation is being submitted to fulfil requirements for the degree of Doctor of Philosophy, through the Institute of Transport and Logistic Studies at the University of Sydney. Some material in the Literature Review is supported by citation of (Emerson et al., 2016) EMERSON, D., MULLEY, C. & BLIEMER, M. C. J. 2016. A theoretical analysis of business models for urban public transport systems, with comparative reference to a Community Franchise involving Individual Line Ownership. Research in Transportation Economics, 59, 368-378. I attest that this is a true statement of the authorship of this paper. Signed
Professor C Mulley
vii
ABBREVIATIONS AND ACRONYMS ARCC Alpine Resorts Co-ordinating Council, Victoria Australia ARMB Alpine Resort Management Boards, Victoria Australia BusNSW The peak industry organisation for the bus and coach sector, NSW Australia CS Consumer Surplus MNL Multinomial Logit Model PS Producer Surplus ROC Regional Organisation of Councils, NSW Australia ROI Return on Investment TfNSW Transport for NSW: The lead agency of the NSW Transport cluster, Australia Thredbo Thredbo Conference Series on Ownership and Competition in Land Passenger Transport
viii
ACKNOWLEDGEMENTS I acknowledge the opportunity presented to me by the Institute of Transport and Logistic Studies in accepting me as a doctoral candidate. In hosting this somewhat unusual study, it has provided a supportive and stimulating environment that has greatly assisted in the development of the ideas in this thesis. I am deeply indebted to my supervisors, Professors Corinne Mulley and Michiel Bliemer; for their guidance throughout my study period. Professor Mulley was most supportive and suitably strict in her review of my techniques and theoretical understandings. She was the midwife of this study and provided constant and valued reality checks. Professor Bliemer showed extraordinary patience and provided invaluable input with his authorship of parts of the economic experiment. He provided wise counsel in many aspects of game functioning, experimental procedure, and interpretation of the results. Thanks, are especially due to those academics, transport practitioners, and friends who acted as a sounding board when I was establishing the problem and exploring the nature of the study. This includes Prof Stephen Glaister, Prof John Preston, Prof Andrew Terry, Prof Winjand Veeneman, Didier Van de Velde, Prof Peter White, Charles Curnock, Sir Peter Hendy and Mr Peter Bruliseuer. Special thanks are due to Dott Gerhard Vanzi of Dolomiti SuperSki who provided the essential information and understandings of the franchisor company that he was intimately associated with for decades. Also, to Mr Mario Longhitano who arranged access to key figures in Italy. As well, the other ski field management personnel, Dr Peter Tschiderer, Messes Paul Anderson and Ian Swan, who were instrumental in providing for me the perspective of the functioning of the other ski field organisations with which the Italian regime was compared. The anonymous examiners also gave crucial and helpful advice, with pivotal suggestions which led to conducting the expert survey. I also acknowledge and thank my colleagues, friends and associates of all sorts, who assisted in the conduct of and participated in the computer experiment. This was an undertaking of some complexity and time but well worth the effort as the results were of substantial importance in the understanding of the Community Franchise phenomena. Special thanks are due here to my thorough, dedicated and long-suffering wife, Margaret, who not only conceded the personal space to undertake this doctoral study in the first place but was critical in the organisation of the ‘back office’ of the game conducted over many months. Special thanks are also due to her as well as to Messes Don Steele and Ron Ringer for their roles in the proof-reading and writing style, to Drs Phillip Emerson and Nick Stevens for their guidance on the statistical aspects of the game, to Graeme Jessup who provided feedback to early versions of the expert survey and Dr Ronald Gaudreau for critical feedback on the expert survey as well as for technical and writing aspects of the final draft of this thesis.
ix
x
CONTENTS CHAPTER 1 Introduction ...... 1 1.1 Study Background ...... 1 1.2 Study Motivation ...... 1 1.3 Academic Oversight ...... 2 1.4 Study Organisation ...... 3 CHAPTER 2 Ski Field Case Study ...... 5 2.1 Chapter Content ...... 5 2.2 Case Study Methodology ...... 5 2.3 Existing Academic Coverage of the Ski Industry Business ...... 6 2.3.1 Community and “‘Integrated” Resorts ...... 7 2.3.2 Competition within the two different Ski Regimes ...... 8 2.3.3 Social Dimensions of “Community” Organisations ...... 8 2.3.4 Asset Allocation in the two Regime types ...... 9 2.3.5 Summary of Ski Field Business Regimes ...... 9 2.4 Choice of Comparative Ski Fields ...... 10 2.5 Interviews ...... 11 2.5.1 Dolomiti SuperSki Italy ...... 12 2.5.2 Serfaus Region Austria ...... 14 2.5.3 NZSki New Zealand ...... 15 2.5.4 Victorian Ski Fields, Australia ...... 17 2.6 Analysis of the four Case Studies ...... 18 2.6.1 Overall Control ...... 19 2.6.2 Responsibility for the Transport (ski-lifting) Services ...... 19 2.6.3 Line Ownership ...... 20 2.6.4 Payment for Lifting Services ...... 20 2.6.5 Ski Field Planning ...... 21 2.6.6 Associated Activities to Encourage Patronage ...... 22 2.7 The Essential Business Regime Differences ...... 22 2.7.1 Generally ...... 22 2.7.2 The Profit Motive ...... 24 2.7.3 Ownership and Loyalties ...... 24 2.7.4 Consequences of Control Mechanisms ...... 26 2.7.5 Alignment of Interests ...... 26 2.7.6 Setting of Fares ...... 27 2.7.7 Lift Numbers per Regime ...... 28
xi
2.8 Summary of Regime Types ...... 31 2.8.1 Community Franchise Regime ...... 31 2.8.2 Conventional “Community” Regime ...... 32 2.8.3 ‘Integrated’ Model ...... 32 2.9 Summary of Case Study ...... 33 2.9.1 Regime Differences ...... 33 2.9.2 Previous Study ...... 33 2.9.3 Regime Peculiarities ...... 34 2.9.4 Prospects for Investigation ...... 35 CHAPTER 3 Defining a Community Franchise regime ...... 37 3.1 Chapter Content ...... 37 3.2 Principles of a Community Franchise Business Regime ...... 37 3.2.1 Community Franchisor Owners ...... 37 3.2.2 Community Franchisor Corporate Structure ...... 37 3.2.3 Community Franchisor Share Ownership ...... 38 3.2.4 Community Franchise Board of Directors ...... 38 3.2.5 Community Franchise Regime Objectives ...... 39 3.2.6 Community Franchise Regime System/Network Planning...... 39 3.2.7 Community Franchise Regime Rules of Operation ...... 39 3.3 The Community Franchise Regime Algorithm ...... 40 3.4 Summary ...... 42 CHAPTER 4 Literature Review ...... 43 4.1 Chapter Content ...... 43 4.2 Franchising ...... 44 4.2.1 Generally ...... 44 4.2.2 Franchising Operational Differences ...... 44 4.2.3 Franchisor and Franchisee Considerations ...... 46 4.2.4 Franchising in the Wider Economy ...... 46 4.2.5 Franchising in Public Transport ...... 47 4.2.6 The Special Franchising Case of Uber ...... 50 4.2.7 Summary of Franchising in Public Transport ...... 51 4.3 Public Transport Classifications ...... 52 4.3.1 Ski Lifting as Public Transport ...... 52 4.3.2 The Historical Legacy ...... 52 4.3.3 The Late 20th Century Reforms ...... 53 4.3.4 Summary of Regime Status ...... 60
xii
4.4 Demand Management and Subsidies...... 61 4.5 Monopoly Tendencies and Franchising ...... 63 4.5.1 History of Monopoly Behaviour ...... 63 4.5.2 Control of Monopoly Behaviour ...... 64 4.5.3 Monopoly Regulation ...... 64 4.5.4 How Franchising Affects Monopolies ...... 65 4.5.5 Exclusion of Competitors...... 66 4.5.6 Ownership ...... 67 4.5.7 Summary of Franchising and Monopoly Control ...... 67 4.6 Regulatory Capture and Franchising ...... 67 4.6.1 Generally ...... 67 4.6.2 Summary of Regulatory Capture ...... 68 4.7 Transport Network Design and Franchising...... 68 4.7.1 The Current Understanding of Network Design ...... 68 4.7.2 Definitions of Design ...... 69 4.7.3 An Augmented Definition of the Design Process ...... 70 4.7.4 The Design Process Applied to Network Planning ...... 71 4.7.5 Summary of Design and Network Planning ...... 73 4.8 Literature Review Summary ...... 73 CHAPTER 5 Application to Urban Public Transport ...... 75 5.1 Chapter Content ...... 75 5.2 Applying the Algorithm ...... 75 5.3 Detailed Examination ...... 77 5.3.1 Objective of Government...... 77 5.3.2 The Geographic Area of Community Franchise Regime Application ...... 78 5.3.3 Measuring Passenger Numbers ...... 79 5.3.4 Standards to Apply to Movement ...... 79 5.3.5 Estimation of the Effective Value to the Government ...... 79 5.3.6 The Commercially Responsible Body - The Franchisor ...... 80 5.3.7 Curtailment of Franchisor Functions ...... 83 5.3.8 Measurement of Transport Activity ...... 83 5.3.9 Reward to the Franchisor ...... 83 5.3.10 Publicising the Franchise Objectives and Standards ...... 84 5.3.11 Giving Scope to the Franchisee ...... 84 5.3.12 Review and Assessment of Franchisee Proposals ...... 85 5.3.13 Engagement of Franchisees ...... 86
xiii
5.3.14 Provision of Property Rights ...... 86 5.3.15 Commercial Pressures on Franchisees...... 90 5.3.16 Review of Payments to Franchisees ...... 91 5.3.17 Review of New Proposals ...... 92 5.4 Analysis of Essential Aspects of the Defining Algorithm ...... 92 5.5 Significant Distinguishing Features of a Community Franchise Regime: ...... 93 5.6 Summary of Regime Algorithm Application ...... 94 CHAPTER 6 Research Questions and Hypotheses ...... 95 6.1 Chapter Content ...... 95 6.2 The Scope of Inquiry ...... 95 6.3 The Research Questions ...... 95 6.3.1 Measurement of Quantitative Differences between a Community Franchise and a Government Enterprise regime ...... 95 6.3.2 Cooperation between Competing Line Owners ...... 96 6.3.3 Is a Community Franchise Regime transferable to Public Transport? ...... 96 6.4 Research Questions and Hypotheses ...... 97 6.5 Summary ...... 97 CHAPTER 7 Methodology ...... 99 7.1 Chapter Content ...... 99 7.2 Research Question One...... 99 7.2.1 Experimental Objectives ...... 99 7.2.2 Game Theory Considerations ...... 100 7.2.3 Experimental Model ...... 102 7.2.4 Experimental Authorship ...... 102 7.2.5 Experimental Participants ...... 103 7.2.6 Hypothesis Testing ...... 104 7.3 Research Question Two...... 104 7.4 Research Question Three ...... 105 7.4.1 Available Methodologies for Assessment ...... 105 7.4.2 Asynchronous Anonymous Computer-Based Surveys ...... 106 7.4.3 The Expert Evaluation Method ...... 108 7.5 Methodology Summary ...... 111 CHAPTER 8 Experimental Gaming ...... 113 8.1 Chapter Coverage...... 113 8.2 Detailed Methodology ...... 113 8.2.1 The Artificial City ...... 114
xiv
8.2.2 Game Operations ...... 115 8.2.3 Participant Suitability ...... 116 8.2.4 Genetic Algorithm ...... 117 8.2.5 Optimisation of the Network: The Decision Variables ...... 118 8.2.6 An Outline of the Excel Model ...... 119 8.2.7 The Calculation of Social Welfare ...... 120 8.2.8 Participant Interface ...... 121 8.2.9 Feedback to Participants ...... 122 8.2.10 Travel Demand ...... 123 8.2.11 The Multinomial Logit Model (MNL) for Passenger Allocation ...... 124 8.2.12 Calculation of Revenue and Profit ...... 126 8.3 Playing the Games ...... 126 8.3.1 Choice of Participants ...... 127 8.3.2 Data ...... 127 8.3.3 Individual Participant Differences ...... 127 8.3.4 Game Introduction...... 128 8.3.5 Operation of the Community Franchise Regime Game ...... 128 8.3.6 Operation of the Government Enterprise Regime Game ...... 128 8.4 Potential and Inherent Biases in Experimental Design ...... 129 8.4.1 Sampling Bias ...... 129 8.4.2 Researcher Biases ...... 130 8.4.3 Order of Play Biases ...... 130 8.5 Explanatory Potential and Limitations of Experiment ...... 131 8.6 Experimental Proceedings ...... 131 8.7 Summary ...... 133 CHAPTER 9 Results of Experimental Gaming ...... 135 9.1 Chapter Content ...... 135 9.2 Suitability of Participants ...... 135 9.3 Results from the Experiment ...... 137 9.3.1 Analysis of Research Question No 1: Social Welfare ...... 138 9.3.2 Regime Ranking ...... 139 9.3.3 Effect of Experience on Social Welfare ...... 139 9.4 Genetic Algorithm (Evolver) Result from the ‘Final’ Experiment ...... 142 9.5 Further Considerations of the Experiment ...... 143 9.5.1 Passenger Number Considerations ...... 143 9.5.2 Highest found Resource Allocation ...... 143
xv
9.6 Further Observations of Human Participants ...... 145 9.6.1 Community Franchise Regime Game Strategies ...... 146 9.6.2 Government Enterprise Game Strategies ...... 148 9.6.3 Participant Performance ...... 149 9.7 Discussion ...... 150 9.7.1 Planning and Performance...... 150 9.7.2 Regime Performance ...... 151 9.7.3 Complexity of the Model ...... 153 9.7.4 Analysis of Results by Order of Play and Age ...... 154 9.8 Research Question 2: Competitor Cooperation ...... 155 9.9 Summary ...... 157 CHAPTER 10 Expert Survey ...... 159 10.1 Chapter Content ...... 159 10.2 Dimensions and Questions ...... 159 10.2.1 Ownership and Direction ...... 159 10.2.2 Budgets and Corporate Control ...... 160 10.2.3 Network Planning and Operations ...... 161 10.2.4 Control and Maintenance of Standards ...... 162 10.2.5 Fares ...... 162 10.2.6 Alignments of Interest and Regulatory Capture ...... 163 10.2.7 Overall Feasibility and Critical Features ...... 163 10.3 Respondents ...... 164 10.4 Mechanism for Data Collection ...... 164 10.5 Research Question Three Summary ...... 165 CHAPTER 11 Results of Expert Survey ...... 167 11.1 Chapter Contents ...... 167 11.2 Participants ...... 167 11.3 Written Results ...... 167 11.3.1 Ownership and Direction ...... 169 11.3.2 Budgets and Control ...... 170 11.3.3 Network Planning and Operations ...... 172 11.3.4 Control and Maintenance of Standards ...... 173 11.3.5 Fares ...... 174 11.3.6 Alignment of Interests ...... 175 11.3.7 Overall Feasibility and Critical Features ...... 177 11.4 Dimensional Group Figures ...... 178
xvi
11.5 Feedback on Results from Participants ...... 179 11.6 Discussion of Major Respondent Issues ...... 179 11.6.1 Ownership and Direction ...... 179 11.6.2 Budgets and Corporate Control ...... 179 11.6.3 Fares...... 180 11.6.4 Network Planning and Operations ...... 180 11.6.5 Treatment of Developing Areas of low demand ...... 180 11.6.6 Alignments of Interest and Regulatory Capture ...... 181 11.6.7 Pilot program ...... 181 11.6.8 Asset Value Realisation ...... 181 11.7 Summary ...... 182 CHAPTER 12 Conclusions ...... 185 12.1 Chapter Contents ...... 185 12.2 Review of Study Process ...... 185 12.3 Answers to Research Questions ...... 185 12.3.1 First and Second Research Questions: Quantitative Evaluation ...... 185 12.3.2 Third Research Question: Expert Evaluation ...... 186 12.4 Limitations of this Research ...... 186 12.5 Further Research ...... 187 12.5.1 Field Study of Community Franchise Regimes ...... 187 12.5.2 Laboratory Experiments ...... 188 12.5.3 Testing a Real-Life Application of a Community Franchise Regime ...... 188 12.6 Contribution to Knowledge ...... 189 12.6.1 Documentation of the Community Franchise regime ...... 189 12.6.2 Definition of Design ...... 189 12.6.3 Effects of Regime on Social Welfare ...... 190 12.6.4 Applicability of Regime to Sydney ...... 190 12.7 Final Word...... 190 Appendix A – Case Study Ethics Approvals ...... 193 Appendix B Expert Survey Ethics Approval ...... 195 Appendix C Expert Survey as sent to Participants ...... 197 Appendix D Responses to Expert Survey ...... 210 Appendix E Numeric Data from Expert Survey ...... 228 Appendix F Summary of Respondent Comments on Expert Survey ...... 230 References ...... 232
xvii
xviii
FIGURES Figure 1 The three business models of Public Transport (van de Velde and Preston, 2013) ...... 53 Figure 2 The four business models of public transport following Thredbo 2015 ...... 54 Figure 3 The Artificial City with Public Transport Network ...... 114 Figure 4 Input Screen for Community Franchise regime experimental game in Excel . 122 Figure 5 Feedback portion of Community Franchise regime screen on the Input tab... 123 Figure 6 Demand and Intensity of Line Use shown in a ‘mimic’ panel of the network .. 124 Figure 7 Ranking of all Social Welfare results from both regime games with Human Participants with that of Evolver ...... 139 Figure 8 Actual resource allocation of the best solution found by Evolver ...... 144 Figure 9 Actual resource allocation of the best solution by humans in the Community Franchise regime game...... 145 Figure 10 Actual resource allocation of the best solution by humans in the Government Enterprise regime game ...... 145 Figure 11 Comparison of Group Producer Surpluses for highest scoring Human Participant Line Group Profit Results with the Highest found Evolver Result ...... 152 Figure 12 Coding of Participant Responses by Expressed Confidence Level ...... 168 Figure 0-1: Image of Completed Ethics Approval https://irma.sydney.edu.au/production/content/rqf ...... 193
xix
TABLES
Table 1 Details of Ski Field Numbers and Business Regime Type ...... 11 Table 2 Details of Case Study Interviews ...... 11 Table 3: Dolomiti SuperSki's figures on lift types and numbers ...... 13 Table 4 Response of Dolomiti SuperSki to Franchising Requirement ...... 23 Table 5 Summary of Case Study Findings ...... 24 Table 6 Lift Numbers per Ski Run Length ...... 30 Table 7 No of Lifts/Run Length, Comparison of European, US and Dolomiti SuperSki fields ...... 30 Table 8 Application of Dolomiti SuperSki Community Franchise regime workings, and the generation of Descriptive Algorithm...... 42 Table 9 Regime Characteristics ...... 61 Table 10 Public Transport Regime Differences ...... 74 Table 11 Application of Community Franchise Regime Descriptive Algorithm to Urban Public Transport ...... 77 Table 12 Sydney in the State of NSW Australia (sourced from Google Maps 2016) ...... 78 Table 13 Branching Tree Typology under the Current Government Enterprise regime (simplified and in concept only) ...... 88 Table 14 Possible routes for lines in a Community Franchise regime with examples of some other potential lines (shown light) ...... 89 Table 15 Research Questions and Hypotheses ...... 97 Table 16 Comparison of Techniques for Group Evaluation of Complex Ideas ...... 106 Table 17 Dimension of Inquiry and Discipline Areas for Recruits ...... 108 Table 18: Results of social welfare of Government Enterprise ‘beta’ game Non- Professional Group (N) and Professional Group (P) ...... 137 Table 19: Social welfare results of the ‘final’ Government Enterprise (GE) game and the ‘final’ Community Franchise (CF) game ...... 138 Table 20: ‘Final’ Government Enterprise regime game results for Social Welfare, divided into Experienced and Inexperienced Participants ...... 140 Table 21: Production of Social Welfare by ‘Experienced’ Participants only, from the ‘final’ Government Enterprise game and the ‘final’ Community Franchise game ...... 141 Table 22: Comparison of ‘Inexperienced’ Participants in both Government Enterprise regime and Community Franchise regime ‘final’ games ...... 142 Table 23: Social Welfare production from ‘beta’ games: Results due to Different Orders of Play ...... 155
xx
Table 24: Proxy Social Welfare production from ‘beta’ games: Results of Age Grouping ...... 155 Table 25 Dimensions and Participants ...... 167
xxi
CHAPTER 1 INTRODUCTION
This thesis is motivated by a desire to investigate a phenomenon in public transport which seems to have escaped the critical review of academic study and which appears to have the potential for wider application than in its current local manifestation. This introduction explains how and why this study came about, the specific issues it seeks to address, and an outline to provide the structure of the study that follows in the twelve chapters that constitute this thesis. 1.1 Study Background The genesis of this study was in the ski fields of the Dolomites in northern Italy. Without any anticipation of discovery apart from traversing the mountains of the Trento area of the Italian alps, we listened to the guides who took us on our ski holiday providing ongoing commentary on how we were able to take advantage of this largest of all single ticket ski fields. Expatriate Australian guides, familiar with conditions in our native country, provided an insight into the different ways that this part of Italy is organised. Their explanation included aspects as diverse as the range of foods, to the way that building construction was approved and is contracted, the local governance of the valley communities but in particular, and of relevance to this study, to the way that the ski lifting was organised as a business. The described business function was profoundly different from the way that ski lifting was organised in virtually any other ski field that could be easily recalled. The Dolomiti ski fields is a facility that distinguishes itself by the way it has significant and continual improvement in the lifting infrastructure each passing year. Ultimately, this was found to be the result of entrepreneurial franchisee lift owners responding to the demand as they observed it and having the opportunity to propose and undertake new lines and improvements in lifting equipment and service they see as necessary or an improvement over existing arrangements. There is no reliance on a centrally directed plan like all the other ski fields that were known to this researcher. The distinguishing feature appeared to be how franchisee operators respond to market forces with the outcome of great convenience in layout, a massive choice of options of ski lifting with a high standard of the ski lifts that were on offer at prices that were some of the lowest in Europe. 1.2 Study Motivation Having just completed an MPhil with a focus on the ways that people move around Sydney, it was evident that most of the urban travel in Sydney is by car, in the order of 80%, and only a small percentage is by public transport, in the order of 10%. The reason most cited in official surveys of the travelling public for this modest public transport use was a lack of services to desired destinations. As a result, many people have no alternative other than to rely on their cars, which of course is a self-reinforcing phenomenon. Public transport as a function is not dissimilar to that of ski lifting, but the business method of the ski lifting observed had certain features that were profoundly different from the way that public transport in our cities is organised. The most important difference is
that each ski lift, which is, in fact, a fixed-line of public transport, is a separately owned small business that provides the owners with their living. This is a crucial difference in the operations of public transport that exists in other ski fields and in our cities. In the Dolomites, appropriate incentives seemed to be built into the system including the need for each line of transport to:
• attract and retain as many paying passengers as possible, • have realistic regard to the amount invested in the infrastructure, and • satisfy the demand most economically with regard to both capital investment and operational expenses. These are features not evident in most public transport systems where the operators are not necessarily directly dependent for their living on the line for which they may have responsibility. It was this feature of tangible responsibility that provided the stimulus to investigate whether this unusual way of organising the ski lift business had any application to an urban public transport service. This became a key part of the puzzle in understanding the nature of what is termed a Community Franchise. This investigation comes at a time of great change in attitudes to urban mobility. There is significant growth in public transport use at the same time as the introduction of disruptive technological change, from driverless cars to new business models of taxi engagement. At the same time, there does not seem to be a consensus as to how the future should be approached. In Sydney, Australia, there is large scale urban expressway building seemingly at odds with major heavy and light rail infrastructure, alongside new forms of mobility being trialled as on-demand services. This seeming lack of consensus about how to approach urban mobility provides some relevance to this study. It provides a different perspective from those business regimes currently being considered for public transport in our cities and suggests mechanisms for the resolution of conflicting approaches to its provision. This study aims to subject the ideas of this form of a public transport business regime to a rigorous in-depth analysis, such that it can be evaluated as to its potential use in urban public transport. 1.3 Academic Oversight It is understood that personally held beliefs can be, at best, encumbrances and, at worst, dangerous so this study is approached with the spirit of what has been quoted by Tettlock and Gardner as beliefs being hypotheses to be tested (Tetlock and Gardner, 2016 p127). This idea of the observation of a different way of organising the business of public transport is being explored and evaluated, not promoted. It was thus considered most fortunate that the opportunity was presented to do this study as a disciplined, structured, and supervised investigation at a respected transport academy, the Institute of Transport and Logistic Studies. This organisation is a distinct entity within the Business School at the University of Sydney and has a global reputation for scholarly excellence. This study strives to meet its standards of thoroughness, impartiality, and objectivity.
Page 2 of 259
1.4 Study Organisation This thesis is organised into three broad sections. To provide a definite object of study, the first task of this thesis is to present the case study of the ski fields and define the operations of the business regime that is the focus of the study. As well, in the first section, when the object of study in its native setting is evident, the literature is then reviewed to establish the extent to which this business regime is present in urban public transport and to explore the types of issues that have been brought to this researcher’s attention as requiring investigation. Care is taken to clarify the different uses of the term ‘franchising’ as its use differs in many fields of application, but particularly in the public transport industry. In this context, franchising has a meaning not consistent with some other disciplines and requires careful navigation when used in the context of ski lifting and taxi services, as well as many service industries. Only when it is evident that this franchised regime of public transport is either unique or unreported and has not been academically scrutinised, the application to the urban public transport to the city of Sydney is proposed. The application of the definition to the Sydney metropolitan area establishes the object of study for the remainder of this thesis. The methodology then shows how this object of interest, this business regime, is to be interrogated. The investigation is reported on in the next two sections, following the case study and regime definition of the first section. In the second thesis section, the defined regime is examined quantitatively by a computer experiment in the form of a game. In the third thesis section, it is examined qualitatively by a survey of relevant experts. Conclusions are drawn when these two contrasting aspects investigating the phenomenon are completed. So, to establish what the peculiarities of this business regime in question are, this thesis now turns directly to the case study of Dolomiti SuperSki to compare the workings of this regime with that of other ski fields which organise the transport of their skiing patrons in more conventional ways.
Page 3 of 259
Page 4 of 259
CHAPTER 2 SKI FIELD CASE STUDY 2.1 Chapter Content This chapter investigates the business regime underlying the operation and control of ski fields in the Italian Alps so that it is evident how the regime operates and exactly what a Community Franchise, being the term used by the owners of Dolomiti SuperSki to describe the business regime under which they operate, actually is. By the completion of the next two chapters, the regime will be described, defined and the peculiarities identified for further study in the remainder of the thesis. The methodology of this case study is first outlined, then a review of the literature about ski business organisation establishes the extent to which this field has been previously covered. The findings of the case study of Dolomiti SuperSki in the Italian Alps and three other ski fields operating under different business regimes in other parts of the world are then presented. This subject was presented as a paper to the Thredbo Conference on Competition and Ownership in Public Transport in Chile in 2015 (Emerson et al., 2015a). Although substantially based on this paper this chapter is augmented to consider the feedback provided by participants in that conference and its location within this thesis. The four businesses discussed represent a geographic and an organisational diversity in the way the business of ski lifting, as a system for the transport of the public in that environment, is arranged. The criteria used to enable a comparison of essential business characteristics are outlined. Then, the interviews for each case study are reported, with each interview analysed for:
• the ski field’s approach on overall control, • how their operations work, • which parts of the organization have responsibility for the transport services, • the ownership of the lines and other components of the ski service, • how lift operators are paid/rewarded, • how improvements to the ski fields are planned, i.e. what is the method for resource allocation, and • how associated activities related to the ski lifting function. The chapter continues with a discussion on the findings with a central focus on where the essential interests of the participants in each regime lie. The significant findings of this chapter are the confirmation that the business regime run by Dolomiti SuperSki is a traditional franchise regime with ownership of the operating franchisor being the local community, (giving rise to the terminology of a Community Franchise), with individual lines of transport being in the ownership of independent franchisees and that this business regime appears to be unique in the world of ski resort business regimes. 2.2 Case Study Methodology There are several ways of conducting case studies. The nature of the phenomena being studied here influences the approach that is adopted. Case studies are often used for
Page 5 of 259
exploratory, explanatory, or descriptive purposes (Ghauri and Grønhaug, 2005). They expand this to say that case studies are a “useful way of investigating phenomena that is difficult to study outside of its natural setting” and when the “concepts and variables under study are difficult to quantify” (p109). The principal object of this study fits these criteria. As well, the business model of interest has not been extensively, if ever, studied and one of the first tasks in such a situation is to first describe the phenomena to provide a suitable objective focus for later analysis. A case study approach enables this to occur. This investigation is framed as a qualitative study and follows the procedures of case studies as identified by Creswell (2012). He notes that there is a range of legitimate approaches to any case study and the circumstances of each investigation determine the most appropriate structure. He also notes that the design of the research process is best kept loosely defined to be able to appropriately react to what is discovered in the process and extract the most from an emerging situation of discovery. The chosen structure was influenced by the fact that it was both necessary to investigate the phenomenon in its natural setting and to contrast this seemingly unique case of the Dolomites with other ski lifting business regimes. Accordingly, after first describing the Dolomites ski lifting business in its natural setting, the study compares the Dolomites example against other ski lifting regimes. Three additional ski lifting cases were selected for examination and comparison. The selection process involved finding ski fields that are of significant size and importance in any market, as well as those being representative of different business models, are fully described. It was the original intention to have as many Community Franchise business regimes in the sample as possible, but as reported, it appears that Dolomiti SuperSki is the only representative of this regime type in the world. In outline, a semi-structured format was used to interview a member of the senior management of each ski field organisation. The objective was to have each interviewee describe in their own words, the type of organisation that operated and the reasons why this regime may have been adopted. The chapter dealing with the case study provides details of the specific interview techniques, questions, and results. The Ethics Committee of the Sydney University Business School, (Project No: 2014/976) reviewed the project and their approval is attached in Appendix A. 2.3 Existing Academic Coverage of the Ski Industry Business The literature on ski field business regimes is very small, but there are several who have given it some attention. The economic aspects of ski field development have been addressed by Clydesdale (2007) who found that practically all literature on skiing was on marketing or environmental issues but not on economics or business. Apart from the very limited contributions he could identify in the economic field, he pronounced that “the cupboard was bare”. Clydesdale reported on aspects of ski resorts where “geographic terrain is transformed into customer value”, a very apt description of what ski lifting does.
Page 6 of 259
2.3.1 Community and “‘Integrated” Resorts Clydesdale’s academic paper is one of the few to examine skiing as a business, but it does not provide much illumination on lifting as transport. Clydesdale does go over the considerations of the types and characteristics of the different ski lifting devices, identifies up to twelve different types of lifting devices, and the considerations that go into the choice of them, such as height, length, demand, comfort, terrain, the difficulty of slopes at the destination, wind effects, etc. He describes how the European alpine resorts are generally arranged in community marketing organisations with individual ownership of the different businesses such as accommodation, food and beverage, and ski lifting. Generally, the European examples are known to be ‘Community’ organisations with a non-‘Integrated’ business model whereas the US examples are nearly always ‘Integrated” resorts with a single owner of the whole ski field. This form of ownership provides control over all operational aspects including investment in any aspect, including ski lifting, and allowing cross-subsidisation of other activities or infrastructure, e.g. accommodation, hire or food, for overall benefit to patrons and the owner. However, he does not mention the peculiarities of the Dolomites community example with the specific relationships established with marketing and recompense for lift owners. The “Community” model in parts of Japan is slightly different from the European ones; these Japanese examples have been reported on by Patchell (2014). He examined the effects of the approach to differentiated common pool resources as a means of explaining the behaviour of the organisations that occur in that part of rural Japan. Skiing was a post-war phenomenon that was in decline after a boom that peaked in the 1980s. The area at that time was sufficiently successful to host the Winter Olympic Games in Nagano in 1998 which was in part an exercise in encouraging its resurgence as a winter destination and activity. He reported on the way this group of separate communities had reorganised to meet the demands of declining demand by forming community organisations that allowed for the development and presentation of a large ‘Integrated resort in the manner of the European and North American models. When demand collapsed from the high point of the Olympics the existing order suffered and also collapsed. Then a more “inclusive, trusting organisation” evolved to exploit the different approaches and requirements of locals, city refugees (from Tokyo), and expatriate ski entrepreneurs (mainly from Australia). Significantly, there is local ownership of many of the separate businesses as in the European “Community” model, but there is not the ticketing or resort integration as occurs in the Dolomites. Patchell distinguished between the ‘integrated’ and the ‘community’ regimes and also reported that there is insufficient evidence to say if one regime type, the ‘Integrated or the “Community” is superior in economic terms: ski fields are located in diverse situations all with their particular problems and possibilities and this confounds any analysis by making the direct comparison of fields difficult. ‘Integrated’ resorts have much more efficient methods of allocation between competing requirements, e.g. between ski lifting, equipment hire, and accommodation. Where there is separate ownership of businesses
Page 7 of 259
in each of these categories, there is not the opportunity for cross-subsidisation to provide an optimised overall positive experience for the visitor, something that the ‘Integrated’ resort does regularly. A major finding was that the disparate elements of the community come together on key issues to jointly solve common problems in ways that maintain competitive selection of individual firms, which in this case has facilitated downsizing, and at the same time maintained profitability and created social capital. The mechanism that he identifies as adding to social capital, being the creation of the “Community” organisations, brings diverse interests together to influence the resort’s evolution and development. This, he shows, allows these diverse businesses to coordinate and sublimate selfish interests for overall and possibly delayed individual gain. 2.3.2 Competition within the two different Ski Regimes Transaction cost theory would suggest that a single organisation would be more efficient, but as Beritelli et al. (2007) show with European examples, this is not at all clear. Large organisations suffer from their size with the management of each separate component becoming over time more remote from the central direction of the head office the larger they become. There is then a tendency for managers to seek to look after selfish interests ahead of the group interests. On the other hand, the competition between the separate businesses of the “Community” organisation allows for a more optimised allocation of resources within each business and this appears to compensate to a large extent for the lack of coordinated integration. This approach puts the positive effects of property rights theory in a direct comparison with the negative effects of principal/agency theory and it appears unclear to those who have investigated this in ski field performance, which theory prevails. However, it appears certain to Beritelli et al. that the operations of both theories compensate to some degree for each other. They outline the different ways of analysing these ski field organisations using a simple dyadic view for the hierarchical ‘Integrated’ large enterprise model, and a network view for the multi-component “Community” model. Viewing it from one perspective only does not provide a sufficient basis for the complexities of the “Community” model and may overlook the essence of its workings. Using a network approach allows a dynamic point of view and captures dimensions such as trust and knowledge. Beritelli et al. examine the two regimes taking into account transaction costs, power asymmetries, interdependence, trust/control, knowledge and informal/personal connections. They conclude that such a study allows a reasonable explanation of the mechanisms at work in the various ‘“Community”’ and ‘Integrated’ organisations that predominate in Switzerland’s destination communities and is a necessary approach to obtain a reasonable understanding of the dynamics of competition operating in the two major models of ski field business organisation. 2.3.3 Social Dimensions of “Community” Organisations The ways that the ‘Community’ regimes operate are shown to largely compensate for the perceived and obvious advantages of a single directed ‘Integrated’ regime model. These social dimensions show the very real effects of some practices that are not explained by
Page 8 of 259
theory based purely on the traditional ‘greed and fear’ of classical economics. Wider considerations that include reciprocity are shown to be crucial to a fuller explanation of economic behaviour and are increasingly being used in addition to those basic concepts of the economics of greed and fear (Frijters and Foster, 2013 pgs.143-298). The network concept of Beritelli et al. allows these wider views to be incorporated. Other academics active in the immediate area, both geographic and discipline, have examined the changing role of the community organisation and its capacity to serve the current tourist trade (Franch et al., 2006, Franch et al., 2003). There is no reference to the business of ski lifting regimes in their analyses, but there is more general comment on the capacity of socially based organisations to function effectively. The sparse situation on academic papers on the topic of the models and operations of ski field businesses, as reported by Clydesdale, is not disproved by this review of the literature: the cupboard is indeed bare. It has been found that most studies are looking at the technical aspects of skiing, e.g. of bindings, or chair design, or are concerned with tourism, sporting/health, or environmental issues. The commercial aspect of skiing is simply not a topic that has been of great interest to academics. However, the nature of the organisations that run ski resorts as being either “Community” or ‘Integrated is clear, as is the need to adopt a comprehensive economic viewpoint in examining and understanding their functioning. Also, there is no comprehensive coverage of the business arrangements for the public transport of ski lifting that has the specific arrangements of the Community Franchise regime of the Dolomites: it appears to be is an unstudied phenomenon. The asset allocation arrangements of an integrated scheme may be different from both a Community scheme and from the specific example of the Community Franchise scheme. This peculiarity of the Community Franchise and the differences to both other regime types identified here is explored in this thesis. 2.3.4 Asset Allocation in the two Regime types The need for potentially competing owners to jointly develop shared assets is a particular problem of the “Community” schemes. This cooperation is handled differently in fields with differing business regimes, with the Japanese community fields operating in a way that may appear suboptimal, but which Patchell (2014) describes as having coped with extreme demand fluctuations in recent years and today still maintains viable operations. The need to cooperate with competing commercial operators in ski field asset allocation is avoided in the ‘Integrated’ schemes where all decisions are made by the one organisation, with absolute discretion as to resource allocation operating within the control and command regime of a normal commercial or government enterprise. This integrated control enables direct subsidisation of different operations at the discretion of the owning organisation to optimise overall financial performance. The integrated fields’ main concern is with other ski fields with whom they compete. 2.3.5 Summary of Ski Field Business Regimes This review of the academic literature of ski lifting businesses has established a relative paucity of interest. However, some matters are clear. There are conceptually two types
Page 9 of 259
of business regimes that can be identified operating in ski fields. One is the “Community” type with widespread ownership of assets and geographic-specific local ownership of assets and business operations with no tight control over their integration or planning. This regime predominates in Japan and Europe. The second is the ‘Integrated’ type which has strong corporate control of the whole or most of the ski field operations, such as occurs in North America and Australasia. There is no obviously better economic performer in either operation although they both have specific mechanisms for the resolution of problems and the allocation of resources. 2.4 Choice of Comparative Ski Fields It was not immediately obvious when first looking for comparable fields, i.e. ski fields of the same scale as those in the Dolomites; that the Dolomites are, by a significant margin, the largest ski field in the world in terms of the number of lifts serving an integrated ski field accessible on a single ticket. Dolomiti SuperSki in its company information record that the number of lifts able to be accessed by a single ticket is 460. The 12 separate valleys in which these lifts are located have slopes totalling 1,200 km in length. However, it is not necessarily recognised as the world's largest in terms of area of terrain or slope length. The largest ski field as recorded by Skiresort Service International GmbH, a German company self-proclaimed as specialising as a “…..leading publication and distribution company for ski resort information and data“, uses a different measure. They treat each valley in the Dolomites as separate fields and record the largest field measured by total slope length and note that this is Les 3 Vallées – Val Thorens/ Les Menuires /Méribel /Courchevel in France with 600 km of runs. However, that field has only 148 lifts to serve an area that is in the order of 1200 square kilometres of terrain. The different methods of measurement are of marginal importance. They were of importance only for selecting comparison fields and the operational reach of their markets. This study is looking at the business regime of operation and for this specific purpose, different business regimes may be appropriate at different scales. For this study, the ability to access the lifts on the single ticket is taken as the relevant criteria by which to judge the size of the overall field. As a result, to obtain relevant examples with which to compare Dolomiti SuperSki, certain criteria were employed. Ski fields had to be of significant size in the market that they serve. As well, it was felt necessary to get at least one other field in the same market as Dolomiti SuperSki, i.e. that of the European Alps, within the study group. Austria was chosen to provide a comparison with a field in the same European market as the Dolomites and in that country, fields with similar business models to that used in the Dolomites were sought. Other community-based models were found in Europe, but not with exactly the same business models for the ski lifting task. That field in the ski area of Serfaus has a significant number of lifts under a single ticketing system, is close to the Dolomites and represented another community-based regime, although one that is different in significant detail from that of the Dolomites. The fields of Australia and New Zealand were targetted, partly for practical reasons of proximity and because it is known that their business models were different from those in the European fields, being the
Page 10 of 259
‘Integrated’ model common in the US (Patchell, 2014). The Australasian fields are smaller than the European fields but of the same scale of those in North America. Accordingly, the largest available ones in Australasia were chosen. The fields in Australia and New Zealand were combinations of several separate fields under common ownership, (NZ) or multiple fields administered by the same government organisation (Australia). Not all fields approached were able or willing to participate in the study, but those that were eventually included have the similarities of national importance and scale to justify inclusion. Each one has a business model that is unique to the study. Relevant numbers for each field studied are shown in Table 1.
Ski Field Number of Total Number Business Regime Separate Fields of lifts Dolomites Super Ski 12 connected 460 Community Italy Franchise Serfaus Region Austria 3 connected 67 Community New Zealand Ski 3 separate 10, 7, 4 Integrated Victorian Ski Fields 5 separate 14, 14, 22, 7, 2 Integrated Table 1 Details of Ski Field Numbers and Business Regime Type The business Regime classification is based on the analysis of Patchell (2014) where he notes the two basic types of approaches to ownership and development as either “community“ or ‘integrated’. In the case of Dolomiti SuperSki, it is a “community” organisation but also is distinct in being a franchise regime within that broader classification. The Victorian Ski Field has elements of both integrated and community, in that it is owned by the Victorian government but is leased out to individual contractors who undertake the work in each field on an ‘integrated’ basis. 2.5 Interviews Interviews were arranged with a senior executive of the controlling organisations of each ski field and a summary of those interviews was provided to each interviewee to confirm the accuracy of the understanding gained. In all cases, personnel were board-level representatives, or equivalent, of the organisations responsible for the overall business. Interviews were conducted in the period from February 2015 to March 2015. The detail of those interviews is shown in Table 2 below:
Ski Field Position Date Dolomite Super Ski, Italy Marketing Director March 2015 Surfaus, Austria President Serfaus Ski Region Tourism March 2015 Organisation NZSki, New Zealand Chief Executive Officer February 2015 Victorian Ski Fields, Executive Officer Alpine Resorts Co- April 2015 Australia ordinating Council Table 2 Details of Case Study Interviews Each interview followed a semi-structured technique. The fundamental approach sought the views of the interviewee as to the reasons why the business regime of their organisation was used, how it may have originated, developed and generally operates, the methods of payment for the operation of the lifting devices and to provide special
Page 11 of 259
comment on the planning of new infrastructure. It was possible and was found necessary in some cases to arrange for questions to be asked by email after the initial relationship had been established by personal interview. The summarised results of each interview follow. 2.5.1 Dolomiti SuperSki Italy The valley communities of the area had formed the equivalent of ‘chambers of commerce’ several decades previously, acting as marketing organisations to promote and protect the tourism of the areas in which they are located. Dolomites SuperSki is an incorporated, commercial enterprise owned by a federation of twelve of these local marketing organisations created to promote and facilitate the tourism activities of a federation of the twelve valleys. Dolomiti SuperSki has the exclusive responsibility to approve and regulate the ski lifting system; no individual or organisation can undertake ski lifting using Dolomiti SuperSki’s ticketing system without the approval of Dolomiti SuperSki and acceptance of the form and terms of payment for the service provision. In this sense, it has a regulatory/coordinating role as well as undertaking the principal activities of ticketing for the lift system and marketing that would be less effective and more expensive if undertaken by individual valley organisations. Lifting is not undertaken by Dolomiti SuperSki itself, but by franchisees which are independent firms or individuals who propose, finance, and operate the lifting devices. Dolomiti SuperSki acts as the franchisor, setting conditions and evaluating private commercial entities that propose specific transport arrangements under its ticketing system. They advise that there are 130 separate lift and slope management companies all with managerial and financial independence. All these companies make their own decisions as to investment choice and risk management. Dolomiti SuperSki was one of the first lift ski operators to install contactless ticketing that is retained in the clothing of the individual skier and remotely operates the entry turnstiles. One ticket can buy access to the whole system of 460 lifts, or cheaper tickets can be purchased for access to just specific areas. This revenue finances the whole ski lifting operation as well as the running of Dolomiti SuperSki which retains a percentage of turnover for administration purposes, and which, at the time of the interview, was set at 3% of ticket sales. There is no contribution by participating valley organisations to Dolomiti SuperSki operations; it is all financed out of the 3% retained from ticket sales. Recompense by Dolomiti SuperSki to each lift operator is purely on a per passenger basis; each patron who activates the turnstile sends an electronic message to Dolomiti SuperSki who then pays the operator on the agreed basis. The level of recompense for each patron carried takes into account the height difference that the lifting device has from top to bottom, the length of the run, and the type of lifting device, e.g. gondola, chair, T-bar, etc. The rates used for each category are updated yearly. These considerations enable the formula and the passenger charges to be agreed upon by all twelve valley associations. This formula and the level of recompense is common knowledge to all participants in the industry.
Page 12 of 259
An effort to encourage patronage is at the discretion of each lift owner who makes a commercial assessment of possible investment and the additional recompense that is expected by the use of the formula for the carriage of passengers. The operator takes all risks associated with new investment and in all cases, there is no involvement by Dolomiti SuperSki in the planning of new investment. All improvements are as a result of individuals making a commercial assessment of their opportunity and the recompense available for the carriage of patrons. Proposals by franchisees are formulated as a result of this consideration. Using only this control of the amount per passenger that the formula provides, the fields have evolved over the 40 years of Dolomiti SuperSki’s existence from 270 lifts to 460 with a ten-fold increase in passenger numbers. There are multiple instances of line improvements over the years, e.g. several lifts being replaced by a smaller number of larger faster lifts, upgrades from chair lift to gondola, etc. The following figures in Table 3 were supplied by Dolomiti SuperSki to show the way these adjustments to capacity were introduced over twenty years.
1992/93 1999/00 2008/09 2014/15 Ski Lifts 248 198 143 127 Chair Lifts 142 139 128 116 Detachable Chair lifts 23 59 93 100 Detachable gondola ropeways 22 58 73 Cable cars 25 21 22 23 Totals 460 457 444 439 Table 3: Dolomiti SuperSki's figures on lift types and numbers They advised that these changes to type and capacity of lifting infrastructure have resulted in fewer facilities with greater coverage. Dolomiti SuperSki assumes that these are satisfactory decisions for the transport providers involved, although there is no way of Dolomiti SuperSki knowing if this is the case, as Dolomiti SuperSki is not a party to those investment decisions. Lifting operations are assets that are owned by their franchisee operators and can be bought, sold, mortgaged, bequeathed, etc. similar to any piece of real property. The investment in each asset is purely a matter for the lifting operator and the value for each line is a direct result of its net return based on passenger numbers, its capital, and its running costs. Dolomiti SuperSki does not put any restrictions on ownership of assets by individuals or companies. There is a great variety of ownership groupings, from single companies owning many lifts in one area, to multiple companies owning just one or two in other areas. The regime operational characteristic of fares being set and collected by the franchisor ensures that monopoly tendencies cannot take hold amongst the franchisees, regardless of their geographical concentration. A distinctive feature of this ownership arrangement is the way that the lifting companies take responsibility for snow-making and slope grooming and have these activities undertaken by contract or by use of their own resources. This ensures that the areas in which the lift companies operate are attractive as a ski destination which encourages
Page 13 of 259
patronage of the slopes and the lifts that operate in the area. It also ensures that this aspect of expenditure is most efficiently undertaken as there is a desire to have this done at least cost but to the best effect. The grooming aspect of operations can be complicated by the existence of multiple lift owners in any particular ski area with different levels of investment. These lift owners negotiate with each other as to individual contributions to the service and the nature of that service, (e.g. where it operates, standards of snow-making, slope grooming, etc.). Despite this complication, the system works effectively as it is in the interests of the lift operators to have the most attractive groomed slopes. Dolomiti SuperSki is not involved in this aspect, nor ski schools, restaurants, accommodation, etc. These latter services tend to be provided by completely independent operators, unrelated to either Dolomiti SuperSki or to the lift operators. Dolomiti SuperSki collects detailed information on the operation of the whole system and uses this information to pay each operator. It shares the information, which is relevant to each valley, confidentially with each individual valley organisation. As well, it shares information with individual companies, but only that which directly affects each company. 2.5.2 Serfaus Region Austria The business organisation of the Serfaus regional ski area has similarities to that of Dolomiti SuperSki but is different in several important ways. The local tourist association, representing two significant ski mountain areas and the two local communities of about 6,000 persons each, organises the marketing and the ticketing of the two ski regions with electronic detection and direct payment to the two lifting companies as for Dolomiti SuperSki. The Serfaus tourist organisation has the same function and the same objectives as Dolomiti SuperSki, i.e. it is serving the economic ends of the local region and all functions that it provides are subservient to this. Any profits are reinvested in the region and thus benefit the inhabitants of the region. One of the significant ways in which the Serfaus region differs from Dolomiti SuperSki is in the manner in which recompense to the lifting companies is calculated. In the Serfaus region, there are only two lifting companies and they share the revenue allocated to lifting in proportion to the time spent by skiers in each of the two separate regions that the two companies serve. This is seen to be a workable division of the income even though it is unrelated to the extent to which the actual lifting devices are used. This method of payment has the acknowledged effect of rewarding aspects of the two regions that could distort direct economic efficiency. People may spend time in one region because of various factors including the weather, the restaurants, or other facilities, but it is the lift companies that receive the proportion of revenue that is split in proportion with the time spent by patrons in each region. History and circumstances at start-up appear to have been instrumental in this arrangement being adopted. Provided the two regions are approximately equal in size and influence, it is reported that this
Page 14 of 259
arrangement is judged to not cause sufficient distortion of incentives to require a change to the basic business regime and resource allocation. Another aspect that is distinctively different from the Dolomiti SuperSki Regime is how planning is conducted. In Serfaus, the planning is a deliberate and conscious process that is conducted by a small committee of interested participants and includes the president of the tourist organisation, his second in command, the CEOs of the two lifting companies, and the mayors of the two communities. This committee plans the extent and overall direction of the improvements that, in their opinion, will have the best economic effect on the region. The formal organisation of the two regions under a single marketing banner only occurred in 2002, and since that time has, in the eyes of the participants, been a great success. There is a common appreciation of the effect of ski field size on the appeal of a region to visitors; a minimum overall slope run is required to provide sufficient variety and interest to skiers for a region to prosper. It was reported that there is a tendency across the whole European Alpine region for smaller areas to amalgamate to achieve this minimum slope run and that this has produced a more stable future for those regions that undertook this organisational change. It is apparent that smaller ski fields only survive or prosper if they are conveniently located next to a large city, e.g. near to Innsbruck, where there is good access to the field and other reasons to visit the city. There has not been any substantial overall growth in the number of skiers since around the 2007/8 period and maintaining numbers is considered a good outcome for any particular region. Serfaus is also different from Dolomiti SuperSki in how the snow-making and slope grooming is conducted. In the Dolomites, each lifting company or group associated with a particular area has direct responsibility for the slope preparation. In Serfaus, this function is conducted by the marketing organisation and this appears to work because, like the lifting companies, there is a common objective of having the most appealing slopes at the least cost. A final difference, of relevance to this study, is the manner of payment of the marketing organisation. This is not paid out of lift revenue (as with the 3% of turnover in the case of Dolomiti SuperSki) but is paid for by a direct levy on all businesses in the area and applies to all the activities of the marketing body for general tourism in the area, not just ski lifting. In the interview, it was emphasised that there is no risk taken by the marketing organisation for any development of lifting infrastructure: all risk, financing, etc. are being borne by the two commercial lifting companies in Serfaus which are owned by the local community organisations. 2.5.3 NZSki New Zealand The business regime of this skiing area in the Queenstown area of New Zealand functions with centralised ownership of all assets, including lifting, snow-making, ski schools, food, and catering, etc. This is a typical ‘Integrated’ resort common in North America and Australasia. It results in the capacity of the commercial entity to have overall control to operate in a monopolistic way within the defined area of the ski resort.
Page 15 of 259
The scale of operations was the main reason reported for retaining ownership of all assets, including transport. All aspects contribute to overall company performance. Assets are productively extended into as many areas as possible, e.g. for in-town equipment rentals at Queenstown, the remote pickup point for ski field visits, bus transport to the fields from Queenstown, etc. Planning of investment is very much on a directed, holistic basis. The principal question asked is: where is it best to spend the scarce investment dollar? Is this in snow-making, restaurants, ski school, access road, etc.? There are technical experts who advise on, or who have responsibility for, the various infrastructure options. They look across and beyond what each specific manager of any particular responsibility area does. The ultimate objective of all investment is to maximise customer satisfaction within the cost constraints; every decision is measured against its contribution to the overall return on equity invested. When describing the Community Franchise regime of the Dolomites, the CEO of NZSki commented that such an arrangement would be totally unacceptable to them because of the inability to centrally control all expenditure and to optimise spending between competing demands. It would, he thought, result in sub-optimisation of resource allocation, (at variance with the findings of Beritelli et al. (2007)). It was reported that planning is essentially by ‘counterfactuals’. What is the opposition up to? How much do I need to spend to match what they are capable of? What is the most productive way of matching what they are going to spend? What are the consequences of not doing something? Can our productive expenditure put pressure on the opposition whilst still making economic sense for our objectives? There are potential problems if other ski fields were to have uncontrolled or undisciplined capital spending on say lifting, (or any other specific aspect). Their market appeal must be matched in some way if NZSki is to maintain its market share. This would lead to an overall reduction in return for all participants if its spending is not as productive as required by normal commercial criteria. The incentive is for the NZSki owners to restrain capital spending and not engage in these ‘wars’ where there can be no clear winner among the competing ski fields. If excess spending was to occur, all the ski fields suffer. This was described as being similar to pricing in any business where each participant is restrained in fare reductions. There is a limit which if crossed would be in essence a race to the bottom. The small number of ski field participants provides a measure of stability in this regard whilst still maintaining a very competitive market. Pricing of tickets targets the longer-term commitment of skiers. A comparison of the value of one day pass to a four-day pass shows an attractive rate for longer periods which tries to lock in patronage to the longest time-period to capture that patronage to the particular field. Operating expenses tend to be fairly flat, running the management and support operations and rising only slightly in the middle of the season due to a build-up of certain classes of temporary on-slope staff. This contrasts with the income stream which only starts when winter is well underway, tends to peak with holiday periods, and then
Page 16 of 259
subsides quickly at the finish of the season. Revenue has to pay for the invested capital as well as the yearly operating expenses, so, while expenses are fairly flat, there is a tendency to prolong the opening and closing to meet every opportunity for patronage. Variations in income over seasons can be significant and are absorbed to a certain extent by the larger parent body which copes with continual fluctuations in anticipation of good overall paybacks. 2.5.4 Victorian Ski Fields, Australia There are five separate ski fields in the state of Victoria, and all are located within National Parks. However, the geographic area that they occupy is excised from those parks. The Victorian state government retains ownership of all ski fields which are administered by separate Alpine Resort Management Boards (ARMBs) which are appointed by and report to the responsible Government Minister. The ARMBs act as Crown land managers and in a similar fashion as a local government authority concerning the provision of the resorts’ ‘municipal services’. ARMBs are in place at Mt Buller/Sterling, Falls Creek, Mt Hotham, Lake Mountain and Mt Baw Baw. A special Alpine Resorts Co-ordinating Council (ARCC), which consists of the chairs of each ARMB and government-appointed independent board members, has responsibility for coordination, overall sector marketing, scientific research, strategic planning for the alpine areas. It also disseminates the information that it gathers for the ARMBs, the involved commercial bodies, the local communities, and other interested parties. The ARCC is funded by mandated contributions from each ARMB which in turn are funded principally by gate entries and lease fees. The resort operations and infrastructure in the two smallest alpine areas, Mt Baw Baw and Lake Mountain are fully owned by the Victorian Government. The ARMBs for these two resorts have the same responsibilities and obligations as the larger resorts, however, in both cases, all defined activities and services are contracted out to commercial operators. These two small resorts are not commercially viable and are subsidised to remain open and achieve wider government objectives. All alpine activities in the three largest alpine areas are conducted by private owners with those private owners who hold leases over the areas needed for their operations. There are no head leases. Each activity is separately leased by the ARMBs to each lift operator, accommodation provider, etc. The companies at the three largest alpine areas take all financial risk and reward for all activities that they engage in. There are no government subsidies for these three ARMBs. The private companies assume responsibility for all commercial activities associated with skiing as well as assuming responsibility for the attractiveness of the destination, i.e. snowmaking, slope grooming, etc. Other commercial activities in the location, not assumed by the resort companies, include further commercial accommodation, food, ski schools, etc. and are provided by other for-profit commercial operators in separate ARMB leases. It is up to each operator based on its assessment of commercial factors as to whether or not to run lifts in the summer months to cater for walkers and other summer mountain activities, and that decision is left to the commercial
Page 17 of 259
operators. The ARMBs also manage separate leases for all on-mountain, not-for-profit club ski lodge premises. All commercial operators pay a lease fee to the respective ARMBs for conducting their business and this provides, for the term of the lease, the security of tenure to invest in its improvement. It is reviewed periodically in light of prevailing commercial circumstances. The terms of leases vary and are renegotiated with the relevant ARMB in confidential negotiations. New developments in the alpine areas need approval from the Minister for Planning in the Victorian State Government, whose department consults with the affected ARMB (although the ARMB is not a referral authority on planning matters). The ARMB and the relevant minister are responsible for issuing any leases or lease amendments that would be required as a result of a successful planning application. This extends to all new developments including extensions to the boundaries of existing leased areas, new buildings, new or upgraded lifting devices, etc. The statutory boards (ARMBs) have no influence over the activities of the operating companies except those agreed through the lease terms and conditions. The actual planning and all business activity associated with activities in those three largest areas are undertaken by the commercial operator, including the main ski lifting, slope preparation, etc. At Mt Buller, this is an Australian domiciled for-profit company. At Falls Creek and Hotham, it is an international for-profit company, Merlin Entertainments Pty Ltd. This company indicates on its web page that it provides the following services1, including ski lifts, Snowsports Centre, (incorporating children’s and adults’ ski & snowboard lesson programs), various rental and retail businesses, Hotham Daycare, Hotham Airport, Hotham Holidays (resort reservation centre), White Spa and Onsen Spa and Retreat. There are no restrictions imposed on the principal lessees to operate in any particular way, other than the terms and conditions agreed to initially. Thus, the commercial objectives and capabilities of any lessee (and prevailing law) dictate how a field is developed. It appears that Merlin occupies as many profitable niches as are available and which suit its commercial objectives. Those which do not fit its objectives may be undertaken by other commercial operators, e.g. specialist accommodation and some food and retail operations. 2.6 Analysis of the four Case Studies As shown above, the interview process confirms that Dolomiti SuperSki is different in several key characteristics from those of other ski fields. Each relevant aspect of the operation is discussed separately in the following sections.
1 (http://www.mthotham.com.au/all-about-hotham/) accessed July 2016
Page 18 of 259
2.6.1 Overall Control The geography of all the ski fields is controlled in some way by a government or community body but no ski field operation is conducted by government agencies: in all cases that function is leased out or sold to other organisations, or in the case of the European resorts, the responsibility is delegated to commercial entities owned by the communities in which they operate. The four ski fields reviewed can be divided into those that have centralised commercial ownership and direction of the ski field operations, (NZSki), and those that have community organisations who control in different ways the manner of their operation, (Victorian Ski fields (government), Dolomiti SuperSki (Italian local valley marketing boards) and Serfaus (Austrian local community business groups)). The overall control of both Dolomiti SuperSki and Serfaus is provided by locally-based community organisations whose principal aim is to foster the development of the local areas from which they were created. All surpluses are generally reinvested in the local area which has the effect of benefiting the local population from whom the community organisations are drawn. The CEO of Dolomiti SuperSki was quite insistent on describing the type of business Regime over which he presides as a “Community Franchise”. The Victorian ski fields have some of the public ownership characteristics of the European model, but the “community” is at the state level without a solid base of the local population to influence and control the commercial operations. (There are board members who are drawn from the local population, but members are selected for board membership generally on their possessing a particular skill. The ‘local’ members provide some representation of the local community. Not all board members are drawn from this local population as in the European examples, and the board is responsible to the state of Victoria, not to the local region. The Victorian examples are the only ones where the central (State) Government has ownership of the ski fields. This was a consequence of historic circumstances and was not the intention of the government. It was simply the result of rising costs and the need to subsidise the provision of services in those two smaller alpine regions which the government judged desirable to continue. The lifting operations at these state-owned ski fields, however, are still contracted out to private operators. Ski fields in Australasia (generally) and the US are run by private ‘for profit’ organisations without necessarily any local ties. 2.6.2 Responsibility for the Transport (ski-lifting) Services The government, as the ultimate beneficial owner of the various ski fields and associated infrastructure, does not have any direct responsibility for the commercial activity of transporting people. This is in all cases leased to private companies who exercise their judgement as to investment in the mode and operation of the transport infrastructure. These investments can be significant and of a scale to match that put into urban public transport in modern cities. In the Australian and New Zealand cases, the involvement in transport extends beyond lift operations to ownership of railways (Perisher), airports
Page 19 of 259
(Hotham), and bus services (NZSki). In this sense these commercial operations are no different from, for example, mining companies that may own the railways associated with the transport of their product to ports, or from Japanese commuter railways and US freight companies owning the railways and making all investment decisions about these assets. 2.6.3 Line Ownership All ski lifting lines in the NZ and Australian examples are owned and operated by the companies that control the fields. In Serfaus, the ski lifting lines are owned by two separate companies who own all lifts in the two separate regions being serviced. These two companies are themselves owned by the marketing organisations which represent the two separate communities in the area. In Dolomiti SuperSki there is a great variety of ownership arrangements amongst the franchisee companies operating the lifting devices. A single company owns all lifts in four contiguous areas in one part of the overall ski area, and in others, a great number of smaller operators own various lifts without geographic concentrations of ownership. Some lifts are owned by individuals who run these as small family businesses and others are owned by superannuation trusts or private companies that operate them for the profits that they generate. No lifts are owned or operated by the franchisor, Dolomiti SuperSki. 2.6.4 Payment for Lifting Services A crucial difference between these four regimes is the method of payment for the transport services provided. Only Dolomiti SuperSki use a system where payment is solely based on the number of passengers carried: it pays the franchisee lift owners according to the number of persons who pass through their turnstiles; there is no other recompense. The numbers recorded at the turnstile are transmitted electronically to the Dolomiti SuperSki computer system where payments are calculated according to the published rate for the carriage of a passenger. This figure differs for the style of conveyance so that a gondola will attract a higher payment than a chair lift. These rates are the subject of negotiation between Dolomiti SuperSki and all the constituent marketing organisations and are reviewed on an annual basis. These figures are published and can be relied upon by any entrepreneur who may have an interest in providing a new service or improving an old one. The Serfaus regime is interesting in that it rewards the lift companies, not with the number of passengers carried, but with the time that skiers spend in the area of the lift, an indirect and possibly inaccurate measure. It was reported that this arrangement works despite the potential for misallocation of resources, possibly because of the local nature of the participants and the operation of pressures for fairness and equity between community members, and because the companies involved are of approximately equal size. The electronic ticketing with the commercial operations of NZSki and the Australian commercial operators are used to control entry to the lift, not as a means of payment, although that capability is available to management.
Page 20 of 259
All regimes utilise fee arrangements and do not charge on a per-trip basis or by distance. This seems a pervasive force in the ski field business that seeks to satisfy customers with a single fee for access to the whole field regardless of individual usage. The use of access fees is similar to many other network industries, e.g. telecommunications, TV, postal services, etc. and including some public transport pricing, e.g. the metro system of Mexico City. 2.6.5 Ski Field Planning Dolomiti SuperSki is distinguished from all the other regimes by how the planning and development of the ski field are managed. This is a specific result of being a franchised industry. At Dolomiti SuperSki there is no central planning. Dolomiti SuperSki does not get involved in whether a particular piece of infrastructure is required, or would be useful, or is necessary; that is left entirely to the market they have created for the movement of people. The principal role of Dolomiti SuperSki is to set the purchase price of lift access offered to skiers which then determines the price paid to the individual lift owners to carry one passenger. The payment to operators is the price signal that any private entrepreneur uses to determine if it is worthwhile to install a particular piece of infrastructure or to calculate whether upgrading an existing one is worthwhile. The ticket access price to individual skiers ensures that the ski fields for which Dolomiti SuperSki are responsible remain competitive in the overall international ski market. Supply and demand at the macro level determine the price that tickets can be offered to the public and this, in turn, determines the price that can be paid to each operator. In the other ski field business regimes reviewed, the price of carrying a skier on a particular line is only part of the many considerations that are taken into account for the allocation of resources for possible investment. This consideration may direct capital to snowmaking, ski schools or food catering, etc. as alternative directions for investment. In the case of NZSki, for example, over-investment in any particular area of a field is seen to be detrimental to all competing ski field operators as it represents capital that has to be both matched by competitors and serviced by the overall charges to skiers. This reduces the general rate of return to the whole industry. These considerations could be considered to have oligopolistic tendencies, but this issue is not pursued here. With a small number of industry participants, with not much-perceived loyalty to each other, there is the tendency, (as reported by NZSki), to make as little investment as is consistent with continued return on the capital already invested. The anecdotal observation of the situation in Australian ski fields would support this conclusion. Investment in new lifting infrastructure is infrequent and parsimonious but improvements will be matched by alternative ski destinations. Some infrastructure in Australian ski fields is so old as to attract heritage listing and does not match the standard expected by most visitors. As there are limited alternatives for skiers, there is no loss to any of the dominant operators, who watch each other’s moves closely and will not allow too great a gap in service offerings to build up but will rarely initiate any investment that is not necessary. This is a sufficient explanation of the lack of current
Page 21 of 259
investment in the Australian ski fields’ lifting equipment where all field operators generally have similar commercial business regimes: only the minimum investment is made so as not to lose customers to adjacent fields, all of which are under the same capital pressures. There is not a slope by slope or a valley by valley competition for individual passengers which drives the upgrades of equipment in the community controlled commercially operated ski lifting systems of those examined in Europe. 2.6.6 Associated Activities to Encourage Patronage The attractiveness of any particular ski area is maximised by the lift owners in the area in the European ‘community’ regimes to increase the chances that they will benefit commercially from current investment. It is interesting to note that despite the possibility of so doing, the lift operators in Dolomiti SuperSki are not generally involved in food catering, restaurants, or accommodation, hotels, etc. and seem to have gravitated to only those specific aspects that are essential to ensuring patronage of their areas, i.e. snowmaking and slope grooming. The other related functions are adequately catered for by other mainly independent providers. This effect is absent from the ”Integrated” business regimes, where considerations concerning the whole ski field are settled at the topmost level of management and where food, ski lessons, most hotel accommodation, and all ancillary activities are participated in by the ski field owner to maximise income from all available streams. In this regime, investment in any one area is compared to the return which that same investment may yield if directed to another area of operation, e.g. the competition for resources between, say, new lifts and food or ski hiring facilities. In the Dolomites investment by the line owners are contained to a smaller set of priorities. With franchisee lift owners, it is confined to only the lifting or improvements which attract people to the area in which the lifts are located, i.e. snowmaking and grooming. 2.7 The Essential Business Regime Differences 2.7.1 Generally Both the regimes of Dolomiti SuperSki and Serfaus are ones that qualify as “community” organisations identified by Patchell (2014): they are not ‘Integrated’ resorts as occurs in most of North America and Australasia. However, Dolomiti SuperSki is significantly different from community organisations as well as all the others in several significant ways. Most importantly this is in its governance structures, where the ownership of the lines, the manner of rewarding the individual line owners and how the planning and financing of new infrastructure are carried out. By the relevant criteria established from the interviews, the business regime of Dolomiti SuperSki is both a community enterprise and one that is owned and operated as a franchise. This franchise function is as it is described by the managing director of Dolomiti SuperSki as the franchisor: the business regime meets all the essential criteria for that description. This is demonstrated by examining each of the essential elements of franchises, summarised in Table 4 below:
Page 22 of 259
Criterion Response of Dolomiti SuperSki to Franchising Requirement
Branding The “brand” of Dolomiti SuperSki with its entire marketing and ticketing infrastructure is owned exclusively by the head franchisor and participation as a transport provider is only available to operators through that organization as a franchisee.
Standards and Franchisees must accept the payments and standards of Dolomiti Payments SuperSki to participate in the scheme and are only paid based on the number of people moved, identified by Dolomiti SuperSki as the relevant criterion for recompense.
Infrastructure Proposals for any infrastructure come from franchisees or Proposals potential franchisees, thus utilizing the local knowledge of franchisees.
Individual Operations of each individual line are controlled by the franchisee Operations who must meet the essential criteria set down by Dolomiti SuperSki but which, after that, can be organized to run its lines in the manner best suited to that individual franchisee.
Beneficiary of All savings and costs incurred in operational expenses remain Operational with the franchisee as the amount for remuneration for a Economies passenger is a set amount.
Funding of The capital resources of the franchisee are utilized by the Infrastructure franchisor with all funding of infrastructure and operating capital provided by the franchisee proponents.
Table 4 Response of Dolomiti SuperSki to Franchising Requirement It is also evident that the Serfus organisation is a community organisation but does not operate as a franchise. A summary of findings on several other measures that contrast this Community Franchise business regime with the operations of the other regimes covered is as follows in Table 5 over:
Page 23 of 259
Dolomiti SuperSki Others
Overall control A for-profit commercial Variety of controlling entities from entity owned by the government and private companies to local community community organisation (Serfaus). Operations By others By others.
Line ownership A variety of individual By centralised commercial entities. entities Payment For passengers carried For the provision of the service, or by only indirect means of usage (Serfaus). Planning No central planning or Top-down direction and planning in all direction cases. Other activities By others A variety of responses, from the controlling entity undertaking all activities, to allowing other entities to participate. Table 5 Summary of Case Study Findings These will now briefly be discussed regarding criteria relevant to commercial enterprise. 2.7.2 The Profit Motive The transport functions in all four examples are run by commercial companies that will only survive if able to make profitable investments. There is a tension between maintaining the price of vertical transport (i.e. lift tickets) as high as the market can bear and keeping it at levels that are attractive to visitors who have alternative destinations and activities available to them. This applies to all cases studied, with the possible exception of the smallest fields in Victoria where the Government chooses to keep these facilities open for other reasons. However, even in Victoria, the actual lifting is tendered out to private interests to run and the tender process will, if open and transparent, ensure that investment is economically efficient. 2.7.3 Ownership and Loyalties A noteworthy feature of Dolomiti SuperSki and also of Serfaus is that the ownership of the franchisor is by organisations and people who are local to the area. There are no studies that show that physical proximity to others ensures that decisions taken by one group will benefit all those in a particular community, even if all people involved are located in the same area. Intuitively, it would seem that there is no connection between proximity and interest alignment. Most political systems are designed to counter these types of human conflicts at any geographic scale. With this understanding, there is no reason per se to think that the locally owned Dolomiti SuperSki has an alignment of interests that is any different from that of any other local commercial organisation. However, a significant factor that differentiates the various regimes in the four case studies is the locality of the shareholding of the companies involved. In the case of Dolomiti SuperSki, the local business community-oriented marketing organisations are the owners of the business i.e. the shareholders. The shareholders not only benefit from a return on their holding in Dolomiti SuperSki, but also from the dispersed benefits that flow to all the other businesses dependent on tourism in the area. There is a diversity of
Page 24 of 259
companies representing the marketing organisation which owns the shares in Dolomiti SuperSki and this ensures that this diversity of views is represented when decisions are made that affect the whole community from which these business interests are drawn. This is like Serfaus where the marketing organisation owns the two lifting companies and again, decisions taken will be considered by those local people making up the ownership of the marketing organisation. This situation is fundamentally different from the case of SkiNZ or the Merlin group in the Victorian ski fields. There the shareholders are not necessarily from the local area. They are in most cases remote and have no binding interest in the local area or the various businesses that they use, or to the individuals, they employ in the areas of operation. Faced with decisions that may result in losses to themselves, it would be unrealistic to expect those commercial organisations to operate for the benefit of local communities ahead of their shareholders. This type of situation is common wherever large companies, for example, mining companies, have to trim investment in the locations where they hold their remote investments. There is no expectation that investment will be kept going for the sake of the local population. There is an acceptance that local interests are beyond the concern of large, and particularly, foreign-owned companies. The operations of Australasian ski fields are all leased to commercial organisations for a fee and run by companies who have offices in the area of their operations with consequent employment opportunities. However, they repatriate their profits to their shareholders who could be in many parts of the country or the world. It is not known if there is an element of taxpayer subsidy in the Victorian cases in the lease amounts paid for occupation, as details of these figures are confidential, but it could be assumed that whatever amounts are negotiated will ensure that the broader objectives of the Government are met by the commercial operations providing the services that are needed by the alpine regions. The two European ski fields, on the other hand, have locally sourced companies and individuals who own the companies involved which are run essentially for profit. The effect of having a large, remote and detached commercial firm with a crucial role to play in the provision of an essential local service, vital local employment and investment are easy to appreciate. Large corporations are not known for putting the interests of a local population ahead of those of shareholders and theory concerning economic self- interest would support the existence of this phenomenon. Many cities and regional areas that have become dependent on a small number of employer/investor corporations have suffered when these industries are closed down for economic reasons. Detroit (motor vehicles) in the US and Newcastle (steel) in Australia are examples where great economic disruption occurs when there are failures in the smooth economic functioning of any dominant industry. The reason for this is partly explained by the way that modern corporations function with those who are led by a CEO and a Board displaying a capacity for dispassionate decision making being the ones most likely to survive and prosper (Kets de Vries and Manfred F.R.).
Page 25 of 259
Of the examples surveyed here, only the European style local community organisation- based regimes with their locally based ownership of the controlling body has the overarching incentive to invest locally if there are difficult times in the industry as a whole. This is a characteristic shared with the Japanese community-style regimes. Commercial organisations, especially multinational organisations could not be expected to have this same commitment to local issues. The ownership by people in the local communities serves a local area in good times as well as bad, as the example of the Japanese community-run ski fields, previously mentioned by Patchell (2014) does. In that case, it actually encouraged the behaviour of local area organisations to find alternative ways of investing locally when their industry declined. It is also evident in the way that the “community” regimes have invested heavily in summer-based tourism to balance the winter skiing in addition to the encouragement of art and craft-based tourism. In summary, a key feature of those fields with community control is that the economic wellbeing of the local area is paramount to their decision-making process. 2.7.4 Consequences of Control Mechanisms The alternative behavioural response of successful integrated commercial operations is to spread the risk through having a collection of different operations under the one corporate umbrella. This enables firms to ride out temporary downturns in any particular industry or geographic area. This is shown in the case study responses of NZSki and also in the US-based Vale which owns ski fields on several continents. This spread of businesses, which potentially are subject to different economic cycles, smooths out the effects on the overall corporate performance. Merlin Entertainment which owns the right to operate two of the Victorian ski fields amongst many different geographically spread businesses have diversified both geographically and the type of businesses owned as a way of spreading risk arising from any of their component operations. The end effect of profit maximisation in the ‘Integrated’ ski fields is that it is not strategically beneficial to give up any portion of the operation to allow other commercial entities to profit from any activity. There appears no intention of giving up the monopoly control of any element of their business since all parts of the operation are deemed necessary for financial wellbeing or even survival. In contrast, in the “community” regimes there is the reliance only on the business in the local region. With this reliance is the imperative to allow the local population to share in the provision of the various services that together make up the ski experience for tourists. In ensuring that the local population benefit from the opportunities to engage in business activities, the organisers allow the benefits of participation to spread through the community which then increases the flow-on effects to the local population. 2.7.5 Alignment of Interests How each regime compares to each other is most starkly revealed when considering where each of the interests of the commercial entities conducting the business is likely to be placed.
Page 26 of 259
With a locally owned and broadly representative ownership structure in place and the monopoly rights to organise the transport service, it does not matter to a Community Franchise regime that the actual providers of the lines of transport are foreign-owned, local or anything in between: the franchisor represents and conducts the business in the interests of the local populations. The commercial organisation led by the franchisor is well placed to ensure the activities generated are of benefit to the owners of the business, i.e. the local population. In the case of the Serfaus in Austria, there is still a connection with the local community in that the owners of the two ski lifting companies are the local marketing organizations and so are controlled by local people who may take the interests of the local population into account. However, this is different from Dolomiti SuperSki where economic forces cause the alignment of all interests, not just the interests of the personnel involved. In Serfaus, there is an incentive for the ski lift operating companies to maximize their profits and reward the operators of the company before other businesses that have no involvement in ski lifting. Only proximity to the rest of the community may moderate this effect and make it different from a more remote shareholding. In the case of SkiNZ, there is an alignment of interest between patrons of the ski fields and the controlling commercial organization. This is caused by the company striving to provide an overall experience that meets patron satisfaction. The transport function costs are being balanced over other service costs, e.g. equipment hire, dining, etc. with cross- subsidies being applied to provide the maximum perceived beneficial return to patrons. The transport function is subservient to the overall function of the whole system and this is a feature that prevails in many forms of ‘Integrated’ business regimes. It is a relevant consideration of a Community Franchise regime that the transport function, being a separate commercial consideration, may cause a sub-optimization of resource allocation as the individual line owners compete, possibly needlessly, with each other to secure their own best interests. This is an issue taken up in a later chapter on theoretical considerations. In the case of the Victorian ski fields, there is this same alignment of interests as in SkiNZ in the commercial organization providing the overall package that, to their mind, best satisfies customer expectations. 2.7.6 Setting of Fares Several factors are operating to set fare levels, and thus the attractiveness of any ski field and the profitability of the companies involved in the commercial operations. All resort operators seek to keep fares as high as possible, but all are constrained by the realities of the market that ensure that they keep fares sufficiently low to be able to compete with the other ski fields in their market. This situation has slightly different implications for the Community Franchise regime compared to the other regimes. All regimes other than the Community Franchise regime of Dolomiti SuperSki, have the luxury of being able to internally subsidise fares by shifting costs within the many aspects of their overall business. This is not the case with a franchised ski lifting operation.
Page 27 of 259
In the case of Dolomiti SuperSki, the franchisor, representing the various tourism bodies of each valley community, has a clear desire to maximize the number of tourists who come to the area and use its facilities. To the extent that a visit to the area is influenced by the cost of a visit, Dolomiti SuperSki, as for all other resort operators, has an incentive to keep the price of vertical transport as low as possible. However, it is in Dolomiti SuperSki’s interests to keep the prices as high as possible to maximise the income for both the franchisees, which encourages investment and for the income of the franchisor, which is a set percentage of overall turnover. Dolomiti SuperSki will, therefore, set the fare price at levels that balances these two divergent influences. Each franchisee lift operator will accommodate that level of its income determined by the recompense per passenger with the technology they can afford and that minimises their running and capital costs. The forces of economics between the three levels involved in a Community Franchise regime, the franchisor, the franchisees, and the paying customers, will tend to keep the fare levels that are acceptable to all three parties. There is a distinct pressure on the cost of service provision in these circumstances, as there are no cross-subsidies, and the franchise regime has a structure that allows that pressure to be firmly placed on franchisees, who to survive must operate efficiently. Fare levels will not be distorted by considerations of cross-subsidies. The possibility of cross-subsidies creates uncertainty about the true profitability of the ski lifting operations in an ‘integrated’ resort. 2.7.7 Lift Numbers per Regime In absolute terms, Dolomiti SuperSki has many more lifts than other ski fields in the European market but especially so from the different markets of Australia and North America. A review of lift numbers and areas of ski fields in the following table 6 shows these differences: Resorts Europe No of Lifts /Run (without DSS) Length of Run Km No of Lifts Length Les 3 Vallees, France 600 140 0.23 Portes du Soleil, France 580 172 0.30 Arlberg ski area, Austria 306 88 0.29 SkiWelt Wilder Kaiser- Brixental 284 90 0.32 Le Grand Massif 265 63 0.24 Mageve/Saint-Gervais 263 80 0.30 La Plagne (Parakiski) 225 90 0.40 Tigners/Val d'lsere 300 82 0.27 St Moritz 155 24 0.15 Aletch Arena 104 36 0.35 Ischgl/Samnaun 239 41 0.17 Arosa Lenzerheide 225 43 0.19 Col Rodella 55 36 0.65 Falcade 54 15 0.28
Page 28 of 259
Alpe Lusia 27 8 0.30 catinaccio 14 6 0.43
Average 231.00 63 0.30
No of Lifts /Run Resorts North America Length of Run Km No of Lifts Length Park City 250 38 0.15 Big Ski Resort 250 28 0.11 Snowmass 237 17 0.07 Vail 234 25 0.11 Steamboat 165 17 0.10 Breckenridge 153 23 0.15 Beaver Creek 150 16 0.11 Winter Park Resort 143 22 0.15 Key Stone 135 14 0.10 Aspen Highlands 135 5 0.04 Powder Mountain 135 9 0.07 Copper Mountain 126 18 0.14 Crested Butte 121 12 0.10 Sugarloaf 119 13 0.11 Alta 116 9 0.08 Jackson Hole 116 12 0.10
Average 161.56 17 0.11
No of Lifts /Run Dolomiti SuperSki Italy Length of Run Km No of Lifts Length Cortina d'Ampezzo 120 36 0.30 Kronplatz 119 32 0.27 Alta Badia 63 21 0.33 Val Gardena 175 79 0.45 Alpe di Siusi Val de Fassa 110 57 0.52 Carezza Arabba 44 28 0.64 Marmolade Three peaks Dolomites 110 32 0.29 Val di Fiemme 112 45 0.40 Obereggen San Martino di Castrozza /Rolle Pass 60 21 0.35 Gitschberg Jochtal- Brixen 100 25 0.25 Alpe Lusia - San Peliegrino 100 23 0.23 Civetta 72 22 0.31
Page 29 of 259
Dolomiti SuperSki Italy2 1185 421 0.36
No of Lifts /Run Length of Run Km No of Lifts Length Average (Resorts Europe (without DSS)) 231 63 0.304 Average (Resorts North America) 162 17 0.106 Average (Dolomiti SuperSki Italy) 1185 421 0.355 Table 6 Lift Numbers per Ski Run Length There is a vast difference in the Dolomiti SuperSki lift numbers per run length compared to those from ski fields in North America and Australia (approximately 350% more). In contrast, there is only a marginal difference in the Dolomiti numbers (approximately 16% more) compared to those other ski fields in that European market. This comparison is shown in the following Table 7:
No of Lifts /Run Length 0.400 0.355 0.350 0.304 0.300 0.250 0.200 0.150 0.106 0.100 0.050 No of Lifts /Run Length 0.000 Average (Resorts Average (Resorts Average (Dolomiti Europe (without North America) SuperSki Italy) DSS))
Table 7 No of Lifts/Run Length, Comparison of European, US and Dolomiti SuperSki fields A plausible explanation of this could be that the ability of franchisees to suggest routes under the Community Franchise regime, means that more routes are established in that Italian ski market than when the initiation of new lines is the preserve of the centralised decision-making lift owner. This abundance of lifts in the Dolomites must be countered by other ski fields in the same market so that those other European fields remain attractive. This would not hold for fields in distinctly different markets like those on other continents. Consequently, fields in North America and Australasia would not have to match the number and
2 All figures were obtained from https://www.dolomitisuperski.com/en/Live-info/Open-lifts 9th January 2019
Page 30 of 259
coverage of lifts that have been developed in Europe. The SkiNZ response suggests that retaining control over capital expenditure on lifting is too attractive as a lever over their corporate profitability to want to relinquish this and that the effectiveness, or limitations, of the resulting network, is something that patrons accept. It was not able to be determined if this patron acceptance means that there is no discernible difference between the number of lifts per run length. However, the reality of ski fields in Europe matching those numbers of the Dolomites suggests that if exposed to market forces then patrons will choose the greater number of lifts per run length and that Dolomiti SuperSki has led the market in this metric and forced others to match their example. 2.8 Summary of Regime Types This case study investigation shows that Dolomiti SuperSki has a regime with a combination of features that makes its business organisation different from other “community” models and also ‘integrated’ models. This case study investigation has described three different business regimes:
• A governance structure of Dolomiti SuperSki that is essentially a franchise regime where the beneficial owners and controllers of the franchised transport network are local and community-based and the transport function, as evidenced by the individual lines of transport, is planned, financed and undertaken by independent franchisees. • A governance structure for the Serfaus organization of a commercial operation fully owned the local marketing organization and local businesspeople. • A governance structure of the ‘integrated’ providers, NZSki and the Victorian ski fields in Australia, which comprises separate commercial operations owned by companies with remote shareholding unrelated to the local area and transport as an integral element of their business that can be cross-subsidized to or by other commercial activities. These different business regimes have different operational consequences which are summarised below. 2.8.1 Community Franchise Regime The Community Franchise regime structure of Dolomiti SuperSki promotes an alignment of interests between the overall franchisor, the travelling public, and the community in which the lines of transport are based. It provides a direct link between passengers carried and the livelihood of the entities that are responsible for the lines of transport. There is appropriate market signalling to each franchisee through its income based on passenger numbers. Consumer demand and costs of service provision drive the adjustment of services to maintain or increase passenger numbers. The Community Franchise regime structure also establishes the property rights of the commercial enterprises involved, enabling the owners to invest in what they regard as the most appropriate infrastructure to meet or increase the demand. Each owner feels the effects, both positive and negative, of all investment decisions made.
Page 31 of 259
This independence of the separate businesses creates the ability for the system to evolve into a more efficient one through commercial pressures. This characteristic establishes a mechanism whereby all proposals for new or upgraded lines of transport arise from the entrepreneurial line developers without the need for direct top-down involvement in infrastructure planning, choice, and investment. This appears to encourage a proliferation in suggested and actual lines of transport. 2.8.2 Conventional “Community” Regime The Serfaus “Community” model structure is different from the Dolomiti SuperSki structure: it is not a franchise model. In the Serfaus business model, conventional business enterprises undertake the business. A distinguishing feature is the local ownership of both the controlling organisation (the local marketing body) and the two lift operators with both of these commercial organizations having obligations for profit maximization. In the case of the marketing organisation, it has obligations to the local communities, just as in the Dolomites or Japan. However, the lift operating companies have obligations to their shareholders that may be sourced from the local area, but like any private business may only be a small subset of that community. In this arrangement, transport is a cost that will be most profitable if minimised. There is an incentive to minimise costs but not of maximising passengers, as payment is related to where skiers spend their time, not on their use of the lifts. There are close ties between the marketing company and the lift companies, where overall planning of the whole ski field infrastructure is discussed and decisions are made, but this is a different relationship from that which exists in Dolomiti SuperSki. In Serfaus, senior management reported that they rely on trust and shared interests. It does not have the transparency of the Dolomiti SuperSki regime where pricing signals are public knowledge and open for any person to take advantage of. In Serfaus, the pricing of the ski lifting service is a matter controlled by the marketing company, and, because of its dominant position, it has monopoly control in its geographic area. Pricing is however strongly affected by the location within a region of intense local competition. 2.8.3 ‘Integrated’ Model The ‘Integrated’ model of NZSki is a phenomenon common to many ski fields and is a system whereby central planning of the whole enterprise, including lifting is possible. In this arrangement transport is a cost that must be minimised; it is just an expense that will receive minimal investment consistent with maintaining consumer acceptance. There is perceived value in being able to create cross-subsidies to invest in say equipment hire or food over improvements to ski lifting. Investment is made where it is considered most profitable to invest scarce capital. It was emphasised in the interview with the management of NZSki that it would never voluntarily give up this flexibility and control over the fine-tuning of the economic performance of the whole enterprise. For any particular ski field, the ‘integrated’ model has monopoly rights and the monopoly operator’s business interests extend from the actual ski field into the surrounding towns where related businesses, e.g. ski hire, can be integrated into the overall operations at a
Page 32 of 259
considerable advantage to both consumers and the company. This monopoly right can be secured by several means outside the scope of this study. In the case of the ‘Integrated’ regime, there is no inherent alignment of interests between the operating company and the local community unless the company’s remote shareholding happens to include members of the local community. The operating company is interested only in the maximisation of profit from all of its comprehensive separate businesses from accommodation to catering.3 2.9 Summary of Case Study 2.9.1 Regime Differences There is little doubt that the way the ski fields in the Dolomites is organised by Dolomiti SuperSki is unusual. The business methods employed in the Dolomites are known to other ski field operators in the European area but are not emulated. Even those other community ski fields, e.g. Serface in Austria, do not use the franchise system but maintain their centrally directed methods of asset allocation and operational control of ski lifts. The Community Franchise regime is different from the business methods used in the ski fields in Japan where although community organisations of ski fields in common, a franchise arrangement of ski lifting do not seem to have been tried. And in North America and Australasia, a franchise system is not only not used but was specifically eschewed as a viable option by those in control of integrated fields when considered, as it would deprive that operator of significant options for management of overall activities, e.g. between allocating capital to ski lifting or accommodation or food activities, and the many other permutations of various possibilities open to a central controller. It can be concluded the business regime of Dolomiti SuperSki, is somewhat different from both the other ‘community’ operated ski fields and the ‘integrated’ ski resort business regimes. 2.9.2 Previous Study The Community Franchise regime of Dolomiti SuperSki does not appear to be a business regime that has been studied before and a clear research gap has been addressed by this study. From the experience of having this case study reviewed at the 2015 Thredbo Conference on Ownership and Competition in Land Based Public Transport held in Chile, it is likely this business regime is unique and not previously reported on in the literature. This series of case studies did not reveal any similar business regimes operating in ski fields either directly by contact with the four case studies or indirectly indicated by discussions with the respondents from these four case studies. The business regime that it embodies has characteristics that make it interesting to study, regardless of its uniqueness.
3 Experience with Australian ski fields would suggest anecdotally that this is a characteristic of monopoly or duopoly situations, with ski lifting being decidedly less sophisticated and convenient in NSW ski areas with only two competing fields than that experienced in say Europe where competition is comparatively fierce and consumer choice is vast.
Page 33 of 259
As a regime, it displays an alignment of interests of those operating the ski lifting transport system that is different from the business regimes of other ski lifting operators in both the other community and in the integrated resorts. This elevates the importance of attracting and moving passengers to franchisee lift owners as this is the only way that they can prosper. This contrasts with the resort owner in the role of transport provider and employees in those roles where the incentive of those operators is to undertake the specified task and is not directly dependent on the throughput of passengers as franchisees are. 2.9.3 Regime Peculiarities A significant difference in this regime operation is the capacity of entrepreneurs to suggest new lines of transport, a task specifically eschewed by Dolomiti SuperSki, but one retained by all centrally planned networks. One of the most interesting features of the Community Franchise regime revealed by the case study is the way that it does not engage in any deliberate planning of the transport network. Instead, this is left entirely up to the widespread collection of entrepreneurial franchisees to propose and finance all development. Dolomiti SuperSki has a generally passive role in this regard and it has proven to be a very effective mechanism for the bottom-up evolution of the overall network of lines. It does, however, influence interest in new and improved lines by the level of recompense offered to franchisees a measure over which the franchisor and its owners have implicit control. This feature of network planning from below is a characteristic that sets this type of franchising apart from other forms of franchising where that term covers franchising in public transport as practised in Sydney and the UK. None of the other ski fields examined has this process of outside parties initiating the planning of new infrastructure and it is an important contributory explanation for the size and growth of the Dolomites ski area. Dolomiti SuperSki has operated for a considerable period, since 1974, has grown to be the largest ski field by lift numbers in the world and the individual ski lifting businesses are commercially successful (almost by definition given its franchise nature). The capacity for independent operators to suggest new routes and to participate by becoming franchisees under the umbrella of the overall field organiser/owner are mechanisms for achieving growth in the number and location of new lines that are not shared by any of the other regimes examined. It is noted that this passive role of the Community Franchise organisation is radically different from the role of owners of integrated ski fields and as well as government agencies in urban transport network planning. In those latter organisations, a controlling role is assumed by the central planner of the system and in the case of cities, by government departments and politicians in the selection and justification of network improvements. Generally, urban infrastructure improvements are made based on cost/benefit studies which provide direction for political decisions. This aspect will be returned to later in this study when political influences on asset allocation will be briefly explored.
Page 34 of 259
The continuing operations enable the commercial franchisor, Dolomiti SuperSki to pay dividends to the owning valley communities. The network is continually evolving in the development of existing lines and the introduction of new lines. The process, by which the network of individual ski lifts evolves by gradual profit-motivated improvements over time, distinguishes its operations from the ‘integrated’ regimes and supports the claim that it is commercially responsive. This is also displayed by the efforts put into local maintenance of the slopes and snow by the lift owners to ensure an attractive destination. This bottom-up evolution process is a complete reversal of the top-down directed planning of the other ski fields covered. The efficacy of this approach is a feature amenable to testing by controlled experimentation and the way that this has been accomplished will be described in the main body of this study. 2.9.4 Prospects for Investigation It could be argued that the reasons for the commercial responsiveness are not necessarily obvious as the existence and growth of Dolomiti SuperSki may be due to factors other than the business regime, e.g. politics, location, the popularity of the sport, etc. This prospect is addressed in the second and third sections of this thesis dealing with experimental evidence and the review of an expert panel where deeper explanations of the observed behaviour are sought. There are many aspects of the Dolomiti SuperSki transport operations that might have been useful or interesting to more fully pursue, such as a more detailed history of its development. This was not possible due to personnel changes at headquarters and a reluctance to reengage with researchers. But based on the documented evidence obtained in the interview and correspondence with Dolomiti SuperSki, the establishment that this business arrangement appears to be unique is the most important outcome and is by itself sufficient for the case study and this thesis. In the next chapter, a Community Franchise is defined from the observations of the observed characteristics of the case study subject, Dolomiti SuperSki.
Page 35 of 259
Page 36 of 259
CHAPTER 3 DEFINING A COMMUNITY FRANCHISE REGIME 3.1 Chapter Content In this chapter, the observations of the workings of Dolomiti SuperSki seen through the case study will be used to define the Community Franchise business regime. This will extract the essential characteristics of the regime of Dolomiti SuperSki, for later application in an urban public transport context. A definition of a Community Franchise regime does not exist in the literature and is a construct of this study. The definition will use a descriptive algorithm that will follow this analysis. Several aspects that are fundamental to the Community Franchise regime concept have emerged that can be considered vital: 1. That the exclusive right to provide the service is retained by government and administered by their franchisor organization. 2. That all shares in the commercial entity to act as franchisor and run the public transport are owned by the government of the geographic area in which it operates. 3. That recompense to franchisees is only for passenger movements. This is a means of tying the perceived interests of the provider to that of the organizer. 3.2 Principles of a Community Franchise Business Regime 3.2.1 Community Franchisor Owners A Community Franchise regime is created and controlled by a subset of the civilian society that represents a “community”, whether this is the wider political unit of the state or a more local citizen grouping of, say, a municipality or city. It is also possible for the owner to be industry/community organisations such as cooperatives, chambers of commerce, clubs, professional groups, etc. To be considered as a Community Franchise regime the commercial entity must be owned by a “reciprocal” organisation as defined by Frijters and Foster (2013), i.e. it is controlled by and exists to serve the members of the particular group. It will respond to group emotions, such as attachment, and act in ways that may not be influenced exclusively by the material reward to members. As a reciprocal organisation, the franchisor in the Community Franchise regime, such as Dolomiti SuperSki, is an entity with the community as its ultimate beneficiary. In contrast, it should not be owned by one defined by (Frijters and Foster, 2013) as a ‘hierarchical’ organisation, being one such as a commercial for-profit company owned by private interests and acting in the interests of its shareholders. The board of directors of such a company has a prime responsibility to maximise profit for these shareholders. Another point of difference is that shareholders do not necessarily have ties to the local area in which the company operates, i.e. do not represent a “community” as such. 3.2.2 Community Franchisor Corporate Structure The franchisor of the Community Franchise regime is a ‘for-profit’ corporation formed by and fully owned by the community organisations of the Dolomites; i.e. by a government representing a geographic constituency. This makes it similar in the Australian context to
Page 37 of 259
Australia Post, or the Australian Rail Freight Corporation. All shares in both these commercial corporations are completely owned by one or more governments (Federal, and State). A key difference with a Community Franchise regime is that the for-profit corporation operates as a franchisor, not a direct operator of the business. The franchisor company has a board of directors who have relevant experience and provide clearly defined objectives for management to follow. Being a commercial corporation means that the chief operating officer, reporting to the board, has a clear understanding of what the board’s objectives are and is hired and paid based on achieving those objectives. As well, it will be normal practice for CEOs of these commercial organisations to be suitably incentivized by having their remuneration related to the achievement of those objectives. The benefits related to reaching those goals will vest only in the long term. The CEO’s interests will thus be aligned with the interests of the board. It can be expected that the CEO will have as a prime objective the need to run an efficient profitable operation but will have to take on board whatever social objectives may be adopted by the board in consideration of the obligations of ownership. 3.2.3 Community Franchisor Share Ownership The structure of the franchisor, Dolomiti SuperSki, is that of a commercial company, like any other franchisor in the commercial world. The share ownership of this commercial entity by government ensures that there is an alignment of interest between the rule setting franchisor, and the public which it serves. Ownership will vary with the operation being franchised and ultimately to which group of citizens the Community Franchise regime serves. In the Dolomites, the Community Franchise regime franchisor is owned by the marketing organisations that represent the twelve valley communities in which the ski fields are located. Keeping ownership of the franchisor to populations dependent on the service to be provided ensures that there is an alignment of interests in the goals, activities, and budgeting between the franchisor and the local shareholding population, considerations that would be sensitive to a changing political composition of the board. This sensitivity of the board and government ownership is the subject of the (Castro and Maddock, 1997) analysis of Australia Post. It was concluded that in that instance at least, the board is very sensitive to the feelings of the Australian Parliament, to the extent that charges for common postal services were maintained at levels considered by politicians to be fair and reasonable, and significantly less than that required to operate profitably. It would be interesting to observe if similar conditions if they prevailed in the Dolomites, would show a sensitivity of the board to political considerations of their owners, and of other situations where government has control of commercial operations. 3.2.4 Community Franchise Board of Directors A board of directors represents the community that owns the shares in this franchisor. As such, knowing the objectives of the community is fundamental to being able to discharge the directors’ duties, and directors could be expected to be held accountable to the wishes of the owners, just as is expected in any corporation with shareholders. The board includes a representation of experienced trusted people, many from the population
Page 38 of 259
of the community it represents, augmented by people with special skills associated with the particular line of business and with business in general. This is no different from any commercial organisation. The composition of a franchisor board could be expected to take its community responsibilities seriously and thereby ensure an alignment of interests between the community it serves and the company that operates the business. 3.2.5 Community Franchise Regime Objectives The most important aspect of service provision is to understand the key defining measure of success. As is illustrated by the Dolomites example, when it comes to a transport function, the critical measure is the movement of people, not the provision of a service. Framed in this way, this objective dictates that franchisees are only paid if they actually carry passengers. This payment is the measure by which all activity is assessed by franchisees and rewarded by the franchisor. A complementary defining characteristic of a Community Franchise regime, as evident from Dolomiti SuperSki, is that the franchisor does not carry out the function itself, but only organises a system of rules, payments, incentives, etc. that achieves the aims of the shareholders. The objectives of the franchisor are therefore achieved by franchisees. 3.2.6 Community Franchise Regime System/Network Planning Central planning of the network or of individual operations by Dolomiti SuperSki has not been needed in the Dolomites. It was observed that at least three conditions are operating and that appear to be required for bottom-up planning as it affects resource allocation in the ski fields: 1. the opportunity for potential operators to participate in the suggestion of lines, 2. the incentive for them to do so, and 3. the security of asset tenure. Without these conditions, existing regimes require a government’s intervention and oversight for the planning and implementation of transport functions to occur. This contrasts with the existing Community Franchise regime in the Dolomites, where the planning of the network is undertaken by the existing and potential franchisees, which is also the level where all risk and reward is located. This is noted to be a different use of the term ‘franchising’ as it is used in such networks of public transport as Sydney. 3.2.7 Community Franchise Regime Rules of Operation The franchisor gives direction to franchisees on matters that are of importance to achieving the franchisor's objectives, but generally not on how to achieve them. The franchisor has a set of criteria for operations that exist to control the franchisees. The franchisees are allowed to achieve the objectives of the franchisor within rules of operation by the means that the franchisees see are appropriate for their own line operating organisation. The rules give clear direction as to what the objectives of the franchisor are including the hours of operation, the frequency of service, safety matters, and the amounts to be paid per passenger carried. The definition must document this vital feature.
Page 39 of 259
These identified characteristics of the Dolomiti SuperSki operations will now be applied in detail through the mechanism of an algorithm. 3.3 The Community Franchise Regime Algorithm The descriptive algorithm will comprise a numbered list of essential steps that must apply for any regime to be considered a Community Franchise regime; it is the definition to be used in this thesis. Using this mechanism, the regime can be defined as shown in table 8 below. In this table, the critical features are identified and how it is applied in the original example of Dolomiti SuperSki in the Dolomites is shown. This process enables the algorithm to emerge from the workings of the existing regime.
Factor of Dolomiti SuperSki Situation Resulting Algorithm Regime Interest Step 1 Overall Lift ski customers to elevations to Identify the objective of Objective enable skiing down established government policy. ski runs. 2 Geography The twelve constituent valley Establish the geographic area communities in which the ski that is most affected by the fields are located. activities to be undertaken 3 Identify the Electronic detection of a ticket Establish the means of measure of sold by the franchisor, (Dolomiti measuring the attainment of this success SuperSki) through a turnstile to objective each franchisee lifting device 4 Standards of Commencement and duration of Set standards to apply in Operation service, headway between achieving the identified services and safety standards. objective 5 Value to Value determined by market (cost Estimate the effective value to Community of ticket times sales volume) and the government of meeting the related to the flow-on effects of objective visitors to the ski region. 6 Nature of Dolomiti SuperSki, a commercial Establish a commercially Organisation to for-profit company owned by the responsive body, the franchisor, represent the marketing bodies of the 12 valley keeping this body in Community communities. This body has the government ownership and responsibility to act as franchisor making this body responsible for the lifting operations. for meeting government objectives. 7 Limitations of Dolomiti SuperSki does not Deny the franchisor the Franchisor undertake any lifting operations opportunity to actually itself. undertake activities to achieve the objectives itself.
Page 40 of 259
8 Measurement The electronic detection of the Require the franchisor to of objective tickets sold by Dolomiti SuperSki measure the attainment of the achievement at each franchisee facility. stated objectives such that these measures can be used for gauging franchisee activities. 9 Funding of The franchisor, Dolomiti Reward the franchisor based on Franchisor SuperSki, retains 3% of turnover. their achievement of policy Management team incentivized to objectives through the retention achieve the goal of successfully of a percentage of revenues growing revenue which is directly generated and appropriate related to passenger numbers. employment conditions. 10 Attraction of Published rates of recompense Have the franchisor publicize Franchisees for the movement of skiers which the objectives, the standards, varies with the mode employed and the rate at which it is and the height and length of determined to pay for the travel. objectives to be met. 11 No direction to any franchisee or Give scope to potential Encouragement potential franchisee on how to franchisees in how the policy of Innovation achieve the movement of people. objectives are achieved. 12 Assessment Undertaken by Dolomiti SuperSki Review and assess the of Franchisee when proposals to build or alter proposals for franchisee proposals lines of transport are made by schemes to meet the objectives existing or prospective and determine if they suit franchisees. overall government policy. 13 Acceptance All applicants for franchisee Identify and engage the suitably of Franchisee operations are assessed for the qualified proponents, i.e. the proposals suitability for the task, the franchisees, who will be paid location of the facility, the the agreed amount by the contribution to the network and Government to each franchisee villages in the area, and any in proportion to the success with other criteria seen to be relevant which each franchisee meets to Dolomiti SuperSki. the measurable objectives. 14 Ownership The line is fully owned by the Provide property rights to the of a line of franchisee with full property rights essential elements of any transport which are recognized by banks franchise to encourage and other property owners. investment in the process, long term commitment, and the capacity to borrow against the inherent value.
Page 41 of 259
15 Long term All franchisees must survive Allow for normal commercial franchisee economically by their own efforts. pressures to regulate the selection There are no subsidies provided activities of franchisees to any individual or group of line owners. 16 Achieving Review is undertaken, done Constantly review both the best resource regularly and any alteration to value and the payment to allocation existing rates for the movement franchisees for the attainment of of passengers is published policy objectives to establish the periodically for any interested rates most likely to achieve person to view. them. 17 Long term Continually undertake the task of Continue to review any new system viability franchisee selection and overall proposals for policy attainment viability.
Table 8 Application of Dolomiti SuperSki Community Franchise regime workings, and the generation of Descriptive Algorithm. This descriptive algorithm identifies the key characteristics of the workings of a Community Franchise regime from how those characteristics have been observed to work in Dolomiti SuperSki in the Dolomites. 3.4 Summary In this section, an algorithm that defines a Community Franchise regime has been proposed based on the operations of the only known operational Community Franchise regime, Dolomiti SuperSki in Italy. This provides a clear view of the workings of this business regime of public transport with which to compare those that will be found and considered in an urban setting. In the next chapter, the literature as it affects franchising as a regime of business operations, transport business regimes more generally and the specific issues important to transport operations, including monopoly control and planning, will be reviewed.
Page 42 of 259
CHAPTER 4 LITERATURE REVIEW 4.1 Chapter Content The objectives of this chapter are several. The first is to examine franchising as a business regime, this being a fundamental consideration of this thesis as well as appearing to be the basis of the observed differences between the characteristics of the franchised regime of the Dolomites and other ski resorts operating under conventional business models, a factor suggested by comments of the respondents in the case study. There is also the need to clarify the meaning of the term franchising as it is applied in different transport contexts. The second is to establish a framework for examining the research questions of this thesis as to the potential for a franchise, in a similar form to that of the Dolomites, to facilitate an effective public transport regime. Following this review of franchising, the status of the various business regimes in public transport generally will be reviewed. This review will use the experience of the 28 years of the Thredbo Conference Series of Competition and Ownership in Land Based Passenger Transport that has considered questions of regime function as its’ fundamental raison d’etre. This conference series has an ongoing documentary focus of the systems that constitute the ways that the public transport function can be organised: i.e. the various business models or regimes of urban transport. This focus of the Thredbo Conference series establishes it as a relevant authority in these matters. This review will focus on the extent to which any form of Community Franchising as identified and defined in the previous chapters has been applied to urban public transport. Coverage will extend to the business models of conventional urban public transport, how this has developed over the later part of the 20th Century, what the drivers of those changes were and what the results of those changes have been reported to be. Several topics will then be covered that are identified as important to the consideration of a Community Franchise regime. These are demand management, the related subject of subsidies, and then monopoly control and regulatory capture. Following this, a discussion of network planning, a distinguishing feature of the Community Franchise regime, is discussed for the implications it has both for current operations and for a comparison of planning in conventional and franchised regimes. Finally, the literature review will identify the relevant gaps and thus the opportunities for the research of this thesis. This review draws on a paper presented at the Thredbo Conference of Ownership and Competition in Land Passenger Transport of 2015 (Emerson et al., 2015b) but is altered for this thesis setting. It responds to feedback received at that conference and has been updated and extended to cover and report on matters relating to different public transport business regimes and what has been discovered in franchising.
Page 43 of 259
4.2 Franchising 4.2.1 Generally There are many meanings ascribed to the word “franchising”. In undergraduate economic text-books franchising is described as being the act of a government providing to a single operator the exclusive right to provide a particular service. The government may do so because, in particular circumstances, it could feel the need to allow a single operator the full economic benefit to achieve economies of scale, or possibly because the industry is considered a natural monopoly, e.g. the power or telecommunications industry. (Layton et al., 2011). This narrow view of the concept fails to convey the full depth of what franchising is in a modern economy covering just this single case of a government-granted monopoly. The Business Directory4 defines a franchise as being: “Arrangement where one party (the franchiser) grants another party (the franchisee) the right to use its trademark or trade-name as well as certain business systems and processes, to produce and market a good or service according to certain specifications.” This definition provides the essential feature of traditional franchising which embodies a level of independent operation of the franchisee within constraints imposed by the franchisor. It is the definition relied upon in this study and in particular, it identifies the different arrangements of controls that differentiate the many types of franchising and that of the Community Franchise defined previously. The Franchise Council of Australia Limited 5(FCA) acts as “the peak body for the $146 billion franchise sector in Australia, representing franchisees, franchisors…” Its membership is open to the many recognised franchised organisations, e.g. McDonald’s, 7 Eleven, Jim’s Group, etc. and the national turnover figure quoted indicates that franchising has a large economic footprint in the Australian economy, as it does in most advanced economies. Franchising is not independent commerce as practised by normal businesses, e.g. cafes, manufacturing and the like, and is distinguished in that the franchisee accepts the restrictions placed on it by the franchisor in return for the right to use the brand, methods and other proprietary business methods of the franchisor. The franchisee normally pays for the right to be in that commercial structure and has rights over the business defined in the franchise agreement. The rights and obligations that are purchased are a saleable asset against which banks will lend capital. 4.2.2 Franchising Operational Differences In more considered academic literature, franchising is distinguished by its operational peculiarities. Franchising is noted to be one of those intermediate methods of commerce that stands between intrafirm and interfirm transactions (Rubin, 1978). In intrafirm transactions, decisions are taken by fiat, and between firms, the decisions are taken by
4 http://www.businessdictionary.com/definition/franchising.html accessed March 2019 5 https://www.franchise.org.au/ accessed March 2019
Page 44 of 259
reference to a market. Franchising is one of the few ways of doing business that sits between these two methods. Rubin notes that there are other examples, such as sharecropping where the owner of the land allows other individuals or companies to utilise their knowledge and enterprise with an agreed share of the profit for each party. A franchising relationship occurs when the owner of a trademark, the franchisor, exchanges the right of use of that trademark with other agents, the franchisees, in return for some consideration (Minkler, 1992). Franchising is not, in his view, a typical contractual arrangement. The way that it works is well understood to be different from sub-contracting. The regime has quite particular and different characteristics for both the franchisor and the franchisee. For the franchisor, Minkler establishes that a sufficient explanation of franchising includes principally three specific reasons for its existence;
• that of the capacity for the trademark owner to obtain needed capital at favourable rates, • that there is a financial incentive built in that encourages individual performance of franchisees even though there is no direct monitoring of individual behavior. Franchising is a way of ensuring the performance of elements of the overall business, where the franchise model overcomes the moral hazard problem inherent in principal/agent relationships, and • that, most importantly, the franchisee possesses the discretionary authority from the franchisor to undertake the local business most efficiently. (He goes to some lengths, citing cases of McDonalds and Coca Cola, to show that the franchise regime harnesses the local knowledge of individual franchisees and their specific strengths that may range over various aspects of the whole business). The discretionary authority of franchisees utilises the cost advantage that individuals have in searching for the best ways of operating a local business in terms of capital expenditure, local marketing, supply contracts, remuneration arrangements, and in particular, in product opportunity. (Minkler, 1992) points out that: “…franchisees possess decentralised knowledge about local markets” and that in complex situations of many businesses “….the franchisor does not even know the alternatives, let alone the best alternative.” In large corporations, he is suggesting that the controlling offices can be remote from the coal face of commerce, but the franchisee, being local and opportunistic, can apply local knowledge to the local circumstance. In noting this, he identifies a characteristic of franchising; that franchisees have search costs for opportunities and improvements that are generally lower than those of the franchisor. Notably, the franchisee is also taking financial responsibility for its choices. In these circumstances, there is a good reason why a business operator may seek the involvement of franchisees as it ensures a wide canvassing of potential solutions to the business opportunity as well as the incentive to husband the capital employed at that level.
Page 45 of 259
4.2.3 Franchisor and Franchisee Considerations From the point of view of the franchisee, there are reasons why this situation is potentially advantageous. It is reported by Peterson and Dant (1990 Table 2, p55) that franchisees find, in order of importance, advantages in the following ways: “...
• Training provided by the franchisor • Greater independence compared to being an employee • The strength of an established organization in that they do not have to start from the very beginning to establish a brand. • Lower development costs for their business • The likelihood of a better investment of their capital. “ Interestingly, there was found to be no advantage in lower operating costs or in management involvement, as these were likely to be imposed to some extent by the franchisor and were recognised by franchisees as being necessary system inputs to any business and thus not an advantage of a franchised one: they would have to put in place these systems themselves if they were not specified by the franchisor. This suggests to Peterson and Dant (1990) that operators treat business in a similar way, regardless of whether it is franchised or not. This view of franchising, with its perpetual time frame and capital contribution of the franchisee, is to be known in this thesis as traditional franchising, to distinguish it from different uses of the term applied to certain public transport and other governmental arrangements. These other forms of franchising will be identified as they occur. It was noted by Loeb and Magat (1979) that franchising provides a decentralised method of regulating monopolies. They quote Demsetz (1968), who argued that “…competition at the franchise award stage would be sufficient to reduce the price of service below the monopoly level even though the increasing returns to scale dictated that only one firm actually supply the service.” This point of view is not completely shared by Sharkey (1979) but Sharkey notes that the approach of Loeb and Magat "… has considerable merit.... and ...is worthy of further and more detailed investigation". It is not clear that much academic attention has been given to this area of monopolies regulation and franchising’s application to this situation, but this will be tangentially reviewed in that later section dealing with monopolies. 4.2.4 Franchising in the Wider Economy The characteristics of franchising have been found suitable for the operation of a significant section of most western economies. In Australia franchising is reported to account for $178 billion6 . In the US the franchising sector accounts for $674 billion 7. It is thus accepted by many operating in and governing the wider economy to be an effective tool for certain commercial circumstances. (Hoffman and Preble, 2004)
6 Web address accessed June 2019: https://www.ibisworld.com.au/industry-trends/market- research-reports/thematic-reports/franchising.html 7 See web address, accessed June 2019 https://franchiseeconomy.com/
Page 46 of 259
A governmental view of franchising is provided by Chan et al. (2009), who defines the term to mean: An arrangement whereby a public sector entity grants an exclusive right to a private sector entity to access, occupy, operate and maintain public infrastructure facilities in order to deliver services over a period of time. They go on to note that: Government franchises are a variant of public–private partnerships (PPPs) in which the franchisees typically make no equity contribution to the infrastructure but could possibly raise finance for capital maintenance and extensions. Depending on the contractual terms, such arrangements provide a means of transferring certain performance and demand risks associated with existing infrastructure from the public to the private sector. This is noticeably quite different from a Community Franchise or a traditional McDonald’s’ franchise in two significant regards, one concerning the capital contribution, which in traditional franchising is made by the franchisee and the time frame, which in traditional franchising is generally in perpetuity, subject to regulatory compliance. It can be concluded from the literature that it is the combination of local entrepreneurs and centralised product specification and pricing that provides franchising with its particular performance characteristics and that this has proven a successful predictable business regime. There is no consideration here of the financial merits or otherwise of franchising as a business method, but research indicates that it is not materially different in its profitability from normal business activity (Stanworth et al., 1998) i.e. it is neither significantly more profitable nor less profitable than other business regimes. In short, it provides opportunities for people in certain circumstances, has certain conveniences to the participants and thus franchising is a regime that has its place if the circumstances suit. This provides a picture of franchising in the general economy which is sufficient for establishing the operating peculiarities of this regime. The extent to which this traditional type of franchising is present in public transport will now be addressed, along with a review of the many business models that public transport has previously and currently operated. 4.2.5 Franchising in Public Transport The term franchising is used frequently in public transport, although there are different meanings to that term described as traditional franchising covered in the last section. It is important to clarify these different meanings and to ensure there is a common understanding of the term that is applied in this thesis concerning the Community Franchise. The term franchising has been noted to encompass a very broad range of activities ranging from fast food to time-limited leasing of lines on British railways to the exclusive exploitation of a government-controlled monopoly.
Page 47 of 259
This broad usage of the term includes what is commonly described as sub-contracting where there are limited or no real ownership rights and only a defined period for the right to operate what has been often secured through a tender process. These terms are not used in reviews of world-wide franchising conducted by Hoffman and Preble (2004) and this is consistent with franchise experts having a more restricted view of franchising as compared to its use in transport circles where sub-contracting and franchising are often used interchangeably.
4.2.5.1 Franchise Usage in Different Jurisdictions Regimes described as franchises in public transport are certainly different in the Australian context. Government-controlled public transport franchise regimes do not, as a rule, utilise franchisees for the capital required for the complete operation as noted above. Nor do they provide much discretion in the way that sub-contractors or employees can affect the manner of the route layouts or performance of operations, and offer little or no security of tenure for operators of the sections of the transport business that they operate, beyond the limited terms of their subcontract, or ‘franchise’ period. There are variations of the regimes identifiable in the UK public transport with trains being treated differently to buses. For trains, the system of operation on the UK rail network is referred to as a franchise system on government web sites. In that jurisdiction, there is a divide between above the rail and below the rail. All permanent elements of infrastructure are financed, developed, maintained, and held by a company that provides the tracks for use by others. The network is owned by Network Rail (NR), via its subsidiary Network Rail Infrastructure Ltd. It is a fully owned enterprise of the Department of Transport. This body must periodically produce Route Utilisation Strategy documents, which are described on their web site 8 as being part of their “…long-term planning process (LTPP) strategy” indicating that it is fully involved in the planning of the network. The transport services themselves are provided by separate operators, e.g. Virgin Trains, East Midlands, etc. on the government owned network (Office of Rail and Road, 2017, White, 2006, Emerson et al., 2016, Preston and van de Velde, 2015). Each train operating company must bid for the right to utilise a particular series of lines through track access agreements. These line collections are periodically retendered and can be contested by several train operating companies. If successful these private operators then have the exclusive right to run their trains for the time slots of their lease period, generally in the order of eight years. These companies also run some of the railway stations with Network Rail (NR) having responsibility for only the largest of these. Operations are monitored by the Office of Rail and Road, a body established by the UK government to ensure all aspects of transport are reviewed for market efficiency. This is undertaken through the mechanisms available through the 1998 Competition Act. The private service-providing companies have discretion as to the type of train, level of appointment, etc., but most significantly as to the fare charged, giving the operator of the lines market power with this price control. The trains are branded so that it is obvious
8 https://www.networkrail.co.uk/running-the-railway/long-term-planning/ accessed Oct 2019
Page 48 of 259
that you are getting a train from one provider as opposed to another, and people are encouraged through advertising and fare deals to prefer a particular provider over others. With regard to buses in the areas outside of London, when first deregulated in the 1980s there was no overall franchisor that specified the product. or had input into network layouts, infrastructure type, operational standards, pricing, etc. All examples of operating companies currently involved in the UK regimes are private or public companies in competition with each other and in control of the level of fares for their own lines or time slots.
4.2.5.2 Differences between Franchising Regimes These two aspects, branding and pricing, are enough to distinguish the business model as being significantly different from franchise systems as they operate in other parts of the economy and those in public transport; and are quite different from the Community Franchise regime operating in the Dolomites ski fields. The right to operate being recontested periodically is also a difference from traditional franchising which generally has long term property rights subject to compliance with the franchisor-stated rules (and is noted to be in perpetuity in the Community Franchise regime and many traditional franchises). However, a similarity is that the capital for the rolling stock on UK railways is provided by the private company and this is generally a feature of franchising, but also of, say, subcontracting in the traditional sense of that term. These descriptive differences might just be considered semantics, but railways in the UK are different in operations and structure compared to the Community Franchise regime deployed in the Dolomites with significant differences to incentives and self-interest. There are differences in other matters considered important to passengers, namely operating times and fares. These considerations are controlled by the individual operators in UK trains and buses, but they must operate under the standards set by the Office of Rail and Road. In the Community Franchise, franchisees must operate to the minimum times and frequencies set by the franchisor and franchisees have no control over pricing: there is no discretion for franchisees in these matters.
4.2.5.3 Franchising: Usage for this Study There is a rich literature on franchising and certain definitions (Stanworth and Curran, 1999, Norton, 1988) nominally encompass what is practiced as government leases of the time-constrained monopolies of private service provision. Within business circles, this arrangement is more likely to be considered sub-contracting. It is noted that the definitions of franchising adopted by the peak body of Australian franchisors (Franchise Council of Australia, 2016) would include this sub-contracting practice, so it is important to be accurate in describing the franchise regime being studied in this thesis. It is also noted that, in all the discussion and review of the various transport regimes, there are virtually none that discuss transport franchising in the way that resembles that regime observed operating in the Dolomites. That traditional Dolomiti franchising regime is different from that often referred to as franchising in public transport both rail and buses.
Page 49 of 259
The case of UK buses is used here to illustrate those differences. In this discussion, where the government initiates the service, in the manner of van de Velde and Beck (2010) description of regime definitions, it is to be known as a Government Enterprise, regardless of whether the government undertakes the service delivery itself, or sub- contracts these responsibilities to private parties. It is noted that this regime version as it works in greater London applies broadly and occurs in most first world country urban public transport systems. On the other hand, there are those regimes where services are initiated by private industry and labelled by van de Velde as being a market-initiated service. This is a regime type that is certainly common but may not be as well represented as the government-initiated services. Regardless of the true extent of its reach, this market initiated service type will be known in this thesis as Open Competition, but the extent to which the private operators are constrained in their operations varies widely and has tended, as White (1997) notes, to be more constrained over time. The focus of this study is to compare the Community Franchise, and its specific form of traditional franchising, with the Government Enterprise. Regardless of the exact nature of the franchised businesses being conducted, the franchising literature identifies the person’s ownership of the income-producing asset, i.e. the business, as providing the property rights and security of tenure for the owner to operate without fear of loss of the asset or encroachment by others. Ownership is different in sub-contracting and these sub-contract type business arrangements in transport known as franchises. A sub-contract has no inherent value beyond the potential income from the remainder of the contract period. Restrictions are normally placed on contracts as to the disposal of the contractual obligations to other parties so contractual rights cannot normally be disposed of in any case. 4.2.6 The Special Franchising Case of Uber Franchising as it is practised in the wider economy is now practised in one specific commercial manifestation of what is considered a form of public transport. That industry is taxis, and an example of this is Uber. Franchising in the taxi industry is a recent phenomenon made possible by smartphones and their programmed applications (apps). The Uber phenomenon is not generally referred to as a franchise, but it fits the definition: the business model is such that the brand is that of the controlling company and it charges for the use of that brand by individual franchisees. The franchisor, Uber, organises the marketplace to bring provider and user together and takes a percentage of the cash flow for doing this. The type of company structure utilised by Uber is considered to be a franchise by several commentators, (Giuli and Maselli, 2015, Schneider, 2015, Sundararajan, 2014). The commercial entity Uber (the franchisor) does not undertake the taxi business itself; that is undertaken by franchisees. Uber has created a means of conveniently bringing buyers and sellers together through its marketing and phone apps, and by arranging the collection and distribution of the fare. Being in this controlling position, it can take a percentage of the turnover for its organising activities. It provides the brand to franchisees and they must adhere to the standards of the company if they wish to remain
Page 50 of 259
as part of that franchise and be paid by the franchisor. Uber may be the first and only widely publicised traditional franchise regime operating in urban public transport, (although there are other widely distributed less publicised start-up ridesharing companies like Lyft who carry out similar functions). The basic function of Uber is essentially the same as in ski lifting in the Dolomites where Dolomiti SuperSki has created a marketplace to allow providers to freely choose the means of service provision and for passengers to freely choose to use any provider. In the process, Dolomiti SuperSki handles all marketing, ticketing, and the distribution of revenue to an agreed formula, retaining a set percentage of that turnover for its contribution to the attraction of passengers. A principal difference in the Uber and Dolomit SuperSki regimes, apart from the fixed-line nature of the ski lifting compared to the point to point flexibility of a taxi, is the ownership structure. Uber is a publicly listed joint-stock company while Dolomiti SuperSki is a community-owned private company. This has implications for the alignment of interests between the various parties and is addressed specifically in the expert survey in Chapters 10 and 11. There are reports on the security of tenure of the business assets owned by Uber drivers, with cases of unfair disenfranchising (Wilkins, 2016, Kosoff, 2014). These cases are consistent with a for-profit company having a principal obligation to its shareholders and where these successful companies are directed by people who are not selected on their compassion to parties unrepresented in their corporation (Wellons, 2012). Another important difference between these two examples is the exclusivity of Dolomiti SuperSki to undertake the transport task, with that company effectively enjoying the exclusive right to authorise entrepreneurs to operate because of its control of ticketing which effectively prevents alternative companies to operate. This is not the case for Uber: in the taxi market, other competing brands and operators can more freely enter the point-to-point transport market in a way that is not possible in fixed line ski lifting in the Dolomites. Both Uber and Dolomiti SuperSki can be described as franchise regimes but differ in the way Uber is a privately-owned company listed on the stock exchange and accordingly is not a Community Franchise, fully owned by a community or government. 4.2.7 Summary of Franchising in Public Transport Franchising is a broadly defined method of conducting business encompassing many operations from the granting the right to operate a natural monopoly by a government to a commercial operator, to a more conventional McDonald’s fast-food type of independent business that franchisees operate under an umbrella of corporate governance, customer charges and standards provided by the franchisor. There is substantial variation from one example of franchising to another from the level of control of operations to the methods of collecting and distributing revenue. The type of franchising which is the focus of this thesis differs from subcontracting styles of ‘franchising’ and is at the conventional end of the franchise spectrum. Dolomiti SuperSki’s rendition of the franchising model
Page 51 of 259
differs from most conventional examples in that the owner of the franchisor enterprise is a community organisation with direct associations with the population of the area that it serves. This confers Dolomiti SuperSki a special status, and which has operational effects to be explored in further reviews of the literature to follow. 4.3 Public Transport Classifications This review of public transport regime types will seek to identify the various regimes, (at a high level only), and then classify them in a way to suit the topic of this thesis. It will not enumerate an extensive set of examples of the deregulation, corporatisation, and privatisation of regimes of public transport in major first world economies, but instead is targeted at a narrow set of criteria. If a regime does not have the essential franchise characteristics as occur in the Community Franchise regime of Dolomiti SuperSki, those characteristics that make it different will be established and the dissimilar regimes will be grouped with others with which it shares broad similarities. The search is for those regimes that are similar to the franchise regime being studied here. Firstly, though the applicability of the Community Franchise ski lifting regime to urban public transport, should be clarified. To this end, the concept of ski lifting as a public transport service will now be addressed. 4.3.1 Ski Lifting as Public Transport The concept of public transport including the lifting of ski patrons to the tops of mountains is not necessarily accepted by all as ‘public transport’. However, Veeneman (2002) defines public transport in a way that shows the relevance of ski lifting to that used in urban applications. He identifies several factors that must be present for the activity of urban transportation to be known as ‘public’. These are the carriage of unrelated members of the public by third parties over fixed routes, at defined periods, and times of the day for a fee. It can be established that ski lifting meets all this criterion. Members of the public, i.e. the groups and unrelated skiers who need to be transported to higher ground, are taken on mechanical devices over fixed routes by third parties (with no relationship to those members of the public). The times of operation of the ski lifts are set out like timetables of any urban metro, with start and finishing times. The frequency is a function of the mode of transport adopted by the owner of the ski lift and a fee is charged for the public to use the facility. By this definition, there is no doubt that ski lifting can be treated as public transport. The way that the Community Franchise ski lifting business regime fits into the larger range of historical and current examples can now be examined. 4.3.2 The Historical Legacy All public transport was initially privately developed and run (Barker et al., 1975, Bagwell, 2002). As the motor car made its effects felt in the 20th century and public transport became a less viable business proposition, governments of the day were persuaded to take over the public transport operations. This was seen to be a necessary and justifiable action because of the vital role that public transport had assumed in the running of cities. By the middle of the century, practically all public transport in first-world economies had come to operate as an arm of government, a situation that did not alter until financial
Page 52 of 259
pressure prompted action by governments not willing to shoulder the cost of these systems. 4.3.3 The Late 20th Century Reforms The UK government in the 1980s was one of the first to initiate the process of deregulation and privatisation in the transport industry. The Thatcher government first took the drastic action of deregulating all bus services outside of the London regional area. It is a process that has been closely observed since that time and widely reported on (White, 2006, Glaister and Beesley, 1991, Hensher et al., 2007a, van de Velde, 2013, Mackie et al., 1995, Preston and van de Velde, 2015). This drastic overhaul of public ownership of transport and the discussions this generated was the genesis of the Thredbo Conference series on Ownership and Competition in Land Based Passenger Transport, which has continuously examined this process on a bi-annual basis since the first conference in 1989.
4.3.3.1 Classification of Regimes This Conference is organised into recurring workshop topics. The workshop series that examines the issue relevant to this thesis is the “Governance, Ownership, and Competition in Deregulated Public Transport Markets.” It publishes a report after each conference year and over time has established what is a notional set of business regimes actually used by governments, or prospective business models for future use, to organise their urban public transport, based mainly on market initiative. They identified three regimes that they term “Models” (van de Velde and Preston, 2013), which were summarised as reported in the Conference Papers for the 2013 year as shown in Figure 1 below.
Figure 1 The three business models of Public Transport (van de Velde and Preston, 2013)
Page 53 of 259
This conference workshop theme was the reregulation of deregulated systems. There was an apparent acknowledgement and acceptance that market initiation of service improvements was making a valuable and different contribution to previous government- controlled services but that any approach required a level of government oversight. At the 2015 Thredbo Conference in Santiago in Chile, this list of regime types was considered again following the inclusion and presentation of the paper concerning the case study of ski lifting (Emerson et al., 2015a). As a direct consequence of this presentation, the possibility of including a fourth regime based on the description of the Community Franchise regime from the ski fields of the Dolomites was considered. The concluding summary of the Thredbo Conference indicated the possibility of a fourth regime, as per the following table shown in Figure 2:
Figure 2 The four business models of public transport following Thredbo 2015 Discussion at the Conference ensued as to both its legitimacy for public transport and to its uniqueness, having some characteristics similar to that which had prevailed in parts of the UK bus system before deregulation. Although there was some discussion at the Conference about the exact wording to be used in the description of Regime 4 (as the Community Franchise regime), it was acknowledged by the workshop leaders that the regime was different, quite unusual, and worthy of further investigation. In particular, the regime was acknowledged as being of a type that has not been documented before and was an addition to those known and described regimes up to this time. This thesis presents further information on this regime in ways that address the voiced concerns of that conference.
4.3.3.2 Market and Authority Service Initiation The conference categorisations make the essential distinction on whether regimes are ‘market initiated’ or ‘authority initiated’, factors originally proposed by van de Velde
Page 54 of 259
(1999). This classification was based on whether the service was either market or authority initiated. It was further classified as to the level of consideration, either Operational, Tactical, or Strategic. This classification was able to distinguish between the levels of planning and control in public transport and gave rise to the four groups in Figure 2 above. Van de Velde provides overviews of the major systems in member states of the EU. Regimes range from those like the French with tight central government-controlled Authority Initiative, through the German models of combinations of markets and government control to the more extreme deregulation of the UK Market Initiative systems. He distinguishes the actions of the involved players in their prescribed roles as initiators of services or the providers of predefined services. He also elaborates on their roles in determining fares, vehicles, routes, timetables, standards of operation, etc. in a hierarchical set of decisions at strategic, tactical, and operational levels. In its published form, the definitions were quite strict and van de Velde notes that many regimes in their application bridged between categories. Regime evolution developed as adjustments were made to many operations and authorities reacted to perceived opportunities and deficiencies after review of applications suggested useful adjustments. Accordingly, many regimes do not fit exactly in these broad classifications and have elements of more than one.
4.3.3.3 Contestable Lines and Time Periods UK rail is one area that has elements of several of the Thredbo categorisations, with trains both sharing infrastructure through the exclusive use of time slots and some having exclusive use of rail infrastructure. White (2006) reports that the competitive bidding market is possibly more effective than the situation “…with direct inter-operative competition within the same mode over the same routes”. He raises doubts about the effectiveness of competition in this form of business regime. Further doubt is raised in White’s observations on the erosion of the strict competition policy that initiated the reforms in the 1980s in efforts to allow for cooperation between separate operators on the one line to arrange common ticketing and to coordinate timetables, factors that would be useful for passengers and patronage. These were judged to be anti-competitive and he reports that it is an issue that “... has yet to be satisfactorily resolved”. Within this UK regime arrangement, there has been experimentation with varying degrees of control being given to and then taken back from private enterprise for route planning, vehicles used, etc. (Gwilliam, 2008) The re-regulation responds to the need for government to react to the actual workings of these initiatives. The Thredbo workshop shows that government has tended to generally retain the obligation to plan the broad outlines of the network and the services, but there are multiple variations to the original pure classifications. Extensive requirements are evident in the UK in their Transport Act 2000 (UK Government, 2000) where local transport authorities, as the responsible bodies, are charged with the duty to make plans, establish standards, ensure the equitable provision of services, etc. This has resulted in
Page 55 of 259
a largely commercial provision of public bus services outside of London initiated by private concerns, with a proportion of subsidised routes, initiated by government. This contrasts with buses within London where the network and services are defined by the Government authority and defined portions of this are tendered for private operations in time-limited contracts, in the order of 10 years. In the Netherlands, government-owned commercial enterprises undertake the planning work of government. The process initiated by the Thatcher government was an early example of a world-wide tendency towards less regulated markets together with engagement with the private sector and several examples provide a representative range of the types of business regimes that emerged. Concerning bus services, the UK government initially instituted a total withdrawal of regulation and support for previously subsidised services for all parts of the UK except for the London regional area. This has resulted in increasing profitability of the larger service providers, greater fleet replacement, and services but only a limited degree of ‘on-road’ competition as smaller providers were unable to match the market power of the larger providers (White, 2013). In London, it has remained to the present time a regulated service, subcontracted out to private industry but completely controlled by the government. This Government control is in contrast with the forms of open competition that eventuated in the rest of the UK (out of London) urban and country areas. The process adopted in the UK contrasts with the more cautious and varied approach taken in Europe (White, 2006, van de Velde, 2013, Van de Velde, 2003).
4.3.3.4 Contestable Area Contracts The dominant method of involvement of the private sphere in service delivery in Australia, and one that is common in many countries, is the use of time-limited area contracts, where companies tender for the provision of a defined service for an area of the city for a period, generally in the order of 5 to 10 years. The experience has been examined and reported on, especially for performance-based contracts (Hensher and Stanley, 2003), As well, there have been comparisons made with that in other countries in a general world-wide review of regulations and regime design (Hensher et al., 2007b). This review, it is noted, did not cover any regimes that resembled a Community Franchise.
4.3.3.5 Fare Box Retention One element that is sometimes put to the market in this arrangement is that of fare-box retention. This has been incorporated into several regime structures and has some parallels with the franchising in the Dolomites in that it is in the interests of the company running the buses that they encourage patronage to maximise their income. It is common in the Netherlands (Van de Velde et al., 2008) and is present in some areas of Australia (Hensher et al., 2007b). There are, however, distinctions that make this arrangement as it is used in these regimes, different from a Community Franchise. Farebox retention generally applies where a contractor has won a part of an urban area network to run the service as designed by the authority and this is for a limited period. The most significant difference to traditional franchising is that there are no property
Page 56 of 259
rights pertaining to the area that is being serviced, beyond the security of the operating period contested in the tender. Accordingly, there is not the same incentive to make improvements that might improve patronage, improvements that may be lost at the next round of tendering: a contracted operator does not retain any investment in this activity.
4.3.3.6 The Japanese Business Regime Approach The Japanese took a different approach in the liberalisation of public transport to both the UK and to Europe (Kasai, 2003, Fukui, 1992). There, the vast and extremely costly national rail public transport bureaucracy was dismantled and replaced by profitable vertically integrated public companies listed on the local stock markets on inter-city routes. One significant difference from the UK model of rail ownership in this regard was the arrangement to retain a single company owning the tracks and vehicle infrastructure rather than to provide for separate ownership of the tracks from those who operate the services, a feature reported by Van de Velde (2000). Safety of operations was, tellingly, one of the main reasons why the Japanese did not want to separate these functions, thinking that those who controlled the tracks should operate on them, in a classic attempt to align the interests of participants. Inter-city operations are divided on an area basis with five distinct areas over the whole country each of which is a virtual rail monopoly in its own area. A feature of city urban public transport in Japan is the way that transport companies can become involved in real estate to support transport operations (Van de Velde, 2000). He reported that the European separation of these functions ensures that this is virtually impossible within the EU. The regime adopted by the Japanese is a typical corporate approach with area monopolies, regulated to some extent by the Government and with the corporations planning, financing, and running the portions of the system under their control. Whilst franchising of business opportunities is common in many different industries in Japan (Nishimura, 1986), there does not appear to be any franchising to separate operators within these areas of conferred monopoly control in public transport that fits the Community Franchise regime or in the mould of an Uber-style franchise.
The different regimes for UK and Japanese railways can be summarised as follows:
• In the UK, the government decides where railways will run, using its planning resources and those suggestions for new routes registered by private industry; it plans and builds the network and is obligated by legislation to ensure that proper long-term planning occurs. There is no reliance on, or even encouragement of, private industry to plan or propose lines. Once lines are built, the government will sell off (franchised) rail operations to Virgin and the like, to operate the trains on those lines on the network. The government is not an operator of services itself, except in cases of commercial default. The portions of the network capacity sold to each commercial operator are in time slots. Different commercial firms operate trains over the same lines as others but in different times slots. These time slots are not held by these commercial firms in perpetuity and these must be re-
Page 57 of 259
contested periodically. Each separate commercial operator charges and collects fares for each passenger. • In Japan, the government makes the opportunity available to private industry to expand the network, e.g. the new high-speed line from Tokyo to Kanazawa. The government may allow certain expensive elements in the infrastructure to be funded from state resources, e.g. some tunnels. This extraordinary government funding is used to ensure some desirable connections are made, (e.g. to the relatively earthquake free north shore of the main island,) but basically, each new line is a commercial initiative wholly-owned in perpetuity and operated by the private commercial company which controls the fares charged on those lines in its possession. No other operators use the tracks of these commercial firms. Both contrast with the Community Franchise regime where the commercial operators propose, finance, and construct all infrastructure and then operate each line that they own in perpetuity. In a Community Franchise, the government-owned franchisor does not have any input into line proposals but may encourage certain modes by adjusting the rates on offer for the carriage of passengers on that mode. Vitally, the government franchisor collects all fares and pays franchisees for each passenger carried. In the area of Japanese bus deregulation, comment has been made by Sakai and Takahashi (2013), who summarise in their abstract that “…there has been little change in the structure of the Japanese bus market so far” and that “…it is the internal organisational factors (i.e. governance structure) that effect the operator’s efficiency, not deregulation”. An interpretation of this situation is that there is no open competition for buses as occurs in the UK, but instead, regimes that are essentially government-owned and operated systems with subcontracting as for most of the western world. There is much that can be further researched in the various national systems, including the Japanese and British regimes. However, it can reasonably be concluded from this analysis and the lack of conflicting advice at such forums as the Thredbo Conference series looking specifically at these issues, that the Japanese public transport regime in respect of its rail and bus systems as well as those in the UK, do not reveal the use of business regimes with the characteristics of the Community Franchise as here defined.
4.3.3.7 Developing Country Urban Transport Where urban public transport in developing economies is not organised by Government Enterprises other forms of public transport have evolved to cater for the need. This is most often an informal arrangement of private operators who serve the generally low- income neighbourhoods of urban areas (Cervero, 2000). These systems are commonly called paratransit and are characterised by the lack of or only minimal amounts of formal organisation. They serve the vital function of connecting the poorer car-less populations with their employment and markets. This informal transport sector is made up of self−employed entrepreneurs who generally lack official registration and operate in a competitive somewhat chaotic environment. This lack of formal organisation is a distinguishing feature of this regime and differs from the formal arrangements in first world countries (including those of the Community Franchise regime).
Page 58 of 259
Another significant developing country transport form is the ‘collectivos’. This is a form of transport described by Cervero (2000), who notes it’s ‘atomized, uncoordinated’ nature and the fact that that these are sometimes illegal operations, competing with more formal ‘franchise’ bus operations. These ‘franchise’ operations are long term agreements with city authorities that do not have the protection of enforceable property rights or the longevity of title over specific routes as in the Community Franchise regime. Collectivos would be described as a specific expression of paratransit but a more organised form controlled by local government. The essential form of collectivos is the independent operators who own their vehicles collect their fares and plan their routes. It is often a regulated industry with rules of operation, often with exclusive access to an area, but on a time-limited basis. These rules do not confer ownership or any exclusivity for a route to any of the participating operators. This is a distinct difference to a Community Franchise. Lines of transport in collectivos are flexible and vary as to the clientele and are chosen by the driver to suit passenger needs. It is unlike a Community Franchise in that there is no central franchisor collecting fares and distributing revenue on a per passenger/distance basis. It can be concluded that no regime in the developing economies operates in the same way as the Community Franchise defined in the previous chapter.
4.3.3.8 Thredbo Summary Using the ‘market/authority’ distinction, a Community Franchise regime may be classified as being at the ‘market’ end of the spectrum for the initiatives available to franchisees. But, it has extremely strict restrictions as to the discretion available to the franchisee over those elements of the market available for their imaginative exploitation, which may place it at the ‘authority’ end of the spectrum. It clearly, has elements of both and defining characteristics include:
• no discretion with fares, timing, and frequency of operations but • every discretion with line routing and infrastructure choices. • The franchisor, being the agent of government, has the authority role over the essential elements with social considerations. • the deliberate withdrawal of the franchisor from any route planning and associated modal choices. • strict enforceable property rights over the lines that the franchisee creates. The specific market/authority split of responsibilities gives the regime its defining and unique characteristics. Several of these characteristics are of great importance in regime functioning.
• Title over the asset of a line provides an inherent value with borrowing capacity. Having security of tenure, enables line owners to embark on improvements to layouts, infrastructure, frequencies over and above specified minimums, and any aspect of the line that might attract patronage for them. • The only tangible influence over franchisee service delivery is the rate provided by the franchisor for the carriage of a passenger a specified distance. All other
Page 59 of 259
decisions are left to the franchisee. Crucially this control does not give away discretion on fare levels; this is firmly in the hands of the franchisor. This characteristic ensures that larger more financially robust firms cannot use pricing power to create opportunities for themselves, i.e. they cannot engage in monopoly activities with the lever of fare pricing, a point emphasized by Dolomiti SuperSki in their case study interview. This thesis adopts a classification that differs from the Thredbo classifications and is marginally broader than them. For this thesis, any regime that is designed by the authority and then either operated by the authority or subcontracted by whatever means to the private sector is to be known as a Government Enterprise. Any regime where the private sector has discretion as to routes, fares, and infrastructure, and at the same time, no exclusivity as to line operations, i.e. no property rights over lines, will be known in this thesis as Open Competition. The Community Franchise sits somewhere between these two extremes. 4.3.4 Summary of Regime Status It can be concluded that there are no European or Australian regimes that operate in exactly the same way as a Community Franchise and that this regime adds to the number of ways that a public transport business can be classified. Of significance in this review is the extent to which regimes of public transport, however classified, are similar or not to the Community Franchise regime of the Dolomites. In this thesis, the four classifications of the Thredbo Conference are compressed into three to reflect the essential differences between those with market initiative, government initiative, and those with the franchise characteristics as per the following Table 9.
Regime 1:Community 2:Government 3: Open Competition Type Franchise Enterprise Route By each franchisee By government, who By commercial operators. Ownership with exclusivity to may operate the There is no exclusivity of that franchisee. No route, or sub-contract route ownership in bus other operator can the route to private services; multiple use the line owned operators companies can compete by another franchisee over the same routes. Rail and property rights operators can seek to are in perpetuity. operate competitively in time slots within the franchised or government- run network. Provision for competition for time slots is made within an existing government-owned network of routes.
Page 60 of 259
Network Suggestions for new Fully designed network Commercial bus routes Design lines and routes by by Government. are proposed by franchisees or Potential for portions of commercial operators on potential franchisees. the network to be a route by route basis. All design, subcontracted to private Creation of a network is construction and providers. often provided by public operating by funds which allow for franchisees. tendered supplementary routes for social reasons.
Setting of By Government By Government By Commercial Operator Fares owned Franchisor operating entity.
Payments By franchisor to each By government to By passengers who to franchisee based on subcontracting purchase tickets Operators passenger/distance commercial operators from each separate only. No payment just based on the contract for company to use on for the provision of each line being operated. the services that the service. Generally, not based on they operate. passenger/distance, but Common ticketing as a payment for the being introduced provision of the defined when anti- service. competitive barriers are overcome.
Table 9 Regime Characteristics It can be concluded from this review, that no evidence identifies any public transport operating anywhere in the world under a business regime that works in the same way as the Community Franchise defined in the previous chapter. Mainstream regimes differ in the manner of line ownership, the network planning function, fare collection, and recompense to private operators. This review will now examine several potential operational difficulties that have been raised as possibly affecting the application of the Community Franchise regime to urban public transport. 4.4 Demand Management and Subsidies Different mechanisms can be identified operating in demand management, capacity, and pricing. Nash (1978) assumes that demand will be a function of the price per passenger mile and the number of bus miles operated. He points out some arguments against such assumptions himself 9 but notes the accepted wisdom that the implied objective of maximizing passenger miles will influence demand due to availability. Nash also
9 The mentioned objections include the upper constraint of passengers able to be carried and the off-peak effect of overcapacity.
Page 61 of 259
investigates various ways of predicting the outcome of competition on existing lines, where additional competition for track access is replaced by awarding a monopoly franchise to a single operator (Johnson and Nash, 2012). This work is mainly concerned with lines of transport where access by other parties is allowed, as it generally is in the UK, where it was determined that although passengers may benefit, this would be at the expense of providers. It is noted that access for competing providers on any line would not apply in the case of a Community Franchise as each line, (but not an area), is individually owned and exclusively operated by the franchisee. The concerns of Nash and Johnson are for the UK experience only. Both the maximizing of passenger miles and the situation in the UK of multiple access of rival providers to any line are of relevance to this consideration of a Community Franchise in that neither aspect applies to that regime. A distinguishing difference between the Community Franchise and the prevailing situation in the UK is the way service capacity is determined. In a Community Franchise, once a franchisee has committed to a line, it is vehicle size and speed which adjusts to suit the fixed factors of recompense per passenger/distance and frequency requirements with the variable aspect of demand. The capacity is controlled by the franchisee providers by their choice of vehicles and support infrastructure to provide the appropriate level of service to meet the demand as they see it eventuating. As previously noted, in the UK situation the providers are given greater discretion to control service frequency, time slots to occupy as well as the level of fares charged. This leads to quite different considerations of both regulator and provider with different implications for both. Subsidies are also of consideration in Nash’s analysis and were a concern of researchers like Herbert Mohring (1972). It is accepted as normal by Nash that fare income will be less than the cost to provide the service (Nash, 1978). There seems little evidence to support the alternative. The Mohring Effect, named after his 1972 paper, identified the way that subsidies will encourage the production of the service for the optimization of overall welfare, particularly that of the passenger; it is in the supplier’s interests to supply the service and subsidies assist demand beyond what may be considered by other measures to be economic. And being a political discretion, it may not be subject to normal economic effects. This possibility is mentioned by Nash and also by Frankena (1983). Basso and Jara-Díaz (2010) and Williams (1981) also provide evidence that the Mohring effect is a valid argument for service subsidy. There do not appear to be significant differences in the way that subsidies are provided in most public transport arrangements. It involves the government supplementing payments to the service providers by either direct payments to franchisees by way of the rate per passenger/distance in the case of the Community Franchise, other payments to suppliers in other regimes and in all regimes by directly supporting passengers so that the fares can be afforded. If the method of subsidies and the effects are the same in all regimes, then it is not a factor that should determine the effectiveness of one regime over another because of this concern. It does support that suggestion however of Nash
Page 62 of 259
of the need for further research into subsidies and their effect on the overall economics of possibly excessive ridership. An aspect of subsidies that is different in a Community Franchise from existing area- based regimes is that of cross-subsidization of different routes by a provider. In an area allocation of routes, the one supplier can cross-subsidize over the multiple routes for which it will be paid. This is not the case with a line ownership model, where individual lines will have to be individually profitable. Other possibilities exist for subsidies to be applied in the case of individual line ownership models. It can be assumed that there will be geographic concentrations of need as this is an effect seen widely. Targeting those in need with subsidized or free travel through an Opal card or similar that operates the transport access barriers, will provide the capacity of those in need to have public transport. It is the less well-off who make good use of public transport and so lines in less well-off areas of a city can have good demand as long as the political will is there. 4.5 Monopoly Tendencies and Franchising Whether or not a Community Franchise gives rise to monopoly tendencies is important in terms of its potential suitability for urban public transport. Public transport in many jurisdictions is a state-run monopoly: the state reserves for itself the right and obligation to plan and run public transport. In this regard, the franchisor in a Community Franchise mimics the role of the government as the monopoly provider of that service. 4.5.1 History of Monopoly Behaviour A brief historical review shows that monopoly behaviour has been prevalent in many industries and many periods of history. It occurred with agricultural production and spice trades in ancient times (Frankopan, 2015), and memorably in the late 19th Century when the robber barons of the industrialising US economy seized the opportunity to dominate and control fast developing industries like steel, transport, newspapers, telegraph, etc. (Viscusi et al., 2005). The monopoly phenomenon seems able to emerge with nearly every change of technology or financial circumstance where early movers can gain advantages in a shifting economic and political landscape. Frankopan notes that monopoly behaviour has always been a concern to both rulers and governments but for different reasons. Absolute rulers want to have control of certain activities to secure monopoly rents for the maintenance of their power. Nowadays, in democratic societies at least, governmental concern is more because the activities of sections of society that can gain monopoly rents have deleterious flow-on effects to the rest of the economy and of society (Posner, 1978). Size often confers efficiencies and competitive advantage and those companies that can become large and efficient often see off their commercial rivals and achieve dominance (Layton et al., 2011). Under the economic state of perfect competition, monopoly dominance means the monopolist gains the ability to raise prices, reduce quality and quantity produced, and exclude competitors through effective barriers to entry. Some industries, because of their basic nature, are natural monopolies. Posner (1978) summarizes with the observation that: “If the entire demand within a relevant market can
Page 63 of 259
be satisfied at the lowest cost by one firm rather than by two or more, the market is a natural monopoly, whatever the actual number of firms in it.” He goes on that in these cases, competition is not a viable regulatory mechanism for control as “it is either short- lived” or “it produces inefficient results”. Some commentators consider urban public transport to be a natural monopoly in part because of the advantages conferred on passengers by common ticketing, a feature notable missing from say UK intercity trains (Evans, 1991). Natural monopolies, potentially, will provide the cheapest production of goods and services and so, if they can be managed and controlled, there is some justification for their presence, when controlled. It is debatable whether public transport is a natural monopoly or not, but that is not of importance in establishing if monopolies may develop in the industry. White (1997) notes that there has been a concentration of bus operator ownership in the UK at the time of his writing. But in such matters as ticketing, it is obvious to the regulators of the UK ‘Out of London’ bus servicing, that controls for common ticketing have proven necessary for efficient and passenger-friendly operations which implies an efficiency in that aspect at least due to monopoly control. 4.5.2 Control of Monopoly Behaviour Many authors have examined the effective ways open to governments to ensure that no one participant in any industry can gain such a significant share of that market that it can engage in monopoly behaviour. The literature records (see Viscusi et al. (2005)) that the measures available to government essentially boil down to three options, namely:
• having regulation of private firms, • using a subcontracting system of tendered services, and • the use of public enterprise These three options are common to others commenting on the regulation of monopolies in public transport (Hensher et al., 2007a), this last group looking at the structure of the contemporary public transport industry. One device currently used is to allow a monopoly over a defined area, which is periodically tendered. In contrast, the device used in a Community Franchise is the ownership of individual lines by franchisees, with competition, or the threat of competition being always present with the ability of other franchisees to open lines in competition with existing ones. When a decision has been made to not use a public enterprise, then control of monopoly involves both the prevention of exploitation and of ensuring that the monopolist operates at an efficient scale (Wolfstetter, 1996) and at a lower price. Wolfstetter does not consider the use of a regulator, acting as a franchisor, as a potential solution although that option was thought by several commentators to be a serious contender for inclusion (Rubin, 1978, Williamson, 1976, Demsetz, 1968). by providing government control over monopoly activity while still allowing competition between different industry participants. 4.5.3 Monopoly Regulation The most obvious need for regulation is to influence both the price and the quantity of a product. It is assumed that in a well-functioning market, if those two elements are
Page 64 of 259
controlled then the quantity will adjust to suit demand. As well as the twin concerns, of quality and price, is the need for regulation to prevent concentrations of power in a single or a small number of commercial operators that can allow for abuses of oligopolistic and monopolistic positions. Lowering barriers to entry, by whatever means, is another and fundamental objective of monopoly control. Importantly, the literature also argues that quality issues can ensue and are important. For example, Spence (1975) argues that “...market problems …. arise when a monopoly sets some aspect of product quality as well as price. It is argued that the market failure is associated with the inability of prices to convey information about the value attached to quality….” The three reported fundamental mechanisms for regulation of business monopoly are thus:
• Avoidance of a concentration of power in a small group of companies, • control of prices, and • the regulation of product with the specification of minimum standards Although there are aspects of this consideration, (e.g. concurrent pricing power and production control of a monopolist), any proper consideration of this topic must at least consider these three essential principles. It must do this with a concern for asymmetric cost information, which is shown to be a frequent reason for failures in the regulation of monopolies (Sharpe, 1990). 4.5.4 How Franchising Affects Monopolies Franchise regimes in the wider economy are generally private companies that seek the capital and operational expertise of others in the running and expansion of their businesses. To recap, a franchise involves several essential elements, namely;
• A product or service created and owned by the franchisor • A license whereby third parties can supply the product or service to the purchasing public • The need for franchisees to maintain the brand and all the standards for the product or service established by the franchisor • The payment by the franchisor to the franchisee for the supply of the goods or services to the purchasing public in proportion to that which is on-sold to the public. In a typical franchise, the product itself and all promotion of the product are undertaken by the franchisor, and the price for the product is totally controlled by the franchisor. The franchisee does not have responsibility for or control of these vital aspects. In the case of normal commercial operations, each firm can set prices, product design, etc. to their own estimation of the market and each firm competes with the others within that market. In a sub-contracting arrangement, the sub-contracting firm has no responsibility for the performance of the product or service, it just supplies it and will be
Page 65 of 259
paid for the production of the item or service regardless of whether that is sold or not; the subcontractor has no responsibility to the contractor other than to provide the item or service as specified. A franchisee, on the other hand, is only paid for the achievement of the aims of the franchisor. If it cannot sell the service or product, it will not receive any recompense. This is a crucial distinction as it puts responsibility onto the franchisee to achieve the aims of the franchisor in ways that are not part of a subcontractor’s normal considerations. The structure of the franchise regime gives it certain characteristics for the three considerations of monopoly control. These are discussed in the next subsection. 4.5.5 Exclusion of Competitors From the franchisor perspective, a benefit of the regime is the diversity of people and organisations that can contribute their knowledge, expertise, and capital, to optimise brand effectiveness (Peterson and Dant, 1990). It thus appears to be highly unlikely that a franchisor will end up with a monopoly situation with franchisees. In fact, the very opposite is more likely, is an observable characteristic of franchising in any industry and an outcome that is not contradicted by the literature. Franchisors benefit from a diversity of franchisees. Concerning pricing, the very nature of the franchisor/franchisee relationship in traditional franchising, has been shown to embed all authority over price with the franchisor. The operating franchisees running the individual businesses have no way of influencing the pricing to improve their own competitive position. This is certainly the case in terms of competition with other franchisees of the same brand: they are all selling the same product for the price regulated by the franchisor. Each franchisee, regardless of their size, has security of tenure over their assets with the ability to operate without fear or threat of loss of those assets. If they operate within the parameters of the franchisor’s rules, pricing by the franchisor ensures that they will be able to continue their business without commercial threat from other franchisees. This situation as to pricing establishes a responsibility on the franchisor to set pricing at levels that are sufficient for franchisee activity; without profitable franchisees, the franchisor will not have a viable presence in the market. It is also a safeguard against the abuse of pricing power and other commercial tactics by franchisees which could eliminate competitors and enable a franchisor to secure a more dominant position in the market. With respect to quality, franchising requires the franchisor to set the standards that franchisees must meet to continue to be able to continue as franchisees. Franchisees can only sell the product that is being offered and promoted by the franchisor. Controls for standards vary as to the industry, but it is evident from the many examples of franchises currently in operation, that common standards can apply to all outlets of any franchised brand, even if those outlets are owned and operated by separate individuals and businesses as franchisees.
Page 66 of 259
4.5.6 Ownership It is shown that franchising can provide control over the three essential aspects considered here as being required for monopoly control, i.e. quality, pricing, and standards. However, the ownership of the franchisor itself can have a great influence over the way this power is exercised. The case of Uber shows how a privately owned franchise can control those three elements to benefit the commercial entity over both the passenger and the franchisee drivers, a feature identified by several researchers (Wilkins, 2016, Kosoff, 2014, Schneider, 2015). But a Community Franchise is owned by government which has been shown to be responsive to community concern (Castro and Maddock, 1997). A government, if it owns the franchisor, can provide effective influence over the three important characteristics of monopoly concern, just as any franchisor can. So, consideration of this issue must address the differences between government and private ownership. Ownership by government of commercial entities, from postal services (Australia Post) to railways (Australian Rail Railtrack Corporation), are not uncommon. There are obvious differences between these private and public owners, and this characteristic will be put to specific consideration in the expert survey as part of the final section of this study. 4.5.7 Summary of Franchising and Monopoly Control The way franchising works confers an effective control mechanism to the franchisor over any participant for prices and quality. Franchising has this control built-in: in the case of Uber by control of the cash flow and the customer feedback and rating system, and in the case of Dolomiti SuperSki with cash flow control and oversight of timing and quality. There is thus great franchisor control in these and similar examples as it can effectively enforce standards and only releases payments on satisfactory franchisee performance. For these reasons, it is evident that a government-owned franchising company appears to have effective means of enforcing its will over aspects important to passengers and voters regarding monopolistic behaviour of franchisee service providers, just as Dolomiti SuperSki has over its franchisees. 4.6 Regulatory Capture and Franchising 4.6.1 Generally This topic is closely related to monopoly control. It has been noted by Dal Bó (2006) that in a narrow sense of the term: “...regulatory capture is specifically the process through which regulated monopolies end up manipulating the state agencies that are supposed to control them.” It is this narrow sense of the term that will be discussed here with application to a Community Franchise to urban public transport as suggested in Chapter three. The problem of regulatory capture is that special interests can influence government administration and consequently benefit from either special consideration or some arrangement that is to their benefit from government officers who will be rewarded in
Page 67 of 259
some way, quite often long after the original benefit is conferred. This can lead to firstly companies not providing services at their marginal cost and then to overall inefficiencies and a retreat of the ‘public good’ (Ayres and Braithwaite, 1991). The question for this thesis is if this type of behaviour is more likely to occur because of the structure of a Community Franchise regime than it is with a Government Enterprise or with Open Competition? Does the commercial structure encompass any peculiarities that might lead it to be more vulnerable to these effects than normal companies and government departments? The structure of a Community Franchise is just a commercial company acting as a franchisor. It is no different from any commercial company apart from the ownership, which is government, not private. Any individuals within that structure would be subject to the same temptations and constraints as an employee in any company and it is therefore unlikely that the structure would in any way contribute to an increase in the tendency for employees to engage in or be tempted by personal gain from engaging in illegal practices. All franchisees are private operators with the same constraints and tendencies. This thesis is not primarily interested in trying to answer questions on regulatory capture. However, it is worth noting that the commercial structure of a Community Franchise is no different, and the ownership of that commercial enterprise is no different, to many examples of activity in many other parts of the transport industry and hence regulatory capture should be no more an issue for a Community Franchise than it is for Open Competition or Government Enterprise. 4.6.2 Summary of Regulatory Capture It can be reasonably concluded that there are no special circumstances in a Community Franchise that would set it apart from other current available arrangements for organising public transport that would make it more vulnerable to regulatory capture. This is further explored in the expert survey of Chapter 10. 4.7 Transport Network Design and Franchising When transport planning experts are introduced to the way that the Dolomites ski field developed, i.e. without a central planner, they suggest this may lead to deeply flawed outcomes. This section examines the theoretical aspects of network design and the literature devoted to these questions more generally. 4.7.1 The Current Understanding of Network Design There is an extensive theory and literature devoted to the traditional design of public transport networks, typified by such publications as Bliemer et al. (2016) and of Mees and Dodson (2011). But these publications and others like them devote their effort to demonstrating how to formulate overall plans of transport networks by people and organisations with specific responsibility for this task. This literature tends not to give attention to the way that networks and other urban structure emerge from undirected activity in other parts of the economy, e.g. in the regular and predictable but
Page 68 of 259
uncoordinated development of retail outlets, service depots, entertainment venues, etc. of the wider economy. Transport planning as it is generally practiced is the centrally directed process of allocating resources to meet the expected demand for transport services for access to trip destinations. It involves the estimation of the demand for specific trips, generally by mathematical modelling and the estimation of costs of the various alternatives to meet that estimated demand. It is a field that has developed significantly since the middle of the 20th Century, especially since the development of computers to cope with the mathematics involved in these estimations (Bliemer et al., 2016). The model and programming responses that utilise the increasing capacity of computers to calculate additional complexity includes the increase in the number of travel zones, the increase in the steps taken to estimate demand (from three steps to four steps), the introduction of causal relationships between planning and transport, the use of the gravity models and/or growth factor models, and the introduction of cycling. One aspect that does not change with increasing capability of predictive methods that is enabled by the increased calculating capacity of computers, is the reliance on mathematical models to make those predictions of the behaviour of complex phenomena. Stopher, in his first chapter on the history and theory of urban transport planning, (Bliemer et al., 2016), acknowledges that estimates of future travel demand and network capacities are not an exact science and that caution is required in the use of the numbers provided by any program calculations. Stopher also concludes that the continual increase in the complexity of the models results in a corresponding difficulty in the interpretation and implementation of modelled results. Complexity and uncertainty are treated differently in a Community Franchise, with Dolomiti SuperSki not undertaking network planning as in other transport regimes. Instead, individual franchisees propose new lines which have proven to happen without the need for central coordination. This regime difference makes planning or the design of transport networks an area of importance to understand and explain. 4.7.2 Definitions of Design Design is defined in the Oxford English Dictionary as “the act of “…deciding upon the look and functioning of (a building, garment, or other object), by making a detailed drawing of it; e.g. ‘a number of architectural students were designing a factory’. In addition, they say that design is to ‘…do or plan (something) with a specific purpose in mind; e.g. with object and infinitive ‘the tax changes were designed to stimulate economic growth’. There is a body of opinion that sees design as being more than this, whereby it is seen as a process that reaches beyond the creative aspects and includes both evaluations of alternative solutions and selection from those alternatives in what is an iterative process. Goel and Pirolli (1989) are of one such group of researchers who note the inadequacy of a simple definition to encompass these obvious aspects of design. They comment that:
Page 69 of 259
“Design is too complex an activity to be captured in a one-line definition, particularly a one liner that purports to specify necessary and sufficient conditions. As such, our characterization of design starts with the observation that design as a category exhibits … prototype effects”. Prototype effects are used to accommodate the iterative, gradual, process side of design. Such a step in the process requires evaluation and choice between differing prototypes. This understanding of design as a process is widespread and is encapsulated in the work of the philosopher Daniel Dennett (2017 ch 13) and economist Eric Beinhocker (2007 p187). Their understanding of design specifically includes the evaluation of variation and selection between variations by reference to a standard. The OED definition, on the other hand, is more concerned with the generation of unique and new arrangements and does not give attention to the selection of alternatives from the palette of options in any progression of the design variation process. Dennett (2017 p22) in contemplating design and evolution states that “…I cast evolutionary processes as design processes...” He terms the human, deliberate, intelligent design processes as Cultural Evolution and notes that the creative process of introducing new and innovative variations to ‘old’ layouts, arrangements, component configurations, etc. is clearly only part of the design process. A designer is always modifying the prototypes of the imagined or modelled design, thus creating variation. There is a constant evaluation of alternatives in the designer’s imagination as the process of design proceeds at this personal level. Once a designer has advanced a design in his mind the results will be critically evaluated. This occurs by either the designer, or when the alternatives are expressed by drawings, models, etc. they can be evaluated and commented on by the group of which the designer is a member, and then by the designer’s client. This same process of evaluation is extended to the wider public if these alternatives are tested through a market testing procedure for which the item is ‘designed’. At each point of the process from the interior of a designer’s mind, to a wider group of individuals, a selection process will occur that acts as a filter to reduce the number of variations and to allow the ‘best’ or ‘most fit’ design to eventuate and to be that one which is selected for offer to a wider public. The OED’s definition obscures, or at the very least, does not highlight the fact that the three aspects of design, namely variation, evaluation, and selection, are a constant feature of any designer’s activities. Design can be considered the complete process of the generation of variation, review, and evaluation between alternatives and then selection. This is a minimum understanding of the process of design and a more explicit, complete description than that provided by the OED. 4.7.3 An Augmented Definition of the Design Process Based on this understanding, an augmented definition is proposed for this thesis to be: Design is the three-stage iterative process for the modification of an existing or imagined artefact, article, process, or other definable entity, being 1) the
Page 70 of 259
introduction of variation of any characteristic, 2) the evaluation of those variations against performance criteria, and 3) the selection between variations, to produce a new example of the artefact, article, process or defined entity. This definition covers the normal design process that all creative professionals from architects, engineers, planners, politicians, lawyers to artists, writers, musicians, painters, etc. all follow; creatively think, modify existing arrangements, evaluate, discard, move on to the next generation of ideas and restart the process. 4.7.4 The Design Process Applied to Network Planning A central observation of the case study is that the design of the overall transport network in the Dolomites has evolved without the need for a central intelligent designer. The franchisor exerts control over the process, but variation is generated by franchisees, evaluation is by the travelling public and selection is by economic forces. A broadened definition of design as an iterative process along the lines suggested and used for this study, can accommodate both the central planning of transport networks and the undirected network evolution which occurs in the Dolomites in its ski transport system. This thesis will now examine each regime by reference to variation, evaluation, and selection.
4.7.4.1 Variation In the Dolomites, the capacity to introduce variation to the ski lifting network is not restricted to a central authority retaining the exclusive privilege of providing the network design, as is the case in the centrally controlled private companies of commercial 'Integrated’ ski fields. There is a significant variation to the existing network in the Dolomites introduced because of the franchisee can themselves propose and install new lines of transport, without waiting for the franchisor to suggest opportunities. This exploits the tendency of individuals who have a natural optimism bias and extensive search capabilities to propose new lines. It also allows existing line owning franchisees to improved lines at their discretion to improve appeal to the public, decisions that are theirs alone to evaluate and not subject to franchisor scrutiny in any way. As Minkler (1992) notes, this is one of the key distinctions of franchising and enables the franchisor to benefit from the multiple and unsuspected contributions to solutions that are not necessarily known to the franchisor. Crowding and inefficiencies of any sort on the Dolomites network are seen by the franchisee, as the willing entrepreneur, as an opportunity to provide a commercial solution from which they can benefit. Decisions on new or improved infrastructure proposals in centrally planned networks are not known for responsiveness to public demands or agility in negotiating departmental barriers to change. Bureaucratic structures, necessary in public and governmental organisations, cannot be expected to allow the taking of decisions that are unusual or risky in the same way that multiple competing private entrepreneurial organisations can and do. Thus, one might expect to see more variation in the franchise business model, and this is reflected in the larger number of lines under Dolomiti SuperSki’s franchisee’s control and the greater number of lines per length of ski run.
Page 71 of 259
This wider offering of opportunity in franchise regimes increases the number of solutions that are proposed and attempted: there is no need to wait for public pressure or submissions to a central authority to have an idea tested. This opening up of opportunities to propose design variations by franchisees increases variation in the system compared to where there is just a single, constrained, and (sometimes politically) filtered source of new ideas for inclusions or variations to the network.
4.7.4.2 Evaluation In the Dolomites, the franchise nature of operations provides a fully functioning and effective commercial measure for the evaluation of each line in the network. The income of each line is solely derived from the criteria judged critical to operational success, namely the carriage of passengers, as lines only receive income proportional to passenger numbers. Each line can control and adjust its capital and operational costs to enable fine-tuning of expenditure and this provides the vital ability to remain profitable. Observation of normal commerce would suggest that this pressure will be felt more keenly and reacted to more decisively in private franchisee companies compared to a centralised owner of a whole network. The overall view of a centralised system does not have the same propensity to adjust for the peculiarities of each line, in the same way as the independently owned separate lines of the franchise regime where performance for that franchisee owner is vital. This difference distinguishes the Community Franchise regime in its ability to evaluate differences, in a design sense, effectively, compared to conventional networks.
4.7.4.3 Selection The evaluation of network typology, modes, frequency of operation, etc. occurs both in a centrally planned network, like in a Government Enterprise, and in a bottom-up system such as Community Franchise. However, the method of selection between individual lines is quite different in the two regimes. Failure to carry enough passengers for commercial success in Dolomiti SuperSki will lead to that line being eliminated unless it can change procedures to avoid this. There are different mechanisms at work in the centrally planned integrated ski resorts and Government Enterprise planned urban transport networks, generally reliant on centrally monitored data analysis. In those planned networks of public transport, there is no equivalent disbursed mechanism that serves this function of continual automatic evaluation and selection of component lines. In Government Enterprise centrally planned systems cost-benefit analysis of different possibilities is generally undertaken by the central planner to determine if a particular line or mode on a line, or any other network variation, has benefit to the overall network before implementation. Crucially, the Government Enterprise system can cross-subsidise operations to maintain or introduce operations in loss-making lines. It is noted that this is a functional aspect not available to franchisees in a Community Franchise. In that regime fares are set by the franchisor and cross-subsidisation of lines cannot occur in this way.
Page 72 of 259
Selection of successful lines will occur as lines without patronage that cannot adjust the economics of their own operations, fail. The economic effects of scale and flexibility issues on regime functioning have been examined in the ski fields of Japan (Patchell, 2014), where there are many community style ski fields with centrally planned transport networks. Comparisons in operation have been made of these community fields with the integrated commercial fields of the US. Patchell determined that, in terms of overall profitability, there was no clear advantage to either regime; the flexibility and supportive structure of the community regimes overcame the principal/agent and scale problems of the commercial integrated regimes. There are differences in the mechanism of selection for a Community Franchise where each line must perform or change procedures to avoid bankruptcy and removal from the pool of individual lines of the network. This is fundamentally different to a centrally planned regime where individual lines that do not perform can be cross-subsidised from other more profitable sections of the network, or the wider commercial operations of a ski field owner, e.g. catering, accommodation, etc. In these instances, lesser performing lines could remain in the network without great consequences to management of the overall ownership. The overall system is more tolerant of these inefficiencies. 4.7.5 Summary of Design and Network Planning An augmentation to the dictionary definition of design has been proposed for use in this thesis that considers the three acknowledged and relevant components of design, namely variation, evaluation, and selection. Both ski field examples, i.e. the conventional top-down consciously planned networks of most ski fields, and the bottom-up emergent layout of the Dolomites have undergone processes that encompass the three essential aspects of design as defined here. The way the variation, evaluation, and selection processes operate in a centrally planned Government Enterprise transport network provides different mechanisms for measuring and reacting to indications of inefficiency in the networks. This consideration of variation, evaluation, and selection in network design, explains the different operations and outcomes of the contrasting regimes of the centrally planned integrated ski field transport networks and the bottom-up planning of Dolomiti SuperSki. This is similarly reflected in centrally planned networks of any sort and those relying on market initiative. 4.8 Literature Review Summary The summary following the case study and this literature review is:
• There is a business regime, known as a Community Franchise, operating in the provision of a transport service that is unusual, differs from other regimes used for the provision of urban public transport, and is possibly unique. • This business model of a Community Franchise is a variation of the classical franchising regime with the distinction of the franchisor being locally owned by community members in the area in which it operates and not by private commercial interests.
Page 73 of 259
• This regime confers to the franchisor full powers of regulation on important aspects of consumer interest, notably the fares charged, standards of equipment and operations, and the frequency and times of service. • The Community Franchise regime conditionally delegates powers of line creation to franchisees who, to be part of the network, must agree to the provision of the service as defined by the franchisor. This has the effect of allowing decisions as to network design and development to be assumed by franchisees. • The franchisees are responsible for financial implications resulting from decisions made as to mode and levels of service on the routes in their ownership over and above any minimums specified. A summary of the differences between a Community Franchise of the Dolomites, the traditional regime of a centrally controlled and directed ski field regime and the traditional Government Enterprise public transport regime of most cities in first world economies are shown in Table 10 below.
Regime Type Community Integrated Ski Resort Centrally Franchise Controlled Characteristic Public Transport Ownership of Regime Government Corporate Resort Government Company Geographic Local, network wide Not necessarily local State-wide, area ownership greater than network Board Responsibility Local Area Shareholder base State Ownership of System Government Private Corporation Government Focus of Enterprise Passengers Attraction of patrons to Passenger Movement resort Movement Ownership of Lines Franchisees Resort Owner Government Setting of Fare levels Franchisor Resort Owner Government Payment for transport Franchisor to Internal accounting for Payment for defined function by: franchisees payments to staff service provision to contractors Basis of service Passenger Distance Employee salaries Employee salaries provider payment: or contract amount Network Planning By Franchisee Centrally to suit By central directive proposal corporate objectives Enterprise Control Rules of Operation Corporate Direction Central Direction Table 10 Public Transport Regime Differences This literature review provides the framework to develop an application to the urban area of the City of Sydney. This process will allow a set of questions and hypotheses to be framed to explore and address areas of regime differences. In the next chapter, an application of a Community Franchise business regime to the urban area of the city of Sydney is presented. This will make explicit the regime application to be examined in the rest of this thesis. A set of questions and hypotheses will then be framed to specify and address the areas of regime peculiarities identified above. But to first define what is to be studied in the urban environment, the definition of a Community Franchise will now be applied to public transport in the city of Sydney.
Page 74 of 259
CHAPTER 5 APPLICATION TO URBAN PUBLIC TRANSPORT 5.1 Chapter Content This chapter applies the algorithmic definition of a Community Franchise developed in Chapter 3, to the function of urban public transport in the metropolitan area of greater Sydney in NSW Australia. In so doing, it provides the object of study for the rest of this thesis. It firstly uses the definition to identify the responses to each step in the algorithm and then examines in more detail each aspect of the application. This process identifies the way that the regime is likely to work in the urban setting as opposed to the ski fields, and thus establishes the questions that will need to be addressed if this were ever to be contemplated. 5.2 Applying the Algorithm The essential characteristics of a Community Franchise have been noted to be:
• Monopoly control of the public transport function by government • Government ownership of the commercial franchisor • Reliance by the franchisor on the franchisees for the delivery of the service. • Franchisor control of ticketing, pricing, standards of operation and choice of franchisees The algorithm derived from the general principles of operations of Dolomiti SuperSki is now used to define the full extent of the regime for application to urban public transport. This is as follows:
Algorithm Step Response in consideration of urban public transport 1 Identify the objective of Moving people along dedicated corridors government policy. and roads in a city/urban geographical area.
2 Establish a geographic area that is The urban area which the transport system most affected by the activities to serves. For this thesis, this is the greater be undertaken. city of Sydney, excluding the adjacent cities of Wollongong and Newcastle.
3 Establish the means of measuring The number of passengers carried in any the attainment of this objective. vehicle.
4 Set standards to apply in Australian Standards as they apply to achieving the identified objective. movement in consumer and industrial activity today.
5 Estimate the effective value to the Use the current budget of the government government of meeting the as directed to transport for both ticketing objective. subsidy and capital works, as a starting point. Refine by experience over time.
Page 75 of 259
6 Establish a commercially A for-profit company with shares fully responsive body, the franchisor, owned by the collection of local government keeping this body in government bodies that represent the geographic ownership, and making this body city/urban area. responsible for meeting The board to be made up of local government objectives. government representatives or their paid staff.
7 Deny the franchisor the The role of the for-profit company is to act opportunity to actually undertake as the franchisor and will not be allowed to to achieve the objectives itself. engage in the transport function itself.
8 Require the franchisor to measure The franchisor must implement technology the attainment of the stated to measure the passage of a single objectives such that these passenger and the distance travelled. (This measures can be used for gauging is undertaken by the Opal card in use today franchisee activities. and would suffice for the public transport function).
9 Reward the franchisor based on its Set up the income of the franchisor so that achievement of policy objectives it obtains revenue only from the turnover of through the retention of a the franchisees engaged to carry percentage of revenues generated passengers. and appropriate employment conditions.
10 Have the franchisor publicize the The board of the franchisor to set objectives objectives, the standards and the and means of publicly broadcasting the rate at which it decides to pay for setting of transparent rates for the carrying the objectives to be met. of passengers over units of distance.
11 Give scope to potential The franchisor is not to prescribe methods franchisees in how the policy of achieving the movement of passengers objectives are achieved. but is to leave this to each franchisee.
12 Review and assess the proposals Government policy to be recognized by the for franchisee schemes to meet board of the franchisor in the setting of the objectives and determine rates for different modes, and distances, to whether they suit overall meet policy objectives. government policy.
Page 76 of 259
13 Identify and engage the suitably The franchisor to select and engage qualified proponents, i.e. the franchisees based on their likely capacity to franchisees, which will be paid by achieve the commercial objectives of the the Government in accordance franchisor and their fit with government with agreements which will have policy. regard to the success with which each franchisee meets the measurable objectives.
14 Provide property rights to the Security of tenure of any line that is essential elements of any allocated to any franchisee to be franchise to encourage investment guaranteed, with a register of all ‘property’ in the process, long term so identified maintained as is the case for commitment and the capacity to real property. borrow against the inherent value.
15 Allow for normal commercial No special privileges or support to be pressures to regulate the activities extended to any franchisee; all must survive of franchisees. commercially on their own merit.
16 Constantly review both the value Periodically publish any adjustment to rates and the payment to franchisees for to be offered to franchisees for passenger, the attainment of policy objectives distance and mode. in order to establish the rates most likely to achieve them.
17 Continue to review any new Establish a function in the franchisor proposals for policy attainment. organization to review policy effectiveness and implement change where needed.
Table 11 Application of Community Franchise Regime Descriptive Algorithm to Urban Public Transport 5.3 Detailed Examination Each step in this algorithmic process will now be examined in more detail with solutions specifically directed to the application of a Community Franchise regime to the city of Sydney, Australia, the geographical area extent of which is later defined. 5.3.1 Objective of Government The principal objective of government in considering the use of transportation corridors is mainly to cater for the movement of people and freight over these roads and railways, thus providing access to the urban areas of the city under consideration. The consideration of this hypothetical application is only for passengers in vehicles and does not include either freight, bicyclists or pedestrians. The measure of the principal objective is actual passenger numbers: it is not some proxy for this measure, i.e. it is not for bus operating capacity for example; only for the movement of people who are in the buses. This is noted to be an important measure and
Page 77 of 259
one that distinguishes it from systems that pay for the provision of a service unrelated to the numbers of persons in those vehicles. It has some parallels with those regimes that use fare box retention as an incentive to maximise passenger movements, but those regimes can differ from a Community Franchise in other significant ways. This inclusion of passenger movements acknowledges and aligns the objective with the fundamental purpose of the roads or rights of way. 5.3.2 The Geographic Area of Community Franchise Regime Application To have an alignment of interests of the people to be served, the decision-making body must be at a government level appropriate for the communities to be served, but no more. In the case of urban transport in Sydney, Australia, this body should represent only the greater city of Sydney, not the whole state. The area served by any local transport network is a separate system confined to that specific geographic area and the benefits and costs attributable to that system should be restricted to persons and organisations in that area. In this hypothetical proposal, the urban area of interest is the metropolitan area of Sydney, Australia, the capital city of the state of New South Wales (NSW). It has a population of about 5 million people and occupies an area of approximately 4,000 square kilometres, generally contained within a basin within the barriers of the ocean to the east, the national parks in the Blue Mountains to the west, and extending to the mountainous geography to the north and south of the urban area which virtually enclose this population.
Sydney
Table 12 Sydney in the State of NSW Australia (sourced from Google Maps 2016) This definition of Sydney excludes the adjacent regional cities of Newcastle and Wollongong which would have their own systems in this hypothetical proposition even though these are areas that by virtue of their being connected to the wider Sydney labour market, could be included as they are now for the total Sydney transport network.
Page 78 of 259
5.3.3 Measuring Passenger Numbers Electronically detectable travel cards (smart cards) are best practice for public transport, (including ski lifting), and these smart card systems are suitably developed to allow this one device to be used for all urban travel on public transport of any sort. Smart cards would still be used in a Community Franchise regime applied to urban public transport, and this is similar to the way it occurs in the Dolomites at ski lift entrances. When a passenger is electronically detected at a turnstile and recognised by the franchisor, a payment is processed to the franchisee operator of the line which is being used, with the franchisor taking its agreed percentage of the ticket price. All line operators would have the incentive to ensure that counting people onto their lines of transport was effective, in order to provide them with the remuneration that would be due to them. This technology is like current practice in Sydney (the Opal card). Every passenger entering and exiting a line of public transport is registered and this triggers an electronic calculation of the fare chargeable. This amount is immediately debited to the passenger’s card account. A feature of the Community Franchise regime is that the appropriate payment for the carriage of the passenger would be made electronically and in real-time to the franchisee. 5.3.4 Standards to Apply to Movement As with the Community Franchise regime in the Dolomites, the start and finishing times for public transport would be specified as well as the minimum frequency of operation. This establishes the headway, i.e. the interval between the arrival of each vehicle of public transport at any point, and consequently the frequencies of service. Current government regulations apply to most activities regarding transport, for example safety, and would not need to be altered as a result of the implementation of a Community Franchise regime. 5.3.5 Estimation of the Effective Value to the Government The exact figure for the economic contribution of the public transport function to a city does not need quantification for this construction, dealing as it does only with the concepts and principles of operation. However, there are some estimates of values that can be established in principle. The value of public transport is not a figure that is easily determined but is always a factor that is used to justify the various infrastructure works and the running costs of public transport to facilitate urban activities. Whereas the value may be difficult to determine with any certainty, the cost of providing public transport for the whole state of NSW is a figure that is recorded and known. Taking recent government reporting on the finances of TfNSW10 it is recorded that the broad figures are yearly expenditure of approximately $14b, and yearly revenue from fares and other charges of around $4b. It can be taken as a starting point that a figure around $10b over and above fare receipts is currently spent each year to support public transport operations for the state of NSW. The exact quantum is not important for this
10 http://www.budget.nsw.gov.au/
Page 79 of 259
application as the focus is only in the principles. The budget figures include capital and running costs of operations and this indicates that successive governments of different political persuasion have felt it necessary to expend significant amounts each year, in the order of $10b, to keep the city moving and facilitate network improvements. From this, and taking into account the smaller area of operations for the Sydney public transport ROC to look after the function in this specifically defined geographic area, an amount can be estimated that would satisfy the need for movement of people around the city in a Community Franchise regime. A figure will not be estimated here but it is an obviously possible and necessary task. A principle might be adopted that State Government finances are not disrupted by such a change of regime in Sydney and that the amount would be set so that any net effects on recurrent expenditure would be initially neutral. This figure, i.e. in the order of $14billion, is that which the ROC’s would be directed by the State Government to be made available on an annual basis through the franchisor for distribution to franchisee service providers for the carriage of passengers in Sydney. The figure paid to franchisees needs to be set such that it would be sufficient for the franchisee to support the use of its capital employed for operations. This would be thus analogous in this regard to the current State Government budget, i.e. it is sufficient to cover the cost of borrowings for capital expenditure in public transport. Also, the current State Government figure includes a subsidy to TfNSW in that the fares do not meet annual expenditure. This situation does not have to change under a Community Franchise regime. A subsidy from the State Government to the public transport ROC can, if necessary, be maintained. The subsidy amount would be that figure above the receipts from fares minus the costs of the franchisor’s operations. This figure could be justified in the same terms as it is under the present regime, i.e. by the need to maintain an effective service vital to urban operations. 5.3.6 The Commercially Responsible Body - The Franchisor
5.3.6.1 Regime Structure and Responsibilities The community franchisor would, as previously identified, be a government-owned profit- seeking commercial entity which would place it in similar circumstances as organisations such as Australia Post, (a commercial entity owned by the Commonwealth Government) or the Australian Rail Track Corporation (a commercial entity with shared ownership by the Commonwealth and State Governments). The financial performance of Australia Post provides evidence that government owned corporations can perform profitably and, as was shown in the Literature Review of Chapter 2, can be responsive to owner, i.e. government, concerns. The community franchisor would have the responsibility for satisfying the demand for the movement of people within the urban area. The franchisor organisation would not undertake these functions itself, but as the standard-setting body, make the necessary provisions for them to be carried out by other commercial franchisee bodies in
Page 80 of 259
accordance to the rules that the franchisor specifies and includes in its agreements with franchisees. This arrangement does not presuppose any particular mode by which demand is to be satisfied. It is open for franchisees to use any means of facilitating and satisfying the need to move people along the lines in their ownership.
5.3.6.2 Ownership of the Franchisor The franchisor entity would be owned by the community that it serves, which in this thesis is the greater city of Sydney not the whole state of NSW. This would be similar to the regional transport operating areas of Europe and substantially covering a single labour market. Recognition of the ways of most directly ensuring an alignment of interests of all parties is to then require that urban public transport for Sydney be the responsibility of elected bodies of that geographic area deemed to represent the labour market and urban area. In this example, ownership is to be held by a body that is similar to the many supra-local government organisations that have been formed throughout Australia to deal with local issues affecting more than a single local council area, but less than a whole state. This gathering of local government areas in Sydney into self-interest groups is common and is noted to occur in the formation of Regional Organisations of Councils (ROCs) in such organisations as WESROC or MACROC. There are many such groupings of local councils in Australia (Marshall et al., 2003) and these are performing as quite capable, sophisticated organisations able to function effectively over diverse fields. Marshall et al. note: “These networks (of Councils in a ROC) perform a critical governance function in so far as they provide a coordinating mechanism for diverse views, and find solutions to specialised problems that are not catered for by existing hierarchies and market systems.” A grouping of Council areas, as in a ROC for the whole of urban Sydney (which could be known as TRANSROC for instance), is considered highly likely to take up such devolution of powers for the direction of urban public transport in their own collective municipal area, having demonstrated this propensity in other issues of interest. The State Government would retain its overall surveillance of the whole system, principally because it would have a necessary role in the initial and ongoing funding. The Federal Government would have no substantive role as is the situation today, a situation that has parallels in the UK where transport is an activity devolving to the smaller geographic entities of Northern Ireland, Scotland and Wales. This arrangement can be argued to be the most appropriate as it would be controlled at the level of government, i.e. the elected representatives of the ROC, which is closest to the effects of decisions on public transport. The best economic outcome facilitated by an efficient public transport system is of great importance to local areas, where job creation is one of the most significant functions of any representative body and where the retention and creation of jobs have enormous flow-on effects to local populations
Page 81 of 259
(Moretti, 2010, Acconcia et al., 2014). The franchisor would have responsibility for public transport only in the geographic area of the community being served. This is like that which occurs in the Dolomites where Dolomiti SuperSki is owned by the collective of twelve constituent ski valley marketing bodies and does not represent any other identifiable area of the Bolzano region of northern Italy. Public transport and movement of people and freight would thus cease to be a matter dealt with in a detailed sense at the State Government level on behalf of the various cities and regional areas of NSW. Responsibility for detailed operations would be devolved to the level of the urban area, i.e. Sydney or any of the regional cities that have public transport requirements. This devolution of responsibility ensures that local areas are the ones that stand to pay for and to benefit from any investment made. As well, the people directing overall operations, i.e. the directorate of the franchisor, are either the elected representatives of voters or people selected by them. Being local, they would be more responsive than would departmental policy application at wider (State) levels of government. In the Dolomites, there is a clear alignment of interests for the population of the local area and the company operations and would translate well into an urban public transport application.
5.3.6.3 Capability of the Franchisor An aspect of consideration is the capabilities of different organisations for running a public transport network. Is a ROC sufficiently resourced and capable of running something as complicated as this? As it is now administered, the public transport operation in NSW is a considerable task with a complicated and costly team of experts, planners, operators and administrators, employing thousands of staff and being responsible for approximately $14 billion expenditure on an annual basis (Transport for NSW, 2016). However, the example of the transport franchise operation in the Dolomites and which is utilised in this theoretical application, shows that the task of the central authority is greatly simplified when the bulk of these tasks are not carried out by the central organising authority. In the Dolomiti SuperSki organisation and as is proposed here, the principal task of the franchisor is the establishment and the setting of standards, and of the monetary rates at which passengers are carried, as well as the running of the computer system to collect revenue from ticket sales and distribute payments to franchisees. This is not as complex a business as the Government Enterprise regime of public transport as it is currently organised. Additionally, it is intended that the body carrying out these simplified tasks is a commercial entity able to hire the most appropriate personnel. The tasks falling to the owners, i.e. to the ROC, are thus mainly of stewardship through the directorate and in this regard, the alignments of interests of the governed and the governors are given paramount importance and potential for oversight.
Page 82 of 259
The franchisor company will be served by a board of directors with suitable experience and could include the Minister responsible for transport functions in the State Government, suitable representatives of the local government council areas represented in the ROC, and independent directors as thought necessary to enhance the board’s capabilities. 5.3.7 Curtailment of Franchisor Functions The franchisor company would not in itself undertake any transport functions but would create the financial incentives and conditions to have these tasks performed by franchisees to the rules and standards which it specifies. This split of responsibilities establishes the franchisor, with its locally based board of directors, as the regulator. It creates opportunities for all prospective independent franchisees to suggest ideas for implementation of elements of the system and for their operation. In effect this harnesses what was described as the lower search costs for solutions by franchisees, who are in a better position to see local problems and solutions compared to a more remote centralised bureaucracy, the probability of which is explored in his paper “Why Firms Franchise: A Search Cost Theory” (Minkler, 1992). 5.3.8 Measurement of Transport Activity Public transport usage is currently monitored in Sydney using a distance-based smart card system called the Opal Card. This is like systems in use in many transport systems in other parts of the world and would need minimal alteration to be used in a Community Franchise regime. 5.3.9 Reward to the Franchisor As for the Dolomites, the public transport franchisor would retain only a percentage of the fares collected: this becomes the franchisor’s main source of income. The franchisor sells the tickets and passes on to the franchisee line operators the agreed amount for their carriage of a passenger. The incentive is there for the franchisor to maximise the number of passengers and thereby maximise its income. There may be other income, for example from advertising, but maximisation of passenger numbers is a guaranteed way of increasing or maintaining the owner’s commercial income. In the case of the Dolomites, the franchisor, Dolomiti SuperSki distributes 97% of the fares collected to franchisee operators. Based on the Dolomites experience, an assumption can be reasonably made that retention of 3% of total turnover would be enough for the franchisor to run a public transport system of the size and complexity of Sydney. With only a set percentage of total fares received as income, the franchisor is incentivised to undertake its function most efficiently and economically. All economies of operation directly benefit the franchisor. As a commercial organisation, it would be able to use its profit margins as a factor in the remuneration of its CEO and all staff, so there would be a clear advantage in performing efficiently. The CEO of a public transport franchisor can be expected to drive efficiencies in the organisation.
Page 83 of 259
5.3.10 Publicising the Franchise Objectives and Standards The franchisor acts as the marketing organisation for the whole public transport network. It would be a highly visible entity which publicises its fares, sells the tickets and promotes the use of its network. As with most commercial organisations, part of its public interface, as is evident on most commercial web sites, would be the opportunities that it provides for both employees and, in this case as it is a franchise, for franchisees. The franchisor would have a separate and distinct engagement with the transport industry generally, the franchisee providers of the transport services, or anyone wishing to become a franchisee. It would undertake the appropriate market research to establish the fares which would generate a sufficient total income to sustain the payments to each service providing franchisee. These payments would fund the capital and running costs of the network elements. Rates would be set for the different modes of transport accordingly. The franchisor has a distinct incentive to set fares such that there is a keen interest generated from providers (potential franchisees) to undertake the transport function at the levels set. This iterative exercise would be like most commercial price setting and one that would be refined over time to attract the necessary interest from industry. As happens in the Dolomites, the rates that result from this practice would be made public for any potential franchisee to view. These rates would be updated periodically and reflect the response to prevailing conditions in the industry, the particular ambitions of the franchisor board, and any other considerations that the CEO and the board would take into account, including its government owners. In the Dolomites, the significant amounts of data collected by the franchisor are shared with the different component (valley) areas and individual franchisees in ways that protect commercial confidence. It would be expected that the franchisor in the case of an urban public transport regime would likewise share the considerable intelligence residing in their overall database with the component local shareholding Councils and the individual franchisee companies in a similar financially discrete manner. This would be a significant source of planning intelligence for Councils as well as for franchisee use, facilitated by the investment made by the franchisor in its computer management systems. 5.3.11 Giving Scope to the Franchisee It would be a requirement that the franchisee proposes its own plan for the line route, mode and operations: this is not dictated by the franchisor, only that it must conform with the basic service frequency, times of operation and safety requirements. Any proposal from a franchisee or a potential franchisee would be considered on its merits; the franchisor has the incentive to consider all proposals. Proposals would cover the actual route of a particular line that a franchisee would like to develop and run and would extend to the vehicles to be used to fulfil the franchisee’s obligations as to the standards of operation, to the vehicular stops, line transfer arrangements that would make using their line attractive, and all design variables that might assist commercial viability.
Page 84 of 259
Being a franchise means that any line of transport, if already owned by another franchisee, cannot be then taken or used, even partially, by any other party: it is the legal property of the franchisee. The implications for the network of individual line ownership concept are spelt out in a later section. But once in the possession of a particular franchisee, that party has legal right to exclusively operate on that line without any other provider being able to concurrently use it without the permission of the incumbent franchisee; the legal owner of that line. This exclusive use and legal possession of a line give the owner, (see later discussion), the incentive to invest in a way that will maximise the attractiveness and ultimately the throughput of passengers, which of course may include sharing facilities that make commercial sense, e.g. transfer stations. Ownership and legal possession ensure that the franchisee and the bankers who are involved can have the confidence to invest with the knowledge that all returns can be used to satisfy lending obligations and ultimately to build on the value of the asset. The scope open to the franchisee is evident. This includes the mode chosen to be used, the types of vehicles, large or small, with or without elaborate seating, the staffing needs for the operations of the setup, the infrastructure to be used at stops, the devices to be installed at, for example, traffic lights to speed up the passage of vehicles, the devices to make transfers to other lines convenient, the inclusion of services such as WIFI, etc. All these potential offerings are matters that are likely to have varying responses by different franchisees as to what are the real requirements and the necessary inclusions. It would be most unlikely if all franchisees had identical views about all these matters. In fact, the specification of each line would be a reflection of the franchisee’s opinions as to what would attract passengers and is affordable at the cost of the capital it takes to set a line up in the manner intended. It is in the interests of the franchisor to allow the proposals from the franchisees to have the greatest scope and not be directed in any particular way, so that all the ingenuity of the broadest spectrum of the transport industry, no matter how mainstream or unconventional, can be harnessed to contribute to the transport problems of any area. That is, it is an example of how the franchise regime can lower the search costs for solutions to the franchisor, by using the imagination and experience of franchisees. This system utilises human tendencies of optimism bias fully tempered by loss avoidance, so that, if it follows the progress of network development as observed in the Dolomites, a degree of cautious entrepreneurial opportunism would emerge to drive the progress of network development. The Dolomites experience also suggests the possibility of a significantly higher number of lines of urban transport than occurs in that of a normal centrally planned network, a direct result of encouraging individual franchisee proposals to ‘search’ for self-funded solutions, from which the franchisees can benefit. 5.3.12 Review and Assessment of Franchisee Proposals The franchisor would have the incentive to assess any proposal for a line as it may result in additional passengers being added to the network and thus to the franchisor’s revenue. The experience of the Dolomites suggests that the particular routes proposed
Page 85 of 259
by a franchisee for any line may be of interest but will not be of great importance to the franchisor in their assessment. Rather, it is the experience and financial strength of the proponent, the capacity of the financial backers and the experience of the franchisee in undertaking this type of work which matter. The franchisor may have opinions on what routes may assist in the development of the whole network, but in the Dolomites, it has not proven necessary to encourage any specific development, except by the adjustment of rates of recompense to franchisees for different modes. In an urban setting, likewise, there is be no need seen for the franchisor to scrutinise proposals for layouts of lines or the modes to be used, except for compliance with standards of timetabling, frequency and safety and regulatory compliance, which would in most cases be with other authorities, e.g. local government. The transport operating industry itself has the necessary knowledge and technical know-how to undertake the detail of the planning. The review and assessment of franchisee proposals would mainly be concerned with the financial capacity and the relevant experience of the proponent. The franchisor would have the most interest in establishing the extent to which it is reasonable to trust the judgement of the proposing franchisee. 5.3.13 Engagement of Franchisees Once franchisee proposals have been vetted and selected, franchisees would be given conditional approval to commence operations. Actual operations may have to wait for other approvals, such as local government, health and safety, etc. or for construction of infrastructure to be completed, or vehicles delivered. Essential to the operations would be the installation of the passenger detection technology which records the throughput of passengers. Once these aspects are in place, operations of the line would commence, passengers would start using the line with tickets sold by the franchisor, and payments from the franchisor to the franchisee would start to flow. This process is similar to most commercial transactions in business. It is noted that time in these matters is almost always of the essence and this will encourage the earliest completion of installations to minimise any delay in receipt of revenues. 5.3.14 Provision of Property Rights The provision of property rights as an essential aspect of the Community Franchise regime is worth labouring. It is the existence of property rights that create the value that can be used to borrow capital to invest in line improvements. Without property rights, there is less financial capacity to make a line more attractive to users and to develop the line to add to capacity and patronage to create greater value. The owner of a line through a city CBD would have huge borrowing capacity because of the value inherent in having a somewhat captive passenger base. A lender would have confidence in the ability of a competent owner to service any loan provided to build the infrastructure to satisfy this need, providing always that the franchisor has built and maintained trust in its ability and willingness to pay in accordance with its stated aims. In a Community Franchise regime, the right to ownership of individual lines is threatened only by the inability of the franchisee to operate the line to the standards agreed to with
Page 86 of 259
the franchisor. A line in one franchisee’s possession cannot be used by others except where proper arrangements for this are made. Such arrangements for use by other franchisees may include leasing of parts of a line, easements over shared infrastructure, or other means which could use established mechanisms as frequently happens in real estate and other areas of commerce. Property rights appear to be a concept that is easily misunderstood or the implications of its application to lines of transport in a Community Franchise not fully appreciated. Only the owner of a line of transport is authorised to operate the public transport on that set of roads or other rights of way and this exclusivity of use could create difficulties for the current layout of lines of public transport in Sydney. If a Community Franchise regime was to be applied to the city of Sydney, where at first glance, the existing network does not seem to lend itself to a regime of individual line ownership, the implications of this need to be reviewed. The exclusive right to use a line would have specific implications for many situations that currently exist in the evolved networks of urban areas. For example, where many lines of transport converge on city centres with a sharing of infrastructure, be that roads or rail. In a Community Franchise regime, there is only one owner for any line and so the nature of the network would inevitably change from that which is in existence in Sydney today which can be described as ‘branching trees’, and in many other cities with this network typology. It would inevitably move towards a network typology that is more like a mesh where individual owners can own lines which cross but do not share the same route. This is reckoned by some commentators not to be a detriment to public access. The work of Alexander (1964) where he examines the effects of different physical urban access arrangements, shows that a mesh network has distinct advantages in urban areas compared to branching tree networks. He gives a definitive explanation for the need for flexible layouts he describes as “semi-lattice” type networks as opposed to the restrictions inherent in what he describes as a “tree” structure. The type of public transport network encouraged by a Community Franchise regime is of the ‘semi-lattice’ variety with flexibility and without the congestion and restrictions caused by converging branches. A corollary implication is that a mesh network may increase the need to change lines to complete a journey. In cities with semi-lattice or mesh network typologies, there is no advantage available for being at a singular focus of a public transport tree structure (like Central railway in Sydney) because any focus of development is less defined and concentrated. The examples of Moscow and Mexico City are two metropolises that have mesh typologies for their public transport networks and do not have the intense concentration of development at a dominant CBD core (Emerson and Parolin, 2011). New York would offer a similar disbursal of development opportunity and transport demand over the whole island of Manhattan. The potential impact on the public transport network can be illustrated by reference to a specific instance. Consideration of this instance suggests that the routes of the various lines of transport that exist in Sydney today will alter from the ‘branching tree’ structure that dominates at present to a more disbursed ‘mesh’ structure of routes more amenable
Page 87 of 259
to individual line ownership. This instance to be examined is of the bus network that traditionally and still largely radiates from the central CBD(s) of the city and regional centres. It is enough to examine just a single instance of this layout to suggest what is likely to happen if the lines of transport were to be owned separately by individual franchisees. In Sydney, up until very recently, at least 27 different bus lines started at a single point at Circular Quay in the urban centre and all progressed out of the city centre using a kilometre or more of Elizabeth Street, the principal north/south route to the ferry terminals. These lines branched off at different points along this main thoroughfare to go to their ultimate route destination. These destinations ranged from Watsons Bay in the east down to Botany in the south, an arc of about 20 kilometres as indicated in concept only in Figure 13 following.
Table 13 Branching Tree Typology under the Current Government Enterprise regime (simplified and in concept only) In a Community Franchise regime, the line along the north/south Elizabeth Street would be in the ownership of a single franchisee. Every bus using Elizabeth Street would be the property of that franchisee, which would collect the revenue of all passengers using that line. However, the particular franchisee owning that line would have only a single destination, as lines in a Community Franchise regime do not branch. The line would tend to be part of a mesh typology. For any passenger to reach any of the other 26 destinations that were previously served from Circular Quay, there would be a need to change buses at an appropriate intersection and board a bus on another (crossing) line that goes to their ultimate destination. This is illustrated in concept in Figure 14 below with several other (but not remotely exhaustive) possible responses to public transport opportunities.
Page 88 of 259
Table 14 Possible routes for lines in a Community Franchise regime with examples of some other potential lines (shown light) The mesh arrangement of lines is obviously different from the branching tree of the current Sydney layout. It is, also a layout that would be found wanting in many different ways by observant entrepreneurs, who may see value in additional routes between those shown or, for instance, a better destination for the airport line to proceed straight over the harbour bridge crossing to the North Sydney regional centre, to serve that destination as well as the CBD of Sydney. Those would be the type of alterations made by franchisees who would not be constrained by a central planning authority and would have the capability to do these things by proposals for new lines, as well as line acquisitions combined with line mergers and subdivisions. This mimics what happens in normal business transactions in any discipline where the opportunities are available. This mesh layout is not a radical conceptual design development, as it is the type of network that Sydney is deliberately moving towards as indicated by the intentions of the planning body controlling the network layouts, Transport for NSW. This is instanced by the MetroBus routes that cross the city joining regional centres in the manner of a mesh. This is a feature of network development flagged by the Sydney bus authorities (Transport for NSW, 2012) in their Long Term Transport Plan, of that time. It is a network typology that is found in many cities and is quite a usual pattern of network development, encouraged to overcome the disadvantages of an overconcentration of supply and demand for any single location. These concepts are also outlined in the HiTrans public transport planning guide (Nielsen et al., 2005). In a mesh typology, it is possible to get from any origin to any destination by using only two lines and changing at the appropriate point where those two lines intersect. Mesh networks that are mature and developed in any city provide this same flexibility of access. This network typology of individual lines would almost certainly develop under a Community Franchise because of the nature of ownership. It can also be predicted that if this regime was to be introduced into the current urban environment, then lines of routes that can be individually sold to franchisees would have to be developed whilst still in
Page 89 of 259
government ownership. This would have to happen before a Community Franchise system could allow the individual and separate lines of transport to be successfully acquired and financed by privately owned franchisees. However, in this sense, it is no different to the recent definition of the seven railway lines of the Sydney metropolitan region rail network and the intended sale of at least one of these, the North-West line, to private interests to run: it would be a process that is quite feasible for the current government owner of bus lines to undertake. Property rights require this exclusivity of line ownership and this has financial and practical consequences. For instance, a line that travels through the CBD of any city with much employment would be able to attract many more passengers than other lines with less demand. This will affect the value of that line and the amount of capital that can be invested to move the number of passengers who would want or need to use it. It also affects users with an alternative travel strategy that would significantly affect waiting times. A passenger in Circular Quay in the Sydney CBD would no longer have to wait for the bus going all the way to their specific destination, and in the process, letting up to 26 other buses leave that go to alternative destinations. That passenger would be able to board the first bus and travel on this until it got to the intersection with the line going to their own destination. Such an arrangement may have less appeal because of this need for changing lines but alternatively may have greater appeal because the first bus is the ‘right’ bus and waiting would be substantially lessened. This would depend on personal preferences and their relative valuations of the speed of boarding compared to the requirement to change lines. The time currently spent waiting now for the ‘right’ bus, will be compensated for by just getting the first bus. Any savings will be partly or completely dissipated on a change of line, with the calculation of its benefit being dependent on the half time of the headway compared to the half time spent on interchanging. However, this is an arrangement that works in other cities, the way that the transport layouts of Sydney are potentially being developed in any case and is the way that this hypothetical Community Franchise regime is likely to operate. 5.3.15 Commercial Pressures on Franchisees The only mechanism for the franchisee to increase its gross income is to increase the overall movement of people on its line. It is therefore in the interests of franchisees to continually monitor the attractiveness of their lines as this affects the supply of income from the franchisor based on the passenger numbers using their line. As well, every improvement that would save operating expense would be attractive to the franchisee as it could increase net income without any increase in patronage. These mechanisms would, as reported in the Dolomites, stimulate investment in improved conditions or equipment and/or adjustment in procedures or equipment to minimise the cost of operations. The incentive for the franchisees is built into this arrangement for they will have regard to the usefulness, attractiveness, practicality and efficiency of any line. The owners of lines would be under continual pressure to attend to all these matters and they have both the incentive and the opportunity to do so.
Page 90 of 259
5.3.16 Review of Payments to Franchisees The fares would be reviewed by the franchisor and adjusted periodically to consider overall demand, any material differences in the cost of operations, marketing concerns and other matters that may affect them. The setting of fares would be a judgement of the franchisor and would be possibly the most important decision that is made in the whole business arrangement. Franchisee response to this setting would be of critical interest and would affect the setting in further revisions. Raising the rate would encourage investment, lowering it may curtail investment and hurt existing franchisor profitability. On the other hand, lowered rates to franchisees would mean lower fares to passengers and may increase the volume of passenger numbers. The effects of adjustments to these variables could be expected to be closely monitored by the franchisor. This arrangement of a simple rate per passenger/distance ensures that all operational savings are retained by the franchisee operator, which becomes a significant incentive to run each line in the most economical fashion and gives scope for operators to trim costs if capital expenditure, for instance, may have not been as effective as thought. The payment to the franchisee for each passenger using a line would be estimated by the franchisor to be enough for the franchisee to make a reasonable return on the capital invested to run the lines in its possession. In this sense, the alignment of interests between the franchisor and the franchisee is coupled: they both want a profitable system, but this can only be profitable to the extent that it is still attractive for passengers to use it and franchisees to invest in it. In the same way as the franchisees would need to monitor the wellbeing of their own lines, the franchisor would need to monitor the wellbeing of the whole system. The only effective mechanism available to the franchisor to increase its gross income is the level of payments to franchisees of which it receives its percentage. Setting the levels of recompense to franchisees becomes a vital consideration the level of which over time could be expected to be refined and improved. The two most important and contradictory factors in setting the fares at the appropriate level are the need to attract paying passengers from other forms of travel, and the amount of interest the rate generates in franchisees. A low rate would attract passengers from other modes to public transport. A high rate would allow franchisees to invest in new infrastructure. There may also be a further consideration of the fare levels compared to that of other cities if there was a desire to make Sydney more attractive as a destination for living, working and tourism. This would be a State Government concern that may affect its desire to increase overall investment and its subsidy for operations. This is a set of circumstances that is finely balanced but will ensure that the rate set for fares is one that is ultimately in the franchisor’s best interests as well as in the interests of the other parties, i.e. the franchisees and the passengers; all parties share a common interest. These considerations would not be expected to result in frequent changes but might be expected to be altered on say a yearly basis, or as needed.
Page 91 of 259
5.3.17 Review of New Proposals It would be in the interest of the franchisor to review all new proposals that are made by existing or new franchisees as this is the only way of increasing revenues; the franchisor is prohibited from undertaking the service delivery itself and must rely on franchisees to generate new business. The franchisor is established as a profit-making corporation, albeit one that is totally government owned, and would have the incentive to maximise its revenues by allowing as many viable franchisees as possible to be given the chance to operate. If follows that the franchisor would have an incentive to engage with existing and prospective franchisees and provide any assistance in organising line routes, providing advice on financing, operational matters, etc. anything, in fact, that would heighten interest and awareness in the possibilities of franchisee participation in the network. 5.4 Analysis of Essential Aspects of the Defining Algorithm To explore the integrity of this quite specific package of controls applied to urban public transport, elements of the defining algorithm will now be investigated for their sensitivity to change. This section is based in part on the work done to prepare the journal article emerging from the 2015 Thredbo Conference (Emerson et al., 2016). There are certain elements of the defining algorithm that would not withstand substantial change without altering the fundamental workings of a Community Franchise. For instance, if it were not related to a specific geography, it would no longer have claim to being a ‘Community’ organisation. Some of the elements of the defining algorithm are more sensitive than others to change. Removing the ownership element of the operational business from the franchisees would have the effect of undermining the essential nature of a Community Franchise. Without long-term equity ownership of the asset by the franchisee, there is no inherent value of the business recognised by financial institutions to be mortgaged to facilitate the creation and maintenance of value by the franchisee. Ownership in perpetuity is a fundamental difference to the other regimes in the Thredbo comparison which also compromises the necessity for franchisees to be vitally fixated on maintaining an attractive and effective service provision. Long term asset ownership could thus be considered immutable. Another example is if the franchisor were to become exclusively responsible for asset allocation. Were this to be the case, there would be no opportunity for the franchisees to independently experiment with different solutions to the defined problem. This would be a move away from one of the key differences between traditional centrally planned systems and those that rely on disbursed franchisee initiative. Having this form of franchising could thus be considered essential to it being a Community Franchise and is a control that could not be relaxed. This also is a fundamental difference to the other Thredbo definitions of regimes. Apart from these essential elements of the definition, the effect of relaxing other algorithm controls is variable. Clearly, the standards of service operation are matters that would be under constant review to ensure that those services to be provided are affordable and appropriate for the levels of risk and say comfort or other measures that
Page 92 of 259
are thought the community being served requires. The elected representatives of the controlling franchisor are sensitive to these factors in the same way as Castro and Maddock (1997) demonstrated with the pricing and operations of the government- controlled Australia Post. But there could be either more or less of this oversight of standards and the regime would still function in essentially the same way. Likewise, the monetary value to the community of the service being operated will be under constant review. This is necessary to ensure the appropriate level of monetary stimulus that is on offer will attract the necessary commercial activity for service provision. These types of adjustment are central to functioning and alterations to their specific values do not call into question the regime operations; the frequency and accuracy of such a value analysis of the service would not alter the basic functioning of the regime. Such matters, and the history of these regime reforms, are prospective areas for further research. 5.5 Significant Distinguishing Features of a Community Franchise Regime: This definition and the review of other regimes of public transport, plus the analysis of the sensitivity to an adjustment of regime features, enables a picture of the differences in regimes to be readily identified. A summary of the major differences is as follows:
• Control of all fares, operational times and frequencies, and equipment standards are clearly with the government in a Community Franchise. This makes it different from say UK operations where prices and frequencies are set by the rail and bus operators. • Property rights; A franchisee in a Community Franchise owns the line it operates as property equivalent to that of real estate. Private bus operators in say Sydney do not ‘own’ the lines on which they operate a service. These services in a typical government enterprise are normally tendered out on an infrequent but regular basis, normally in the order of ten to fifteen years. Property rights may be present in that they have been awarded a contract for a defined operation, but this is for a limited time period and is not able to be sold on to another operator. These operations have most in common with what is described in other contexts as sub- contracting. Operators in ‘colectivos’ and other paratransit systems, do not have property rights over routes; they compete with all other providers to service the market and can use any route to achieve that. • Government Function: The role of the government owner in a Community Franchise is limited and is excluded from actually providing the service itself; it has to rely on the franchisees to achieve its aims. Other regimes have varying roles for the government, and this can range from full provision of the service by government agencies to just planning the routes and times, and allowing private enterprise to tender for the defined services or opportunities within certain time slots and routes without specification of specific times or standards.
Page 93 of 259
5.6 Summary of Regime Algorithm Application The regime as defined from observations of ski lifting operations in the Dolomites is discernibly different from the existing regimes described in this thesis as Government Enterprises or as Open Competition. From the initial observations and analysis, it appears quite able to be applied to urban public transport. The question as to if this is in fact the case cannot be reasonably assessed by the party proposing its application. Not only may that person be too close to the ideas, but the regime’s performance is subject to a vast array of forces too complex for any individual to reasonably foresee. However, several aspects can be examined without a full-scale application of the regime to find out how it may operate. It is evident from comments received from colleagues and conference participants that there are questions about the social desirability of private ownership of public transport assets and the way individuals would steward these assets in a situation of competition and rivalry. As well, there has been enormous variation in the range of opinion as to how the regime would function and this is, as mentioned, not something a single individual can usefully illuminate. The next chapter will examine the research questions that emerge from this hypothetical construct and to state the hypotheses that are proposed for each question. This will enable the methodology of this study appropriate to these questions to then emerge.
Page 94 of 259
CHAPTER 6 RESEARCH QUESTIONS AND HYPOTHESES 6.1 Chapter Content This chapter will identify the research questions to be posed and the hypotheses and expected answers to emerge, following consideration of the literature review and from having constructed the hypothetical application of the Community Franchise regime to urban public transport in Sydney. 6.2 The Scope of Inquiry The underlying questions of the thesis can be expressed broadly as to if the Community Franchise regime delivers better outcomes for government in public transport than existing regimes. Other aspects of this overall question are if the Community Franchise regime applied to public transport allow for an alignment of interests between the government and the service providers such that both parties and the public benefit from the arrangements? There is a need to answer whether this business operation could work to produce a coherent transport network from a collection of competing individual operators. This is allied to the question whether these individual operators, being focused on their own welfare, could produce a network that was in the passengers’, i.e. society’s, best interests. Finally, there is the overall question as to whether such a regime with this collection of individual franchisee owners and a government-owned franchisor could have relevance to urban public transport and be applied to that purpose. The considerations to this point reveal several research gaps and the ones that are to be addressed in this study are: 1. the objective performance of the regime to satisfy financial and social objectives, 2. the behavior of competing individuals in a transport network when questions of upgrading shared infrastructure need to be addressed and agreed upon, and, 3. the way that a Community Franchise regime may have relevance to urban public transport. These general concerns are framed into specific research questions to enable qualitative and quantitative methodologies to be used to undertake an appropriate examination. 6.3 The Research Questions 6.3.1 Measurement of Quantitative Differences between a Community Franchise and a Government Enterprise regime Measuring the quantitative differences between the operations of a Community Franchise regime and a Government Enterprise regime is a further aspect of the fundamental question. It is assumed that government agencies planning and operating public transport networks are seeking to maximise social welfare. However, in a Community Franchise regime, each individual line-owning franchisee will be seeking to maximise its profit and is unlikely to consider the greater social good. This question is accessible through a consideration of social welfare, being the sum of producer surplus
Page 95 of 259
and consumer surplus. The producer surplus is an obvious component, is easy to measure as net profit of operations and is an important component of both government enterprises and commercial operations. The consumer surplus is likewise a figure that can be calculated from a consideration of the utility of travel, a figure able to be derived from widely accepted measures of these figures. A relevant measure of social welfare can be established by the addition of these two figures. If it can be established that a Community Franchise regime can produce similar social welfare to that of a Government Enterprise regime, then the case for use of a Community Franchise regime on the grounds of the adverse effects on society, cannot be made. Testing for social welfare is thus an appropriate way of comparing both regimes. It is proposed to examine the question through a consideration of what effects the different objectives of the operators in this regime have on social welfare. This will examine the effect of selfish competition between the operating entities of the network compared to a social welfare focused approach of government planned operations. This research question can be expressed as: What is the impact on social welfare of a group of profit-maximising franchisees compared to a single purpose centralised organisation seeking to maximise this measure? 6.3.2 Cooperation between Competing Line Owners The literature does not directly address whether a competitive transport regime when all the lines are separately and individually owned, would be prejudicial to the creation of a network that was efficient for the whole community. However, this is a view expressed by several commentators when a Community Franchise regime has been presented at internal university seminars. This is an area of importance and needs to be addressed. The question that arises is whether the expected selfish behaviour of individual line owners has a negative effect on the measure of social welfare. This research question is expressed as: Are competing line owners able to cooperate so as to enhance network performance? 6.3.3 Is a Community Franchise Regime transferable to Public Transport? This is a fundamental question for this thesis. What are the advantages and disadvantages that such an adaptation of a Community Franchise regime would bring to public transport? Would this Community Franchise regime, if it shows promise of being relevant to public transport, be able to be applied to this urban task? This research question is expressed as: What aspects of a Community Franchise regime are transferable or could be adapted to urban public transport?
Page 96 of 259
6.4 Research Questions and Hypotheses These questions with appropriate hypotheses and expected outcomes are formulated as summarised in Table 3.1 below.
Research Question Expected Outcomes or Relevant Hypothesis
1 What is the impact on social Hypothesis 1: The Null Hypothesis is that welfare of a group of profit- both regimes will deliver similar amounts of maximising franchisees social welfare (SW), i.e.: compared to a single purpose centralised organisation seeking H0: Mean GE (SW) = Mean CF (SW) to maximise this measure? H1: Mean GE (SW) ≠ Mean CF (SW)
2 Are competing line owners able Expected Outcome 2: That in a game to cooperate so as to enhance context, there will be sufficient cooperation network performance? to provide an acceptable allocation of resources compared to that which is achieved through central planning.
3 What aspects of a Community Expected Outcome 3: The hypothetical Franchise regime are application of a Community Franchise transferable or could be adapted regime to urban public transport can be to urban public transport? demonstrated to have a reasonable chance of success.
Table 15 Research Questions and Hypotheses 6.5 Summary This chapter has identified three research questions that emerge from the original observation of the ski lifting business regime of the Dolomites, its hypothetical application to urban public transport and the literature review of academic study on the subject. The reactions of those hearing of this business model have indicated the need for some experimental evidence to establish the way that the business regime may perform practically and financially and will react to situations requiring cooperation and network planning of competing individuals. The fact that the observations are of a regime in an industry that some regard is only tangentially related to urban public transport, suggests the need for a process of investigating the applicability of the regime to an urban environment. With a definite set of questions, hypotheses and expected outcomes being in place, a methodology will now be outlined that allows these questions to be addressed and tested. This is investigated in the next chapter.
Page 97 of 259
Page 98 of 259
CHAPTER 7 METHODOLOGY 7.1 Chapter Content This chapter describes the several different approaches used to examine the three research questions posed on aspects of the Community Franchise business regime. The approach to these different methods will be outlined here and extended as necessary in each chapter dealing with the different research questions. Quantitative differences to resource allocation between a traditional government enterprise regime and a Community Franchise regime will be examined by a computer game. The game will also enable observation of cooperation between competing players to directly test the impact that a lack of cooperation might have on the sharing of infrastructure in the competitive situation between different transport lines in a Community Franchise. The overarching objective is to examine whether a Community Franchise regime could be an appropriate application of some benefit in a public transport setting, and this question is to be examined through a survey of experts. This chapter is organised into three sections relating to these complementary methods, that of the computer experiment (as an application of experimental economics – Q1), observations of cooperation (Q2) and an expert assessment of a hypothetical application of the regime to an urban situation (Q3). 7.2 Research Question One Research Question One states: What is the impact on social welfare of a group of profit-maximising franchisees compared to a single purpose centralised organisation seeking to maximise this measure? The approach to examining this question relies on the fact that aspects of running commercial enterprises can be quantified, in particular producer surplus and passenger numbers. This characteristic enables numerical comparisons to be made between the Community Franchise regime and the Government Enterprise. A computer model of an artificial city was constructed to allow for numerical aspects of an urban public transport system to be quantitatively tested experimentally, as follows. 7.2.1 Experimental Objectives The problem to be investigated deals with potential sub-optimisation of operations, where individual line owners in an urban public transport context may maximise their own profit to the detriment of the overall optimisation of the network giving rise to lower producer surplus and/or lower passenger numbers, and thus lower social welfare. Social welfare has many meanings depending on the context. It is defined popularly 11, as “..the study of how to distribute income in order to achieve social good”. In the Meriam- Webster dictionary, it is defined as “..a branch of economics dealing with
11 Web search 2019 https://financial- dictionary.thefreedictionary.com/Social+welfare+(economics)
Page 99 of 259
human welfare, the defining of wealth, and the establishment of guides for social policy aiming at the maximization of total individual utility”. But the meaning in the context of this thesis is that specific usage in the discipline of economics used by Tirachini et al. (2013) and described there as being the “…sum of producer and consumer surplus”. This is a clearly definable and derivable figure for many aspects of the economy, including, as Tirachini et al. show, for public transport. The object of the experiment was thus to see if this measure of social welfare of the whole network would be adversely affected by individual line owners working towards the potentially conflicting objective of profit maximisation of the lines in their individual ownership. This was tested by having experimental subjects (the participants) engage with a computer model of an artificial city set up to operate as a Community Franchise business regime. In this ‘game’, participants had to optimise the operations of their portion of the transport network, i.e. to maximise the profits of the lines in their ‘ownership’. The figures for consumer surplus, passenger numbers and social welfare were recorded at the same time. All figures were compared with the same computer model set up for a Government Enterprise, with a single operator of the system. In addition to the different human efforts to maximise either social welfare or profits, a comparison set of figures for producer surplus, consumer surplus, passenger numbers and social welfare, were generated by the genetic algorithm Evolver working within the Excel spreadsheet in which the model was constructed. The objective of the algorithm was set as firstly the producer surplus and then as social welfare, to give perspective to the comparison. This comparison enabled an examination of profits and social welfare of the two operating business regimes. It realistically combines the two important measures of public transit network performance: the utility to passengers and the economy of operations. 7.2.2 Game Theory Considerations It was noted by Evans and Holder (1985) in considering bus deregulation that simple games used in their study do not allow for an understanding of the complexity of real-life and that “…the strategic complexity of games increases astronomically with the number of decision options and the number of players.” However, he recommends the use of more complex games as a way of throwing light on aspects of real situations, but that these should be simple enough to understand. This approach has been followed in this experiment: the games cover just sufficient of those factors which enable the private profit/public welfare question to be addressed but is so much less complex than reality. The level of complexity in decisions available to participants was adjusted in the final version from earlier versions that had much more decisions able to be applied to ‘improve’ services and attract patronage. The more complex versions took an unreasonable time for volunteer participants to spend arriving at their best solutions. The final version enabled play to conclude in about 20 minutes with clear stable situations emerging.
Page 100 of 259
Game theory would suggest that there is likely to be a Nash Equilibrium (NE) for the experiment. The NE can be described in various ways, two common ways being as follows:
• NE is a state of the system involving the interaction of different participants, in which no participant can gain from a unilateral change of strategy if the strategies of others remain unchanged (Oxford English Dictionary). • NE is where there is no incentive for a player to deviate from an initial strategy if other players remain constant in their strategies (Cornell University (Maths)). The complications inherent in the rules of operation and the sheer number of alternatives in this game would make it difficult to determine a NE as well as establish if there were single or multiple equilibria. If there is a mathematically determinable equilibrium, (and finding one was not attempted), it was not indicated from the results of the experiment. All games had similar outcomes in that certain lines would dominate, and some would have consistently limited patronage, but nearly always with different social welfare figures. All games ultimately reached a point of stability where no further move was desirable for any participant. But there were virtually no games that had identical outcomes. This was a situation that was common to both versions of the game, Government Enterprise and Community Franchise. This complex indeterminant situation was for a very simple set of possibilities, significantly simpler than early versions of the game tried before settling on the final version. As well, it is vastly simpler than that of the real world. More complex circumstances in any application would result in a more varied set of possible outcomes. It can be reasonably deduced that outcomes would differ for the separate lines if the rules of operation were to change, e.g. rates for the carriage of passengers. This can be done by the Franchisor as necessary and as it might desire. In real life, outcomes would also change due to the strategies of individual participants who can choose to run their lines with different operations, different levels of equipment and different attractions being served by the line. This would be a constant feature of competitive interaction between franchisees as the most economical and the most attractive ways of operating are experimented with on the separately owned lines. This would suggest that finding an optimum solution by calculation would be at best an uncertain estimate and likely, in any case, to change over time. It was in this context that the use of the optimum outcome was sought by the use of Excel’s inbuilt optimiser. Finding this figure for the specific circumstances of the experiment would enable this stable state to be compared to the human attempts, in both regimes. It will be explained in section 8.2.4 that the inbuilt optimise was not capable of performing this very complicated task and the add-on programme Evolver, a genetic algorithm, was used to obtain estimates of this figure. Predicting the effects of changes can be argued to be a desirable thing to attempt for the franchisor deciding on rates to provide to franchisees, or by government and the franchisor deciding fares to charge passengers. But it is not likely to be any more
Page 101 of 259
accurate than attempts by government to predict outcomes of any complex economic situation, so would need to be interpreted accordingly. It is noted that the Evolver solution was always higher than those of humans, (but different on different runs), but appears to be a reasonable indication for choice between different options and a possible way of assisting in the prediction of outcomes of change to any of the operational parameters. Details of how these two contrasting regimes were tested using a computer experiment follow. 7.2.3 Experimental Model The theoretical context for the experiment/game within an artificial city was based on a “Synthetic Task Environments” approach which provides a research platform that “… bridges the gap between controlled studies using artificial laboratory tasks and uncontrolled field studies on real tasks or using high-fidelity simulators” (Cooke and Shope, 2004). Synthetic tasks are “…research tasks constructed by systematic abstraction from a corresponding real-world task” (Martin et al., 1998). Others give details on experiment design (Dietz et al., 2013) e.g. the Event-Based Approach to Training (EBAT). Although there was some complexity in the tasks that subjects had to perform, (which take it beyond a simple ‘laboratory experiment’) these had as few variations as possible (and thus were not a realistic simulation of a real situation). Doing this minimised the ‘noise’ within the experiment. The choices available to participants enabled the differences between a direct action top-down controlled approach of the Government Enterprise, (where one subject can control all decisions), and the bottom-up undirected approach of the Community Franchise regime, (where several competing subjects control only parts of the overall system and must optimise their own affairs without regard to the overall functioning of the system), to be observed and measured. This is directly analogous to the top-down government-directed public transport systems of Sydney and the individual ownership of the lines of transport with no direct influence of any one body to optimise the system, as occurs in the Community Franchise regime of the Dolomites ski lifting transport network. The program itself is described in more detail in Chapter 6. A Multinomial Logit Model (MNL) allocates the mode and routes to be taken by each passenger in maximising their utility in satisfying their travel demand. The MNL model is used to allocate passengers to the various lines based on travel and waiting times, crowding, fares, line transfer reluctance, etc. so that their choices as to the routes to travel from one origin within the city to the predetermined destination can be realistically modelled. This MNL model resides within the spreadsheet to provide the financial outcome of the various decisions that can be made to adjust the model. 7.2.4 Experimental Authorship The basic concept of the game, the inputs and graphic interfaces, determination of capacities, performance and financial aspects of the system have all been the unique
Page 102 of 259
contribution of the candidate. However, Professor Michiel Bliemer, a co-supervisor of this PhD study, is acknowledged as the principal author of the allocation of the demand side of the experiment, in particular the use of the MNL model that provides for the realistic allocation of passengers onto the transport lines of the artificial city as well as the calculation of consumer surplus. Without the allocation of the passengers in a manner that was realistic to those playing the game, the degree of realism achieved, or the acceptable measure of the social benefit of each regime, would not have been possible. 7.2.5 Experimental Participants Participants for the experiment were chosen with specific criteria. Firstly, players with particular expertise in public transport planning and operations were sought. This criterion was adopted due to a prevailing understanding of the relevance of professional planning at the Institute of Transport and Logistic Studies (ITLS), the institution at which this study was conducted. Many connections to industry and government were available through the ITLS, thus securing professional participants was seen to be an achievable objective. To this end, participants were recruited from the government department responsible for this function, Transport for NSW (TfNSW) and from the professional organisation of the bus operators, BusNSW. To supplement this core group of experienced professionals, a further group was recruited from the academic community of the ITLS, including PhD students who are involved in studies of their own, all with a transport focus. Additional participants representing a non-professional group, necessary to achieve experimental numbers, were recruited from those who have or had a background in the professions of any sort; from teachers, engineers, chemists, actuaries, psychologists, etc. with a further group coming from business people who have demonstrated a level of success in their working arrangements. The participant characteristics, either transport professional or not, and their age grouping was noted for all recruits. This approach was embraced as a way of testing the business regime as there are parallels with the operations of the Community Franchise regime. This regime does not rely on professional planners to suggest new lines or investigate potential cost benefits from any particular line configuration or modal decisions: it is purely the entrepreneurial activities of interested sections of the business community using their own intuitions and judgment. Using both the professional planner and ordinary members of the public was seen as a way of measuring the potential of both groups to undertake planning and to make a direct comparison of a centrally designed network with the evolution of the unplanned Community Franchise regime network. The above outline is a brief description to give the conceptual basis of the experiment. A complete description of the computer model, the procedures of play and all aspects of the collection of results and their analysis are presented in the chapters on the computer game.
Page 103 of 259
7.2.6 Hypothesis Testing The experiment was run a sufficient number of times to obtain statistically significant results. The data collected were analysed to enable a comparison of the social welfare generating capacity of both the Community Franchise and the Government Enterprise regimes. Hypothesis 1 outlined in Chapter 3 reads: The Null Hypothesis is that both regimes will deliver similar amounts of social welfare. The null hypothesis tested was that the mean social welfare (SW) generated by the Government Enterprise (GE) centrally planned model will be equal to that which is generated by the Community Franchise (CF) regime with its uncoordinated, selfish profit- motivated operators. This was expressed as:
H0: Mean GE (SW) = Mean CF (SW)
H1: Mean GE (SW) ≠ Mean CF (SW) The results are presented and discussed in Chapter 7. 7.3 Research Question Two The other aspect to be explored in the computer experiment was how competition between the individual line owners affects their propensity to cooperate when the need for this arises over shared infrastructure. This was established to be a question of some significance given the complexity of Sydney’s transport networks and the evident need for cooperation for effective functioning. The question was raised as to how would competing participants co-operate or does the situation of multiple owners present a barrier that only explicit direction from a central authoritative body can resolve: i.e. would competitors actually cooperate in ways that may allow the overall system to evolve as an efficient overall network and not just for the benefit of a particular owner? These questions were formalised as follows as the Second Research Question which was stated in Chapter 6 to be: Are competing line owners able to cooperate so as to enhance network performance? The expected outcome for this question was: That in a game context, there will be sufficient cooperation to provide an acceptable allocation of resources compared to that which is achieved through central planning. This is a question that was explored using observations of the participants in the computer experiment. The observed participant behaviour was analysed and commented on. Of specific interest was whether participants forgo selfish benefit for the good of the whole system and whether or not this matter in the production of social welfare. A complete outline of the method is given in Chapter 8 when the experiment is described.
Page 104 of 259
Observations of this aspect of the computer experiment enabled an answer to be given to this research question and the expected outcome compared. 7.4 Research Question Three Research Question no 3 states: What aspects of a Community Franchise regime are transferable or could be adapted to urban public transport? The expected outcome of this research question is: The hypothetical application of a Community Franchise regime to urban public transport can be demonstrated to have a reasonable chance of success. This question was explored by an Expert Survey, an asynchronous anonymous computer-based survey, as a means of establishing multiple viewpoints from different relevant perspectives. 7.4.1 Available Methodologies for Assessment There are several accepted ways of approaching a qualitative evaluation of the hypothetical application of the business regime to an urban area. One approach is to organise expert commentary through workshops and focus groups of relevant experts or interested participants. This is a technique that has proven useful in identifying factors that are not obvious to all the attendees to those forums or to the researchers. However, there are practical difficulties in arranging the simultaneous location of all those who could or should be involved. It also suffers from the tendency of ‘group think’ where individuals’ viewpoints can be overwhelmed by the majority or dominant thinking to the detriment of a collected balanced view (Linstone and Turoff, 2011). These disadvantages of the workshop approach are effectively overcome by tools which can canvass the required breadth of expertise in ways that keep individuals separated by place and time. Linstone and Turoff show that this is effectively done by the use of computer-mediated asynchronous communications. This ensures that participants are not influenced by the opinions of others but provide only their own point of view. It enables people to participate in their own time and they are not bound by the necessity to be present at a specific workshop event. This approach suits the particular circumstances of this research project. This use of a technique to canvass a broad range of relevant expert opinion is a supplementary tool to provide perspective to the case study and the experiment and to directly address the third research question of the applicability of a Community Franchise to the urban environment of Sydney. It is thus not attempting to provide a comprehensive or conclusive view on this matter, but one that brings additional points of view to what is essentially a question without definite answers. The intention is to both get a wider view of the suitability of the regime to the urban setting, i.e. a response to the overarching research question, and to establish the types of problems that may be encountered if a Community Franchise business regime was to be applied in an urban setting.
Page 105 of 259
7.4.2 Asynchronous Anonymous Computer-Based Surveys The advantages and disadvantages of both workshop/focus groups and AAC’s can be summarised as follows.
Technique Description Advantages Disadvantages Workshop/ The research idea is Able to be applied Overall view able to be presented to the where there is a swayed by dominant Focus experts who then can convenient collection of individuals; no guarantee Group comment on the experts, e.g. at a of any balance due to application and conference, or at the dynamics of group ‘brainstorm’ to obtain request of the situation. A ‘one-shot’ viewpoints more researcher. Multiple approach needing to diverse than for any approaches and points assemble all experts at individual. of view. the one time and location.
AAC A disbursed expert Well tried and May not establish a survey group who academically accepted definite outcome as there anonymously and technique, flexible is no opportunity for asynchronously structure, responses consensus to be easily comment on written untainted by ‘group achieved. questions, in a think’, wide group of structured series of potential participants, rounds with statistical analysis, responses computer-based, and coordinated by the not requiring researcher. coordinated responses.
Table 16 Comparison of Techniques for Group Evaluation of Complex Ideas The topic of this thesis fits the critical criteria for the use of an AAC survey. The Community Franchise business regime is not present in urban public transport, but it has similarities to existing regimes operating in the closely related field of fixed line ski lifting. Its proper assessment requires expertise in various unrelated areas, e.g. franchising, network design, commerce, government, politics, finance, etc. and thus requires a diverse range of opinion in various areas of knowledge. The appropriate participants for several of these categories are people employed at a high level in commerce and business and would not necessarily have the time to attend a workshop but may be able to attend to the lesser task of computerised responses to a survey. It thus suits the timeframe and the logistical/organisational constraints of the variety of different experts for business regimes of public transport. The use of computer-based techniques allows for the practical implementation of the technique to the range of experts who would be requested to participate. The most recognised method of an AAC is the Delphi survey. A Delphi survey and its variants are described as an asynchronous, anonymous, computer-based (AAC) survey (Linstone and Turoff, 2011). This uses a similar range of diverse expert opinion to a
Page 106 of 259
workshop but in a way that eliminates both the ‘group think’ tendency and the locational difficulties. A Delphi survey is a well-used flexible research tool appropriate for complex situations where there is incomplete knowledge (Benarie, 1988) The technique uses a structured process for collecting and distilling knowledge from a group of dispersed experts through a series of questionnaires interspersed with controlled opinion feedback (Gunaydin, 2006). Hsu (2007) summarised the objectives of the Delphi technique as follows: • To determine or develop a range of possible program alternatives, • To explore or expose underlying assumptions or information leading or different judgements. • To seek out information which may generate a consensus on the part of the respondent group • To correlate informed judgements on a topic spanning a wide range of disciplines, • To educate the respondent group as to the diverse and interrelated aspects of a topic. These researchers show that Delphi surveys are a good tool where a goal is to improve our understanding of problems, opportunities and solutions. Gunaydin (2006) and Linstone and Turoff (2011) also point out that its greatest use is in defining problem areas, of identifying disagreement, even more so than in establishing consensus, an often-nominated benefit of the technique. Far from being a failure of the technique, when consensus is not achieved, it identifies potential problem areas and enables appropriate attention to be directed at those areas in further work on the topic. The Delphi survey techniques themselves are diverse and flexible and will encompass many variants of the traditional form. Skulmoski et al. (2007) show that there is a very wide application of the technique, many of which vary from that they describe as the “Classical Delphi” but still adhere to the basic concepts of having a moderated expert communication without direct contact between participants over a nonsynchronous extended time frame. They endorse the use of non-classical Delphi surveys because of this demonstrated flexibility and effectiveness in the wide range of applications that they observed. They concluded that the Delphi approach can be aggressively and creatively adapted to particular situations. The many variants to the traditional survey arrangement have provided useful results when applied to groups of respondents from as small as four people to many hundreds of people, and when the number of rounds was restricted to as few as one. A modified Delphi survey, taking advantage of the very wide range of applications of the technique and based on that concept of a survey that is asynchronous, anonymous and computer-based, would appear to best suit the circumstances of this research project. It is an appropriate technique to explore the possibility of the business regime being applied to the field of urban public transport and as well to suggest any potential problem areas and barriers to its application.
Page 107 of 259
The technique to be used and described in the next section is a subset of the Delphi technique and can be described as an Expert Evaluation survey. This term will be used hereinafter. 7.4.3 The Expert Evaluation Method The objective is to get a range of relevant opinion on and to identify obvious impediments to the urban application of a Community Franchise. The respondents will be drawn from those with skills in the range of appropriate ‘dimensions’ of the problem. Experts will be sought who will be aware of public transport systems, who are involved in commerce, franchising, in particular, the governmental peculiarities of the regime application and others to be identified below.
7.4.3.1 Dimensions The dimensions to be covered must include all relevant aspects of the problem. It is assumed that persons having a good understanding of say government, do not necessarily have an equal understanding of business, or of franchising as a particular form of commerce, or of, say transport. But each of those separate discipline areas will make a relevant contribution to understanding aspects of the application of the regime that will assist in building a picture from different perspectives. Those disciplines areas thought to cover the essential and relevant dimensions of the problem, and the areas from which potential participants will be recruited were determined to be as per Table 17 below:
Dimension Discipline/Areas Franchising Franchisors, Franchisees, Professional Associations, Academia Government State Government, Local Government, Academia Finance/business Accounting, Business, Academia Planning Government, Industry, Academia Operations Business, Transport Industry, Academia Table 17 Dimension of Inquiry and Discipline Areas for Recruits With the opinions of experts from these five dimensions, a more relevant picture will be provided than if the views of only some of these expert groups were consulted. This more rounded view will then allow the overall questions of the applicability of the Community Franchise regime’s application to urban public transport to be more reasonably assessed. The questions for these dimensions are developed in Chapter 10.
7.4.3.2 Respondents The literature is quite varied in its treatment in establishing the number of respondents to be sought. Skulmoski et al. (2007) in looking at successful, published graduate research using the Delphi method, lists examples with sample sizes from 3 to 126 persons and with the number of rounds between 1 and 3. There are many Delphi surveys with only a limited number of respondents and Hsu and Sandford (2007) state that concerning the number of respondents “….that researchers should use the “…minimally sufficient number of subjects” and look to using follow up research to verify findings. They also conclude that this is “…generally determined by the number required to constitute a
Page 108 of 259
representative pooling of judgments and the information processing capability of the research team”. These guidelines suggest that the number of respondents in this graduate research case should be several from each respondent category to respond to each of the relevant ‘dimensions’ identified. There should be as many as can be generated from approaching representatives of all the categories of people identified as the larger the number the more representative the sample will be. Accordingly, the objective was to get several representatives from each of the 6 dimensions. Choosing the respondents or the participants is a structured procedure (Okoli and Pawlowski, 2004, Hirschhorn et al., 2017). Firstly, the important dimensions of inquiry of the survey were identified, as in the previous section. Appropriate groups of respondents to each of these discipline areas were identified and relevant persons were approached who fitted into the work categories. These categories of organisations contained potential respondents with backgrounds in transport, finance, business, franchising, government and researchers from universities and specialist think tanks. Multiple individual organisations were identified and contacted to establish if management would support one or more of their staff undertaking the survey. Snowballing from this group helped identify additional respondent organisations. When the prospective responsive organisations had been identified, the recruitment of individuals commenced. The head of the appropriate section in these organisations was contacted and it was requested that a qualified, economically literate member of their team be nominated and made available to answer the series of questions on the topic. No participants were sought from organisations that did not want to be formally involved. Contact details of the individual nominated, including direct email connection, were obtained from the responsible officer providing the opportunity. The email address was used to provide the questions and support material to those individual respondents.
7.4.3.3 Number of Rounds The number of rounds required is determined in part by the purpose of the survey and is a very flexible dimension of the technique (Skulmoski et al., 2007). Multiple rounds have sometimes been seen as a means of establishing consensus between participants. Linstone and Turoff (2011), two of the originators of the method, regard it as “…“a method for structuring a group communication process”, not “…a method aimed to produce consensus”. There is never a certainty that consensus between group members will eventuate, and different opinions on many matters are considered valuable in any case as it provides an indication of the range for any outcome. There is no certainty or ‘right’ answer to questions on theoretical constructs of the sort considered in this study and no consensus is thus sought. Fewer rounds were thus seen to be appropriate for this issue. The principal output of this survey is to identify concerns/barriers and positive elements for the implementation of a Community Franchise in the urban setting. The widest range of opinions would be the most valuable.
Page 109 of 259
Hirschhorn et al. (2017) found that a survey structure with three rounds to target specific objectives was effective. In that survey, the objective of the first round was to ‘brainstorm’ so that participants contribute to the list of issues seen to be relevant to the basic question. This list is then ‘narrowed down’ in a second round in which feedback is provided to the participants and responses are sought to reduce the number of issues and their responses. In a final round, this series of reduced issues and responses are ‘rated’ by participants and this provided authoritative rankings. The ‘brainstorming’ round in the Hirschhorn study was seen to be a very relevant contribution to that survey, but for this study, the generation of these issues is seen to be well canvassed in the earlier sections of the thesis and better dealt with by the questions of the participants as to ‘critical features’ that may effectively limit the application of the regime. The free form responses will draw out those aspects which are thought to be essential or critical to regime function in Sydney. Multiple rounds also have problems with attrition. The case study and the literature review effectively suggest the issues which needs to be addressed in this study and it was decided to limit participant consideration to the answering of questions generated by the study author. It was, however, useful to allow participants to see and respond to the range of answers provided by others. Accordingly, a second and final round was included to allow for a response to the curated, consolidated response to all the questions.
7.4.3.4 Survey Respondent Preparation Not all potential survey participants were familiar with a Community Franchise business regime or had any prior interest in it. When the attention of the participant was secured, they were provided with a single document that contained a description of the Community Franchise regime and its urban application. This regime description is provided in the Appendices. Following this description, the document then went straight into the fourteen questions. The questions were grouped into related areas of interest, coinciding with the identified dimensions. Each question was prefaced by a description of the issues pertaining to that group. At the conclusion of each set of questions, the participant was requested to self-assess their level of expertise in that area. The whole survey was estimated to take no more than one hour to read, respond to and return, all through the Word program and their computer interfaced with the internet.
7.4.3.5 Analysis of Responses The written free form responses of participants were analysed in NVivo. This enabled a detailed examination of conformity and differences of opinion and for common themes to be identified and commented upon. The responses to scaled qualitative answers were consolidated and analysed as to the average, standard deviation, highest, lowest responses and any consistency in the responses, using the self-classification of expertise to establish differences, patterns, consistencies etc. As well, the qualification of the respondent to comment on the particular questions was noted and commented on in the analysis. For example, those with an intimate knowledge
Page 110 of 259
of the business methods of transport were considered to give more reliable answers than those without that intimate knowledge as to questions about the possible business interactions of potential regime participants. This approach to the survey was reviewed by the Ethics Committee of the Sydney University Business School, (Project No: 2014/976) and the approval is attached in Appendix B. 7.5 Methodology Summary The hypothetical example of how a Community Franchise business regime could be applied to urban public transport was proposed in Chapter 5. This regime application applied to the city of Sydney was investigated by two distinct methods, namely a computer game and a survey of experts, to examine the three research questions. The computer game enabled a quantitative approach to be explored to examine the development of a network under two contrasting approaches to resource allocation, where participants had different objectives. The Expert Survey enabled the overall question of this thesis, the applicability of the Community Franchise regime applied to urban public transport in Sydney, to be scrutinised and commented upon by a range of persons expert in their own fields relevant to a broad examination of this hypothetical application. In the next chapter, the first two research questions, as examined in the experiment using the computer game, will be reported on.
Page 111 of 259
Page 112 of 259
CHAPTER 8 EXPERIMENTAL GAMING 8.1 Chapter Coverage This chapter describes and documents a computer experiment designed to examine possibly the most important characteristic of the Community Franchise regime, that of dispersed resource allocation, where uncoordinated franchisees make all decisions as to network development. The experiment was set up as a “game” for the quantitative investigation of the different business regimes controlling a transport system in an artificial city with a network of public transport lines. The game was designed to enable a direct comparison of the social welfare that results from the planning approaches of the two different business regimes, namely the Community Franchise regime and the Government Enterprise regime, typical of most urban public transport and of the integrated ski resorts, where all decisions on network development are taken by the central controller. It is reiterated that authorship of the game concept was that of the researcher, but that the associate supervisor, Professor Michiel Bliemer, is gratefully acknowledged as the creator of the demand model for the allocation of passengers to the various lines of transport and the calculation of consumer surplus resulting from the derived measures of utility, through his authorship of the multinomial logit model working to allocate passengers to lines of transport in differing conditions of wait time, travel time, crowding, fares and the measurable propensity to ‘stay at home’. A brief introduction to this experiment was given in Chapter 4 Methodology, and the next section provides a full description. 8.2 Detailed Methodology As outlined in Methodology, the experiment was conducted by having experimental subjects (the participants) engage with a computer model of the artificial city set up to operate as either, a Community Franchise business regime with multiple owners of lines, or as a Government Enterprise regime, with a single participant making all decisions as to resource allocation. In the Community Franchise version of the game, participants had to optimise the operations of their portion of the transport network and maximise the profits of the lines in their ‘ownership’. The figures for consumer surplus, passenger numbers and social welfare were recorded at the same time, but not shown to participants. All figures were compared with the same computer model set up as for a Government Enterprise, i.e. with a single planner and operator of the system whose stated objective was to maximise social welfare. For the Government Enterprise regime, the optimum resource allocation that maximised social welfare was independently established by the genetic algorithm, Evolver, as well. This experiment responds to the Research Question no 2 outlined in Chapter 3: What is the impact on social welfare of a group of profit-maximising franchisees compared to a single purpose centralised organisation seeking to maximise this measure?
Page 113 of 259
The null hypothesis is that both regimes, i.e. the Government Enterprise (GE) and the Community Franchise (CF) regimes, will deliver the same mean social welfare (SW) against a two-tailed alternative hypothesis, i.e.:
H0: Mean GE (SW) = Mean CF (SW)
H1: Mean GE (SW) ≠ Mean CF (SW) 8.2.1 The Artificial City The artificial city is composed of twenty-five population centres which are the 5 x 5 grid of areas labelled A to Y at their centroids in figure 3 below.
Figure 3 The Artificial City with Public Transport Network All population centres are connected via walking segments. The transport grid is the four vertical and the four horizontal lines in between the population centres at the intersections of the walking segments where there is also a public transport stop. People can board and alight at each stop and can transfer to another line, which enables the movement of people around the city from any origin to any destination. People walk from the population centroid to the transit stop and then take either one or two lines, with a transfer to a second line if necessary, and thus to a suitable transit stop adjacent to their destination. They then walk to the centroid of their destination. The number of route alternatives varies across origin-destination pairs. For example, there are only two route alternatives from A to E (that is, a person can either walk all the way or board the line at stop 1 and alight at stop 4), while there are 34 possible routes (with at most one transfer) from G to S. In total, there are 6,744 possible routes explicitly considered in the model. The spreadsheet that models this artificial city uses a graphical interface to allow infrastructure associated with each line of transport to be altered in
Page 114 of 259
ways that affect the capacity and operational efficiency of the system to carry passengers. 8.2.2 Game Operations The financial and numeric workings of this model were formulated in an Excel spreadsheet. Participants in the game were able to adjust the network resources to optimise the running of the network. The resources which participants were able to vary were confined to three service elements only: 1. the number of vehicles used on each line of the network. The number of vehicles establishes the overall line capacity and headway between services and thus waiting times for passengers. 2. the enhancement of traffic light operations. The traffic light enhancement, if invoked, reduced delays of the transport vehicles at intersections 3. a transfer facility to make it more convenient for passengers changing between different lines of transport. The transfer facility if invoked removed the disincentive to change lines at that intersection, thus making a change of lines more probable where this improvement had been implemented. There were different economic consequences of altering any of the three service variables, each affecting capital and/or running costs. Variables included in the model affected labour, maintenance and running costs components. There were additional running and labour costs where more vehicles were added and, high capital costs but no running costs being associated with the two intersection variables. Initially, resources were set at a very rudimentary level, with just a single vehicle for passengers on each line and no improvements to traffic lights or intersection transfer stations. As the three service components were manipulated, the travel demand was affected. Travel choice behaviour was accounted for via the Multinomial Logit Model (MNL) that considered multiple travel alternatives from each origin to each destination. Concerning the transport network, it was assumed for simplicity that a grid network structure and only two modes of transport, namely walking as a private mode, and bus as the public transport mode. While it would be possible to consider multiple private and public transport modes, the approach taken was that this is not necessary to investigate different ownership regimes and answer the research questions. Another assumption is that no congestion is present on the network (although crowding inside vehicles is considered), which also does not interfere with study objectives. The simplicity of the network model was of great importance to achieve a game that could be explained and played with sufficient ease. The utility functions for each public transport route alternative in the model considers in- vehicle travel times (which are reduced if traffic light enhancements are available on the route), access and egress walking times to the public transport stops, average waiting times (depending on the frequency or headway of a line, which in turn depends on the chosen number of vehicles on a line and whether or not the traffic light enhancements had been invoked), the level of crowding inside the vehicle (which depends on the number of vehicles on a line), and a transfer penalty (which is reduced if a transfer
Page 115 of 259
facility is available). Since a grid network structure for the artificial city was assumed, routes with two or more transfers were not considered as a relevant alternative. Instead of using public transport, there was also the choice to walk the entire trip or to stay at home. These two alternatives allow the demand to be elastic, i.e. the number of people using public transport is affected by the public transport service offered. To mimic the desire for performing an activity at the destination (e.g., work, education, shopping), a positive utility was attached for reaching the destination (while such a utility was omitted for the ‘staying at home’ alternative). Most values of the parameters in the MNL model (which reflect preferences towards each of the above-mentioned attributes) were adopted from the literature, (see Chapter 8.2.10). A further restriction of variables was that relating to the fare charged per distance travelled. This would vary in the proposed adaptation of the ski field’s model to urban public transport, as it currently does with ticketing on the Sydney network where each passenger pays for the distance travelled. This would have been a complication to this Excel model that would not have illuminated any aspect of the basic question about how the resources are allocated to attract passengers rather than to walk or stay at home. However, it is noted that the regime currently operated by Dolomiti SuperSki uses the single payment for each day of service provided, regardless of the number of uses, the distance, height or any other variable. This equates to those public transport networks which have a single access payment for access over a defined number of hours, a common way of charging for network use in many cities. 8.2.3 Participant Suitability Participants experienced in the planning of transport networks were sought to be involved in this experiment. This pool of potential participants is small and needed to be enlarged by including persons without direct experience in the transport field. These were labelled non-professionals to distinguish them from the professional group. A “beta” version of the game, before the most suitable measure of social welfare had been properly determined but where the game was performing consistently, was used to establish participant suitability. To test the suitability of the non-professional participants the results of both groups were analysed to examine their relative performances. It was hypothesised that the two groups would produce the same amount of the mean social welfare (SW) measure, i.e.:
H0: Mean Professional Participants (SW) = Mean Non- Professional Participants (SW)
H1: Mean Professional Participants (SW) ≠ Mean Non- Professional Participants (SW) The experiment also addresses Research Question 3 stated in Chapter 3, where it was asked: Are competing line owners able to cooperate so as to enhance network performance? The Expected Outcome no 2 for this research question was:
Page 116 of 259
That in a game context, there will be sufficient cooperation to provide an acceptable allocation of resources compared to that which is achieved through central planning. This issue of cooperation was examined by observing the effect of competition between line owners in the Community Franchise regime and the propensity for them to cooperate in making the network more efficient on shared infrastructure. Cooperation was examined by the inclusion in the Community Franchise regime game of some aspects that required cooperative behaviour of competing participants, to achieve better network outcomes. What transpired during play was observed and assessed as to its role in either hindering or advancing the essential need for cooperation between competing line owners. Their individual behaviour was observed and recorded, with an awareness of any examples where there was no cooperation. 8.2.4 Genetic Algorithm Being a game set up in Excel, it was possible to have the problem of resource allocation resolved within the spreadsheet and hence to determine the ‘best’ or ‘optimum’ solution for this bounded mathematical problem. This was sought to provide an objective maximum against which to compare human participant scores. The inbuilt Solver in Excel was not capable of solving the network problem; it could not handle the number of variables within the formulae or cope with the size of the program. Accordingly, the use of a more powerful genetic algorithm as an add-in program was employed for this purpose. The commercial product Evolver works as an integrated component within Excel and proved to be a successful means of finding the optimum allocation of resources, a finding consistent with the literature (Platt and Mtenga, 2007). Evolver is an add-in for Microsoft Excel that attempts to find the best solutions to optimization problems by adjusting variables and comparing results. The program allows the user to find the maximum value for Social Welfare in Excel for the cost of system variables by this device of automatically adjusting the variables. The given range for each adjustable cell is a hard constraint that is strictly enforced. Working within the Excel game program, Evolver rapidly adjusts, in a systematic way, the decision variables. It retains solutions that provide better outcomes and progresses until a maximum result is achieved. This may only be a local optimum and so, in a manner consistent with its genetic descriptor, makes random changes to the decision variables and repeats the process. In this way, it will explore multiple solutions but quickly settles on strategies that produce results that no humans, or groups of humans, could match. This process can run for as long as it is desired. It was tested in overnight runs, but by and large, these long periods did not do better than runs of less than an hour. A genetic algorithm starts with the initial population of individual “chromosomes” corresponding to different ownership groupings of the problem. The makeup of each chromosome for this study is composed of the numbers of vehicles, transfer stations and traffic lights applied to all allowable segments of all possible solutions. Mutation and
Page 117 of 259
crossover operational adjustments are carried out after several generations without improvement of the objective function. The use of a genetic algorithm was appropriate given the complex nature of the problem (Mitchell, 1998). It enables the phenomena of suboptimum local maxima within the search space to be overcome by varying the conditions with random generation of variables in the parameters and ultimately the selection of the fittest solutions. This leads to a gradual selection of the solution most fit to the circumstances and is only discovered over prolonged time frames and the scaling and discarding of different local maxima that are found not to be the highest. The objective function for Evolver was set to maximise social welfare. The variables able to be manipulated were the same as for the human groups, i.e.: 1. the number of vehicles, 2. the installation of traffic lights at the chosen intersections, and 3. the installation of line transfer facilities at the chosen intersections. The software was run a sufficient number of times and over sufficient lengths of time to give a reasonable expectation that the result was valid and in fact that no better solution could be found. Mitchell (1998) indicates that it is a feature of this search method that there is never absolute certainty that the highest solution has, in fact, been revealed. However, if it is run for a sufficiently long time, the doubts surrounding this uncertainty reduce. The software explored the individual solutions at a rate of about 1 every 2 seconds. This was run for several sessions of over 18 hours and the resulting figures became the benchmark against which the human participant results were compared. No participants knew those values before their play and they were only ever revealed after the collection of results of all games, if at all. 8.2.5 Optimisation of the Network: The Decision Variables In the Government Enterprise regime game, the single authority plans and implements all decisions as to the infrastructure to optimise its operation and maximise social welfare. The individual human game participants could optimise the city’s transport infrastructure as the agent with control over the allocation process for all lines. In the Community Franchise regime game, the participants were given sole ‘ownership’ of particular sets of lines with three persons each having responsibility for two or three lines (2x3 lines plus 1x2 lines), being discrete parts of the whole transport system. In this version of the game, participants were instructed to maximise their own profitability, with the calculation of the social welfare measure and total producer surplus being hidden from view. Each participant had to decide the number of vehicles to be used on each individual line. Increasing the number of vehicles on any line increased that line’s capacity and decreased vehicle headway (the time or distance between a vehicle and a subsequent vehicle) and thus waiting time for passengers. Decisions made on vehicle numbers affected only the lines in the ownership of that participant.
Page 118 of 259
The decisions regarding the traffic light intersection improvement on any line, influenced the frequency of vehicle operation on the two crossing lines affected and thus passenger waiting time on those two lines. This had a direct effect on the efficiency of the existing fleet operating on both those lines but was noted to have also had a more diffused effect on overall network efficiency. The interchange improvement induced people to change at that intersection to the potential advantage of the two lines involved. The intersections that were chosen to allow traffic lights to be modified involved lines which, in the Community Franchise regime game, involved different ownership groups. Upgrading this infrastructure always involved negotiations with other line owners; this upgrade option was not available on intersections involving two lines having a single ownership group. This situation provided the opportunity to examine individual line owners’ behaviour and if and how they maximised their own self-interest where an owner needed to interact with competing owners. The layout required that each participant must negotiate with each of the other line owners on two separate pieces of shared infrastructure, namely, a traffic light enhancement and a transfer facility, both of which were designed to occur only at the intersections of separately owned lines. 8.2.6 An Outline of the Excel Model The two spreadsheets for each game version are accessible to the examiner of this thesis as separate Excel files. A third Excel file with the individual results and the analysis of all the valid games is included as well. These three files are: 1. 2.0 game_experimental_version Community Franchise consolidated Evolver revised CS, 2. 2.0 game_experimental_version Government Enterprise Consolidated Evolver revised CS and. 3. Participant List.xlsx The Excel model was configured so that each section was contained on a separate tab. The ‘Assumptions’ are contained in the first tab. This includes, for example, the length of a segment of the line, the speed of a vehicle, the capacity of the vehicle and financial matters such as capital cost, interest rates, etc. This tab is followed by the ‘Demand’, an arbitrary but considered distribution of the population’s potential travel demand to and from the twenty-five different centres of the city. Once set, this was never altered. The ‘Calibration’ tab allows for the use of an optimisation routine in Excel to obtain the best arrangement of resource allocation to maximise the social welfare measure. This tab was disabled for the running of the experiments with real participants. The participants input their changes of network resources into the system on the ‘Input’ tab, (various screenshots of which appear later in this chapter). These show the sections of this screen enabling both the inputs and the consequences resulting from changes to infrastructure to be seen side by side. The proximity of measures of cause and effect provided the impacts of their decisions in real-time. The network capacity resulting from the choices that participants make is calculated on the next tab labelled ‘Capacity’. This then leads to the next tab, ‘Routes’, in which the
Page 119 of 259
core of the network model is implemented, including the MNL model and assignment of travellers to line segments. Crowding, in which the number of travellers on a vehicle influences route choice, determines the number of travellers on a vehicle, and require the network model to be iterative. Therefore, 14 additional tabs containing the intermediate iterations of the network model to reach equilibrium are required (during testing the model it was found that a stable solution is reached after approximately 10 to 15 iterations using the method of successive averages). ‘Routes’ provides the final figures of the iterative process to give the passenger numbers on each line for any particular allocation of resources. The next tab is ‘Performance’ which takes the actual numbers of passengers travelling on each section of each line in the hour of the day that is being measured for establishing the numbers of people travelling on every section of each line in both directions. This leads to the allocation of revenue and cost implications which are calculated on the ‘Financials’ tab. The financial calculations determine the financial state of each line for the hour the model is assumed to run. It then extends this snapshot of the financial state for that hour to provide the annualized result, the measure of a snapshot of a year’s projected operation. This is not to suggest that actual financial results for a whole year would be consistent with an hour’s economic performance but is designed to provide a consistent comparison across the two regimes being examined by reference to this single measure. The whole enterprise in the Government Enterprise regime game, and the three different groups of lines in the Community Franchise regime game, pay tax and dividends from net earnings and then retain capital from any profits for reinvestment. The tax rate is based on Australian experience and is set at 30%. Dividends were set at 50% of after- tax net margins. All the figures for performance and financial matters are read back to the ‘Input’ screen to provide real-time feedback to participants adjusting the resources of the model. 8.2.7 The Calculation of Social Welfare The social welfare for the purposes of this thesis is the addition of the consumer surplus and the producer surplus. In the spreadsheet, the figure for this measure was shown to participants. It was made prominent in large bold type to the players in the Government Enterprise game, in cell S36 of the Input tab. It was less evident in smaller un-bolded type to players in the Community Franchise game where it was not hidden and was still quite visible at cell S35. In both versions of the game, this cell displaying the figure is a reference to the cell F97 in the Financials tab of the spreadsheet where the calculation of Social Welfare is prepared. The Financial tab is where all the operational costs and margins are calculated. The figure for financial performance on that tab is shown in its two component parts, i.e. the revenues (F95) and the costs (F96) and added to that figure for the Consumer Surplus (F94). Social Welfare is the addition of these three figures.
Page 120 of 259
The Consumer Surplus is prepared through a series of iterative calculations which ultimately is the addition of the component surpluses attributable to each section of each line. The social welfare component is the addition of the component parts of this calculation of utility, i.e. the walk time, the in-vehicle time, the access time, the transfer time, the fare, a crowding penalty and a transfer penalty. These are the components discussed in 8.2.2 above. These utility figures feed into the probability of passengers using the line because of the choices open to them at that point. These probability figures for each line component feed into the first iteration of the logit model, (tab Iteration 1) and the changed circumstances resulting from those decisions feed into the next iteration no 2 and so on. There are thirteen iterations, enough for the figure for social welfare to stabilise in the calculations of the Multinomial Logit Model. The thirteen iterations proved enough for the model to stabilise for any decision taken by the game participants. The final iteration is displayed at the Routes tab, and the figure for social welfare is the addition of all these individual sections of all line segments. This is displayed on the Routes tab at cell AB6751. The Financial tab takes this cells’ input from the Routes tab and displays this in the Financials tab at cell Routes AB6750. When transferred to the Financials tab, it gathered in a single figure the value to which the producer surplus is added. The combined figure is calculated on this tab at cell F97. This information is made available to game participants by the previously mentioned cell on the Input tab S36 in the case of the Government Enterprise game and less conspicuously at cell S35 in the case of the Community Franchise game. 8.2.8 Participant Interface All this happened in the background to the players who were unaware of these calculations but could see the results of their decisions in real-time. Also displayed in real-time were graphical displays of Capacity Utilisation of each line, so that they could receive relevant information on decisions open to them to improve their positions. The resource allocation is controlled on a line by line or an intersection by intersection basis. The intersections at which the network resources could be installed in the Community Franchise regime game was chosen such that negotiations would be necessary between each of the three separately owned groups of lines for both the lights and for the transfer facility. The interface with the participants is shown in Figure 4 below:
Page 121 of 259
Figure 4 Input Screen for Community Franchise regime experimental game in Excel 8.2.9 Feedback to Participants The feedback given to the participant was fed back to the ‘Input’ screen. These were displayed for the effects of any consequences of the changes to the network resources to be immediately apparent. The features most relevant to the participants were presented and what is shown below is the considered result of the numerous prototypes evaluated in early testing of the game before the final layout was determined. The screens were designed to provide all relevant information to participants, adjusting as input was made. The screen displays the model situation with respect to:
• capacity and passenger utilization of each line • borrowings and the ratio of borrowings to turnover • total passenger numbers • profit for each set of owned lines, • the capacity and utilization of each segment of each line and • the overall passenger numbers for each line and globally. A portion of the screen giving feedback to the participants in the Community Franchise regime game and showing the three separate sets of line ownership is as follows in Figure 5:
Page 122 of 259
…..
(……Continued below)
Figure 5 Feedback portion of Community Franchise regime screen on the Input tab Additional information was shown on the input screen, e.g. Return on Investments (ROI) for each line, but it was apparent that the six elements mentioned above were of most interest to participants. Feedback relating to ROI and route flows was largely, but not universally, ignored. In the Government Enterprise regime game, the figure prominently displayed was that relating to passenger numbers, borrowings, and the measure of social welfare. In the Community Franchise regime game, each set of lines in one ownership group had its combined passenger numbers, borrowings and group profitability displayed. 8.2.10 Travel Demand Travel demand was established by the size of the population of each population centre. This was arbitrary, varied over the artificial city but remained the same for every game. Demand for service on every line builds up towards the centre of the city as would be expected in most urban settings. For participant feedback, the utilisation of each section of each line and the overall utilisation of each line is shown on this same screen showing the demand. This feature is displayed in a ‘heat map’ of the intensity of demand shown in Figure 6 below.
Page 123 of 259
Figure 6 Demand and Intensity of Line Use shown in a ‘mimic’ panel of the network 8.2.11 The Multinomial Logit Model (MNL) for Passenger Allocation As previously described, the MNL model was made responsive to various behavioural effects that included a propensity to tolerate ‘walk time’, ‘in-vehicle time’, ‘access time’, ‘transfer time’, ‘fares’, ‘crowding’ and a ‘transfer penalty’. The values for the marginal utility of making a trip (which influences the propensity to stay at home) and the influence of the transfer facility on the transfer penalty were calibrated such that the resulting choice behaviour and traffic flows seemed plausible. All parameters in the MNL model were initially taken from the literature (Wardman, 2004, Tirachini et al., 2013, Shires and De Jong, 2009). They were, however, scaled and adjusted by very minor amounts as described later to provide the most realistic responses to participant resource infrastructure changes. Justification for these adjustments is as follows. The coefficient values established in the literature are known to be estimates dependent on a variety of assumptions. They are dependent on the actual conditions of the observed situations, may differ from one city to another, can vary over time and these conditions may or may not be present in the model city under examination in this experiment. There is also a difficulty in comparing coefficients taken from different times and places and using different currencies with different exchange rates over time in
Page 124 of 259
establishing the costs of travel. Accordingly, it is sensible to be able to modify these estimates to reveal realistic actions of passengers in the artificial city. Coefficient estimation required a process to be adopted to give the MNL model the most realistic feel for this experiment and this was done in the early stages of game development. Adjustments to base coefficient rates were made by including a constant for each of the coefficients that could be independently modified with the results on passenger numbers for any, and all routes observed. The checking technique was as follows:
• Commence by transferring parameter ratios (with the cost parameter in the denominator) established in the literature. Values were obtained for in-vehicle travel time, the value of waiting time, the value of travel time savings, etc. • Scale all the parameter values using the global scale parameter in the Excel spreadsheet keeping all parameter values the same until results of movement are plausible. • Fine-tune the results by small adjustments to individual parameter values. • Observe the numbers for ‘walk only’, the total demand for those sets for the different variables, the demand for those routes not requiring a change of line and noting the most preferred route, variation between routes, etc. Employing this technique over multiple random Origin/Destination combinations enabled the impact of each parameter value on passenger numbers to be observed and a judgement made as to whether the settings provided reasonable, realistic results. Minor and more graduated adjustments to the parameter levels were then made until the results were satisfactorily aligned with several objective measures that matched typical outcomes on public transport, e.g. the number of persons who can be observed to actually use a bus in crowded conditions. (This was taken for game purposes to be as many standing as could sit). Each coefficient was retested after adjustments to other coefficients to ensure that modelling for all variations resulted in consistent outcomes. During the experiment, only one participant was not (necessarily) convinced that the model realistically modelled behaviour based on the presented results. However, many commented on how the decisions appeared to ‘make sense’ and were believable reactions of the passengers to the conditions encountered. This was evident in, for example, crowding, where the adjusted coefficient proved to give reasonable results for participants and accorded with most people’s perceptions of reality. Only very small segments of the overall network ever had passenger numbers where those standing were more than those who were already seated and then only by a couple of percent for limited sections of a line length. This gave some confidence that the model behaved in a way that participants would accept as being realistically responsive to their proposed changes to infrastructure. The feedback that accorded with the participant’s perceptions of reality provided an element of acceptance of the validity of the experiment. The outcome of the parameter settings in the MNL model is that the public transport demand was elastic, and more passengers were drawn into using public transport as resources increase and conditions improved. The initial conditions of the model, before
Page 125 of 259
any modification to resources, with just a single vehicle on each line, resulted in about 1,300 passengers using public transport. The highest number that was achieved with all adjustments made to the resources was just below 20,000. 8.2.12 Calculation of Revenue and Profit With the varying numbers of passengers, there is a corresponding variation to the revenue, calculated on a fare per passenger basis. As additional assets were committed, there was an attendant effect on running and capital costs. Vehicles had to be bought and required a driver with salaries and overheads. The vehicles required maintenance and depreciation had to be accounted for. The traffic light upgrades and the intersection transfer facilities were taken to be capital costs only, without any ongoing labour costs associated with them. Any running costs of these two improvements were assumed to be amortised over the life of the installation and reflected in a single capital payment. Overall margins of financial operations were calculated considering these costs and the revenue derived from the number of passengers using each section of each line. When the net impacts on financial implications were calculated, there was firstly tax to be paid on any surplus, and then dividend payments of 50% of the net after-tax figures to be paid to the owners in both regimes. The revenue which remained, after expenses, taxation, and dividends, was added to retained capital. This was then available for reinvestment in the network. Any retained capital earned interest and all negative capital, (i.e. borrowings), was charged interest and those different rates were set appropriately given today’s economic circumstances. There was a limit to borrowings set at 30% of turnover for both the Community Franchise regime and the Government Enterprise regime games, which was strictly enforced. This is the equivalent of borrowing mandates set by either the government treasurer or a franchisee’s bankers. 8.3 Playing the Games The Excel model is identical in its workings for both two regimes under investigation. The three sets of lines under separate ‘ownership’ were colour coded to enable each of the three participants to easily identify their own set of lines. Many preliminary versions of the game were prepared and played until it was working such that stable conditions were obtained in a reasonable length of time (about 20 minutes). It was concluded from this process that these game procedures were adequate and that stable situations evolved whereby participants had no further desire to make additional changes to the three decision variables. In the Government Enterprise regime game, each participant played separately. They had control of the whole network and could allocate resources in whatever way they believed would maximise the social welfare. Only the researcher was present in the room to advise of playing measures, to remind participants of their objective and, if required, to make the inputs to the computer. When establishing the optimum allocation by using the genetic algorithm, Evolver had control of the whole system with the objective of maximising social welfare.
Page 126 of 259
The Community Franchise games were played with groups of three participants. All participants were present in the same space, in front of the screen being controlled by the researcher, who entered the input as directed. In the Community Franchise regime game, each of the three participants with responsibility for their own subset of the overall network was only concerned with their own net profit as their objective function but could also see the profit of the other two participants. 8.3.1 Choice of Participants The inclusion of a substantial number of people familiar with the planning and running of public transport enabled a reference set of results to be collected with which to compare the other group of participants’ results. Included in the professional group were transport academics and PhD students, government transport planners, employees and owners of private companies operating public transport and employees of the peak operating body, BusNSW. All other participants were treated as non-professional participants and noted thus. These were drawn from a wide range of people and their results were compared with the reference transport planning professional group. The two sets of results of the professional and the non-professionals were statistically compared to see if the results could be pooled. As will be reported fully in the next chapter giving the results, the null hypothesis of equality of population means could not be rejected at the 5% level of significance (two-tailed p-value of 0. 0.2479). On this basis, all results were pooled. Information obtained about participants was limited to the professional grouping, i.e. with or without transport planning experience, and what age bracket, (below 30, between 31 and 60, or above 60). No other information was sought or recorded. 8.3.2 Data The game was organised so that data on capacity, utilisation, financial performance, and generation of the social welfare measure could be collected for each individual line and, as well, for the overall network. Play proceeded until stability in resource allocation was achieved. The game concluded when stability had been reached when participants identified there was no better way to allocate their resources to improve the performance of the network and generate either additional social welfare or more profit. There was no time limit imposed on reaching this stable resource allocation. Figures for these elements were taken as the final settled stable resource allocation and were compiled into each of the three collections of lines for the participants in the Community Franchise regime game and for the whole network in the Government Enterprise regime game to give the results for each participant. 8.3.3 Individual Participant Differences As play was being conducted between people of different skill levels, aptitudes, and enthusiasm it was necessary to design the game to acknowledge and cater to these differences. Accordingly, it was intended for the original (‘beta’) game series that the level of social welfare achieved by any group of three in the Community Franchise regime game was compared specifically with the individual scores of each of those same
Page 127 of 259
participants when playing the Government Enterprise game. This process was modified as described later in this chapter to cater for peculiarities in the spreadsheet calculations. 8.3.4 Game Introduction To begin the process, participants were initiated to the game concepts with a small PowerPoint presentation to put the purpose and background of the experiment in context. This made the connection between the ski lifting in the Dolomites and public transport and established that the game would examine the differences between these two business regimes, the Community Franchise regime, and the Government Enterprise regime. This established the purpose and generated some motivation to provide their input with each participant’s completely voluntary involvement. 8.3.5 Operation of the Community Franchise Regime Game The play for the Community Franchise regime game was conducted with all three participants seated in front of the computer so that the screen was visible to all. The researcher had control of the keyboard and provided the input of all decisions taken by the three participants. The groups of lines for individual ownership were chosen by the participants on as little a whim as a colour preference for the line group colour coding or any other criterion, arbitrary or otherwise, able to be drawn from the group. This was kept completely random and not directed by the researcher. When line groupings were established, play proceeded by allowing each participant to have three decisions on resource allocation and then moving to the next participant. The performance of each line group was obvious to each of the participants so that the impact of any decisions was immediately obvious to everyone. This was important feedback to the other participants who could plan their moves and strategies whilst waiting for their turn. In time the need for a strict number of goes in any turn broke down as the play developed and the situation matured. At later stages of the game, turns as a means of regulating decision making was not important to the participants and a more ad hoc interaction evolved. The objective of the regulation of decision making was to ensure no participant would feel disadvantaged by the way that others took decisions and that each participant was able to truly maximise the profit for their group of lines. Some participants would decline to make decisions at certain times to see what others did or would delay a negotiation on common infrastructure, hoping to gain additional information. 8.3.6 Operation of the Government Enterprise Regime Game Participants were able to sit with the researcher alone to make all decisions about the allocation of resources for the entire network. No others were in the room at the time that each participant was making decisions. If they felt competent, the participant made the computer inputs, or if for any reason, did not wish to input, the researcher was available to do that. The researcher ensured that the participant was aware of the options and was making valid choices but was at pains to ensure that no decisions were encouraged or discouraged.
Page 128 of 259
Play in the Government Enterprise regime game was slightly more efficient than in the Community Franchise regime game as there was no need for negotiations on infrastructure common to an intersection. Many Government Enterprise regime game participants took significant time to optimise their results whilst many were quite content as soon as they reached a stable allocation and no further improvement seemed possible. All participants were given every encouragement to ensure that they had exhausted the possibilities of increasing the figure for social welfare and which had achieved a stable amount that could not be bettered. 8.4 Potential and Inherent Biases in Experimental Design The game as played is acknowledged to have the potential for biases. Certain potential biases in the basic design of the experiment and some biases of the researcher that could affect the results were identified. These concerns were addressed in ways that dealt with them within the experimental time and cost constraints. 8.4.1 Sampling Bias It was initially seen to be important that participants to be included were skilled in the planning of transport systems. It was expected that their experience of the planning function in public transport would contribute to realistic planning decisions in the game. However, having the whole sample from this source was not a practical option because this population was not large and ultimately only nine sets of three participants were recruited from this source. People not experienced in transport planning had to be recruited to make up the numbers and that group totalled nineteen sets of three participants. Potential bias in this situation was overcome by tracking and comparing the results of both groups; those who were classified as experienced transport professionals and those who were not. Tracking the performance of both groups presented the opportunity to establish the capability of the professional group against that of the non-professional group, the members of which made up the majority of the 84 individual participants. The performances of the two groups are reported on in the next chapter. It was found that there were no statistically significant differences in the mean of the populations from which the two samples were drawn. The game is a vast simplification of reality. It was not designed or held to be a reliable guide as to the performance of the task of real transport planning. At the same time, this experiment was not about planning and operating a real transport network: it was essentially about comparing business regimes and the performance of a set of competing individuals against a set of people who were able to plan the whole network with overall central control. It essentially tested centralised resource allocation against bottom-up undirected resource allocation. The mechanism, by which this objective was achieved, is of lesser importance, provided that the essential test was addressed. It is contended that this experiment does address the essential test and that these considerations validated the reliance on the whole sample which then gave results that could be used to evaluate differences in the business regimes.
Page 129 of 259
8.4.2 Researcher Biases The method of conducting the games is inherently open to the bias of the researcher. The researcher was present with each participant in the Government Enterprise regime game and with each participant group when playing the Community Franchise regime game. The researcher could influence the choices made by each group of participants in the Community Franchise regime game, and at times for participants in the Government Enterprise game, when they were uncomfortable interacting with the computer. This need required a certain amount of interaction between the participant and the researcher for play to progress. The choices open to the design of this experiment within the constraints of the budget for research were such that very little could be done to overcome this disadvantage. Using a strict economic experimental design would have removed this bias. The research budget did not accommodate the use of the economic laboratory, the payment of 28 groups of three participants, attracting multiple groups of industry personnel to the University of Sydney economic laboratory, or the use of such software as zTree (Fischbacher, 2007) to enable totally detached inputs from participants and computer laboratory game control. The only way to practically achieve the required sample size was to have the Excel program loaded onto a memory device and taken to the places of work of the various industry groups who were vital to obtaining the experienced planning and operational personal involved. This situation provided the potential for the researcher to ‘coach’ the participants in ways that might enable them to improve their scores. Care was taken to ensure that the treatment of every individual was the same and that any instructions on how to play the game were the same for all participants and not favouring any particular outcome. This bias was minimised by the researcher’s awareness of the need to control it and the knowledge that the experiment might at some time be repeated by others. 8.4.3 Order of Play Biases Participants learned from the experience of playing a version of the game, either Government Enterprise or Community Franchise regime first, and could apply knowledge and understanding of the network performance to the game version which was played second. This potential bias was overcome by altering the order in which the two versions of the game were played. Half the total number of games was played with the Government Enterprise version first. This order may result in a more competent playing of the Community Franchise regime game. To counteract that probability, the other half were conducted in the reverse order, with the Community Franchise regime game played first. This was sufficient to overcome this bias and it was tested in the results to be reported in the next chapter.
Page 130 of 259
8.5 Explanatory Potential and Limitations of Experiment The experiment was designed to explore the potential of directed resource allocation (Government Enterprise) to generate social welfare compared to an undirected approach (Community Franchise) to do the same. The social welfare in the undirected approach the production is just an unintended consequence of each participant’s pursuit of profit, this being just one, their own portion of the producer surplus, of the two constituent elements of social welfare. The results of the games can be ranked to provide a picture of the different ways that the best arrangement of each regime is found, i.e. by the genetic algorithm maximising social welfare, by individuals acting to plan the allocation of resources to maximise social welfare, and by individuals acting alone to optimise their own profit. The range of possibilities established within the Excel program represents the design space of this problem, and it is very limited compared to reality. The design space in a real transport network of an urban situation is vastly greater. There are many more factors operating, e.g. the types of modes and vehicles, the number of lines, (and potential lines), many and differing economic factors, the reaction of passengers to crowding, changing lines, etc., as well as many other factors that affect consumer choice. Design space, in reality, is not only greater by orders of magnitude, but it is also exponentially greater. The experiment had to limit these factors, i.e. it had to limit the design space, to enable programming in a way that the components of social welfare could be measured, and the resources realistically manipulated by participants. This difference in the size of design space is compounded by the consideration that reality is open-ended, it is not a closed system like the design space of the experiment written in a computer programme. This makes design space in real life essentially infinite. A solution is possible to find in the limited space of the programme, but an unambiguous maximum is not possible in real circumstances. Accordingly, the capacity of this experiment to predict realistic outcomes between the two regimes is limited. However, by examining means of each regime, and by ranking in order the performance of the two regime approaches, with a comparison of each to the solution of the algorithm, some measure of this can be provided. 8.6 Experimental Proceedings When the experiment commenced in earnest the coefficients were operating such that travel behaviour was sufficiently realistic to meet participant expectations and the game was not so complicated as to take more than 20 minutes to reach a position of relative stability where participants were satisfied that no further improvements could be made to their social welfare or operational profit scores. These two criteria were the most pressing concerns and when met, the significant task of organising the eighty-four anticipated participants to obtain twenty-eight valid results for the three participant Community Franchise games was undertaken. The game procedures required the donation of time by each group of three participants for about two hours. This time allowed for each participant to play the Government
Page 131 of 259
Enterprise regime game and then to participate as one of the three owners of lines in the Community Franchise regime game. The period in which the experiment proceeded was from June 2016 to November 2016 by which time 28 valid results for the Community Franchise regime game had been obtained with a concomitant 84 valid Government Enterprise regime game results. This ultimately was determined to be sufficient to undertake statistical analysis. However, after all the games had been played, it eventuated that the means of calculating social welfare within the spreadsheet was incorrect in an important detail, although all other aspects of the spreadsheet withstood thorough critical review. This version of the game may have been sufficient if all players were using the same objective measure, but as outlined in the methodology, the objective measure for the games was different. The Community Franchise game participants were targeting their own portion of producer surplus, and the Government Enterprise game participants were targeting social welfare. As a result, the error in the spreadsheet did not allow for a defensible comparison of the ability of both regimes to maximise social welfare. This series of games was useful in other regards however and has been identified as the ’beta’ series. The “beta” game series has a use for certain aspects of analysis, specifically those games that compared performance based on only the Government Enterprise regime game where all participants were playing in the same version of the game. Games played on this basis enabled a defensible comparison of professional participants to the non-professional participants, and for a comparison of age-related performance. When the calculation for social welfare was corrected in the experimental spreadsheet, the correct figure for social welfare was determined for all Community Franchise regime games. The program itself is deterministic in its calculations and any particular resource allocation always results in the same results for social welfare or any other measure. This predictability was true for the Evolver games as well which was rerun for the corrected version of the spreadsheet to produce the social welfare figure for the optimised figure. This series of games is described as the ‘final’ series and 25 results were reliably recorded by this method. The ‘final’ series could not be manipulated to show the human performance in the Government Enterprise regime game because, unlike the Community Franchise regime game, social welfare was the objective function they were seeking to maximise, not components of producer surplus. Participants may have reacted differently to the unfolding game as they progressed through the various alternative resource allocation possibilities if the social welfare had been different. Having human to human comparison was however necessary for testing of the hypotheses. Accordingly, a second series of Government Enterprise regime “final” games was then played with human participants to provide human data that was consistent with that derived from the multiple participants in the Community Franchise regime game. This could not, for practical reasons, be played with all 84 participants. In place of this ideal approach, a matching number of 25 Government Enterprise regime games were played with a new set of participants who were drawn, where possible, from
Page 132 of 259
the original participant list, but with new participants to make up the numbers where necessary. The resulting data, therefore, consists of the actual scores of those ‘final’ games and not the average of the three participants as in the “beta” series. This data may be considered as not as reliable as the original game design intended, but in the circumstances provided a practical way of comparing human to human performance in the two regimes. In particular, it did not fully avert the potential bias for the order of play. However, there was a considerable period of nearly 12 months between the two playing sessions, and this would nullify to a large extent those order of play biases through the passage of time between the two sessions of play. This aspect has been retested and the analysis arranged to suit this reality. It is reported on in the next chapter. The differences between professional and non-professionals were also retested for this ‘final’ game series and as is reported in the next chapter where it is shown that, the null hypothesis of equality of population means could not be rejected at the 5% level of significance (two-tailed p-value of 0.04787) allowing the data to be pooled. 8.7 Summary This chapter has described the computer experiment and discussed the reasons underpinning its design, including the accommodation of potential biases and the limitation of the experimental approach. After many iterations of varying game complexity at the preliminary stages, the game was simplified and refined to its final form where it was able to be played with consistency and with some speed to achieve a stable allocation of resources which participants were not able to improve upon. Stability in the solutions being ‘discovered’ by participants was able to be achieved in a manageable game time frame of about 20 minutes. At this point of stability, all participants were satisfied that they had maximised their objective measure. All participants were allowed to reach this stable state without time limit. The data generated from the human participants in the Government Enterprise version of the game in this original series of games had to be set aside for hypothesis evaluation as it could not be reliably compared to the data generated from the Community Franchise series of games. This first set of results was labelled the ‘beta’ series. However, the data generated by the human participants in the Community Franchise game could be used when modified to conform to the corrected calculation of social welfare and 25 results were able to be retrieved by this method. This recalibrated game was labelled the ‘final’ game in the Community Franchise regime game. Human participants were re-engaged in playing the modified Government Enterprise regime game to provide comparable figures to those results from the ‘final’ series of Community Enterprise regime game. Evolver was used to provide a maximum anticipated result for this ‘final’ game series. The ‘beta’ series of games provided the means of assessing the suitability of non- professional participants as well as examining age performance and order of play issues.
Page 133 of 259
In the next chapter, the results are presented, hypotheses and research questions reviewed, the significance of those results discussed and conclusions about the experiment drawn.
Page 134 of 259
CHAPTER 9 RESULTS OF EXPERIMENTAL GAMING 9.1 Chapter Content This chapter gives the consolidated experimental results. It reports the raw figures of the ‘beta’ games, i.e. that version before the final measure of social welfare had been determined, to establish the suitability of the two groups of participants and the appropriateness of pooling the results of the professional and non-professional participants. Having established this prerequisite, it provides the experimental results of the ‘final’ game series to be relied upon for hypothesis evaluation and then analyses them. Next, it reviews the hypothesis of the Research Question no. 1 concerning the quantitative performance of the two regimes and the expected outcome of the Research Question no 2 concerning the propensity of competitors in business to cooperate. It includes a description of the playing methods and strategies observed to be adopted by participants. Finally, it summarises the experimental conclusions. 9.2 Suitability of Participants The results have been assumed to be dependent on the suitability of all the participants to be included in the figures. The validity of using those who were not experts in the field of transport planning was first tested to establish whether there was significant variation between the two populations, professional and non-professional. Only when this was established could all participants be included in the analysis. The differences in performance, which is dependent on professional status, was only analysed for the Government Enterprise regime game. These participants were acting alone and their results were thus not affected by others as in the Community Franchise regime game. In this test of the results, the null hypothesis was that the mean social welfare achieved by the two groups was equal, and if accepted, then the results of all participants could be pooled to contribute to the experimental database. Testing of this variable was conducted using the ‘beta’ form of the game. A consistent measure of that which was then calculated to represent social welfare was used to determine if both groups of participants, i.e. the professional and the non-professional, were similar. The playing of the games took place in the period July to November 2016. There were 84 results from the ‘beta’ series of games showing the particular measure of social welfare exclusive to that series, which was generated. The results are shown in Table 7.1 divided into the two groups. This table shows the results of the non-professional group the ‘N’ and the professional group ‘P’. Note that for this table and further tables in this chapter, acronyms will be used as follows:
Page 135 of 259
• Community Franchise (CF) • Government Enterprise (GE) • social welfare (SW) • producer surplus (PS) • consumer surplus (CS) The results of all ‘beta’ games for both regimes are as follows: Professional (P) Non-Professional (N) $ 19,879,315 $ 19,851,752 $ 19,851,752 $ 19,839,565 $ 19,851,752 $ 19,828,298 $ 19,809,256 $ 19,791,146 $ 19,761,446 $ 19,712,062 $ 19,712,062 $ 19,613,866 $ 19,688,522 $ 19,247,233 $ 19,678,708 $ 18,904,076 $ 19,672,716 $ 18,712,254 $ 19,635,592 $ 18,679,691 $ 19,308,399 $ 18,605,607 $ 17,549,728 $ 18,182,757 $ 17,227,941 $ 17,731,717 $ 17,182,403 $ 17,616,260 $ 16,066,764 $ 17,606,940 $ 14,685,894 $ 17,573,937 $ 14,394,826 $ 17,557,494 $ 13,757,505 $ 17,475,284 $ 13,736,118 $ 17,459,574 $ 13,670,608 $ 17,423,027 $ 13,590,868 $ 17,358,111 $ 13,041,573 $ 17,347,827 $ 12,141,742 $ 17,347,568 $ 11,621,951 $ 17,262,567 $ 11,485,022 $ 16,758,651 $ 11,447,856 $ 16,702,114 $ 10,403,445 $ 16,136,932 $ 16,066,764 $ 15,697,650 $ 15,569,123 $ 15,485,098 $ 15,363,224 $ 15,341,223 $ 15,292,877 $ 15,274,779 $ 15,262,136 $ 14,896,062 $ 14,812,506 $ 14,777,865 $ 14,685,894 $ 13,638,373
Page 136 of 259
Professional (P) Non-Professional (N) (continued) (continued) $ 13,509,797 $ 13,402,535 $ 13,272,813 $ 13,265,573 $ 13,238,256 $ 13,065,618 $ 12,934,135 $ 12,833,151 $ 12,732,249 $ 12,387,593 $ 12,321,973 $ 12,124,668 $ 11,294,142 $ 11,029,563 $ 10,487,843 $ 9,413,758 Averages $ 16,253,843 $ 15,786,027
Table 18: Results of social welfare of Government Enterprise ‘beta’ game Non-Professional Group (N) and Professional Group (P) The hypothesis for the testing of applicant suitability was expressed as:
H0: Mean SW of Professional Participants = Mean SW of Non-Professional Participants
H1: Mean SW of Professional Participants ≥ Mean SW of Non-Professional Participants The null hypothesis was that the mean social welfare achieved by the two groups of participants would be equal. The alternative hypothesis was that the professional participants would have a greater mean social welfare than the non-professional participants (i.e. a one-tailed test). The null hypothesis of equality of population means could not be rejected at the 5% level of significance (one-tailed p-value of 0.1620) It was thus confirmed that the use of all participants, whether professional or non- professional was reasonable. Accordingly, the results of the ‘final’ game, in which this group of participants played, are now presented. The results from the experiment are now given. 9.3 Results from the Experiment Results of the participant production of social welfare from the ‘final’ game in both game formats, i.e. the Government Enterprise and the Community Franchise regimes derived from the ‘final’ experiment, are shown in Table 7.2 below:
Page 137 of 259
All Government Community Participants Enterprise Franchise $ 13,543,581 $ 18,046,490 $ 15,609,586 $ 17,987,615 $ 17,906,996 $ 17,947,169 $ 12,416,225 $ 17,925,849 $ 11,624,875 $ 17,925,849 $ 13,447,948 $ 17,886,599 $ 12,675,315 $ 17,878,425 $ 13,376,987 $ 17,778,227 $ 17,875,851 $ 17,699,970 $ 13,655,452 $ 17,693,692 $ 17,871,816 $ 17,559,180 $ 11,860,598 $ 16,954,767 $ 14,771,094 $ 16,060,297 $ 13,510,235 $ 15,801,110 $ 11,367,333 $ 15,801,110 $ 17,750,840 $ 15,770,442 $ 17,871,816 $ 15,655,857 $ 15,659,306 $ 15,655,857 $ 17,856,677 $ 13,664,274 $ 17,769,581 $ 13,585,774 $ 15,457,607 $ 13,559,356 $ 15,836,578 $ 13,502,953 $ 13,096,692 $ 13,367,036 $ 12,645,984 $ 11,678,459 $ 17,955,089 $ 11,395,711 Average: $ 14,936,563 $ 15,951,283 Table 19: Social welfare results of the ‘final’ Government Enterprise (GE) game and the ‘final’ Community Franchise (CF) game The figures for social welfare are noted to be lower than those in the ‘beta’ game series shown in table 11.1. This resulted from the initial figures for consumer surplus, which, although being maximised, were initially negative. Being a measure without a real reference point, this does not affect the relative social welfare generated. 9.3.1 Analysis of Research Question No 1: Social Welfare Research Question no 1 addresses the generation of social welfare from the two regimes. It states, in Chapter 6: What is the impact on social welfare of a group of profit-maximising franchisees compared to a single purpose centralised organisation seeking to maximise this measure?
Page 138 of 259
It was hypothesised that the two groups would have equal population means. The alternative hypothesis was that either one regime or the other would have a significantly different mean, implying a two-tail test.
H0: Mean Social Welfare GE (SW) = Mean Social Welfare CF (SW)
H1: Mean Social Welfare GE (SW) ≠ Mean Social Welfare CF (SW) The null hypothesis of equality of population means could not be rejected at the 5% level of significance (p-value of 0.1175). This suggests that the Community Franchise regime, with its un-directed bottom-up resource allocation, can generate similar amounts of social welfare to the traditional Government Enterprise regime with its top-down centrally planned resource allocation. 9.3.2 Regime Ranking It is interesting to examine the ranking of all 25 participant groups from the Community Franchise regime and the 25 individual participants from the Government Enterprise regime for the social welfare achieved in their games. This provides a graphic illustration of distribution and relative levels. With this ranking of both regimes, there is an obvious spread of results from both regimes throughout the whole range of values, as shown on Figure 7 below, where results are measured in $millions and shown together with the result of the genetic algorithm, Evolver.
$20,000,000
$15,000,000
$10,000,000
$5,000,000
Social $- Welfare 1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31 33 35 37 39 41 43 45 47 49 51 Evolver Community Franchise Government Enterprise
Figure 7 Ranking of all Social Welfare results from both regime games with Human Participants with that of Evolver Figure 7 shows a similarity between many of the results – some results are close to the Evolver result then there is a step-down and then a further step-down. The ranking exercise does show similarities in the amount of social welfare that participants in the Government Enterprise and Community Franchise regimes were able to achieve, despite the different workings and different focus of each regime. This view is further explored in other aspects of the data, by examining the effect that prior experience of playing the game had on these results. 9.3.3 Effect of Experience on Social Welfare The effect of playing one version of the game or the other first, i.e. Government Enterprise game first and Community Franchise game second, or vice versa, had been
Page 139 of 259
tested in the ‘beta’ series of games. There it was tested to see if playing one game ahead of the other had an impact on the mean social welfare that could be achieved. The null hypothesis was that in the original ‘beta’ game series, the mean social welfare generated by the different groups was independent of the order in which that game was played. The alternative hypothesis was that the mean social welfare of participants who played any version of the game second would have a greater mean than those playing the same game first because of the experience they had gained (i.e. a one-tailed test). The test was performed for both the Community Franchise games and the Government Enterprise games. The null hypothesis of equality of population means could not be rejected at the 5% level of significance for either game (one-tailed p-value of 0.90 and 0.32 respectively). The conclusion drawn was that there no significant difference in the population mean social welfare generated by the participants for the order in which either version of the early ‘beta’ version of the game was played. With the ‘final’ game series, this player order of experience effect was not able to be tested in the same way as for the ‘beta’ series. There was no before and after groups to provide a similar test for the ‘final’ game series and different definitions were required to differentiate those who had played before and those who had not. The ‘Experienced’ participants in the ‘final’ game series were defined as those who had been previously involved in the ‘beta’ game series. Those who had been newly recruited to play only the ‘final’ Government Enterprise regime game were labelled ‘Inexperienced’. The number in each of these two categories were kept to as close to equal groupings as possible, with 12 participants being newly recruited, the ‘Inexperienced’, and 13 participants having previously played in the ‘beta’ series, the ‘Experienced’. The performances of these two ‘final’ game groups were examined. The results of the ‘final’ game series, divided into these Experienced/Inexperienced categories, are:
‘Final’ Game ‘Experienced’ ‘Inexperienced’ Government Had played in ‘beta’ Had not played in Enterprise series before ‘beta’ series before $ 13,543,581 $ 15,609,586 $ 17,906,996 $ 11,624,875 $ 12,416,225 $ 13,447,948 $ 17,875,851 $ 12,675,315 $ 17,871,816 $ 13,376,987 $ 17,750,840 $ 13,655,452 $ 17,871,816 $ 11,860,598 $ 17,856,677 $ 14,771,094 $ 17,769,581 $ 13,510,235 $ 15,836,578 $ 11,367,333 $ 13,096,692 $ 15,659,306 $ 12,645,984 $ 15,457,607 $ 17,955,089 Average $ 16,178,345 $ 13,584,695 Table 20: ‘Final’ Government Enterprise regime game results for Social Welfare, divided into Experienced and Inexperienced Participants
Page 140 of 259
The null hypothesis was that the mean social welfare achieved by the two groups of participants would be equal. The alternative hypothesis was that the ‘Experienced’ participants would achieve a higher mean social welfare, because of their previous experience (i.e. a one-tailed test). The null hypothesis of equality of population means was rejected at the 5% level of significance (one-tailed p-value of 0.0004). This may have been the result of the small number of participants, but is more likely related to the fact that by the time the ‘Experienced’ participants had played the ‘final’ game, they had participated three times, having already played the two versions of the ‘beta’ game, both Government Enterprise and Community Franchise versions. This contrasts with the ‘Inexperienced’ participants who, being freshly recruited for this ‘final’ Government Enterprise game exclusively, had only this single experience of playing and they had not, and did not go on, to play the other version. Having established that there was a significant difference between these two groups, the performance of the two regimes as played by the ‘Experienced’ groups was then compared. Table 21 shows the social welfare generated from all ‘Experienced’ participants in the ‘final’ game series for both the Government Enterprise regime game and the Community Franchise regime game. It is noted that the definition of ‘Experienced’ for the Community Franchise regime is different, as described above, from that for the Government Enterprise regime.
‘Experienced’ Government Enterprise Community Franchise $ 13,543,581 $ 17,987,615 $ 17,906,996 $ 17,886,599 $ 12,416,225 $ 17,878,425 $ 17,875,851 $ 17,778,227 $ 17,871,816 $ 17,693,692 $ 17,750,840 $ 16,060,297 $ 17,871,816 $ 15,801,110 $ 17,856,677 $ 15,770,442 $ 17,769,581 $ 13,367,036 $ 15,836,578 $ 11,678,459 $ 13,096,692 $ 11,395,711 $ 12,645,984 $ 17,955,089 Average $ 16,184,440 $ 15,754,329 Table 21: Production of Social Welfare by ‘Experienced’ Participants only, from the ‘final’ Government Enterprise game and the ‘final’ Community Franchise game The null hypothesis was that the mean social welfare achieved by the two groups of participants would be equal. The alternative hypothesis was that either one regime or the other would have a significantly different mean (i.e. a two-tailed test). The null hypothesis cannot be rejected at a 5% level of significance (p-value of 0.6689). The conclusion is that both regimes appear capable of achieving similar mean social welfare values.
Page 141 of 259
The situation was further examined by looking at the differences between the ‘Inexperienced’ participants in both regimes. This allows the value of experience to be more fully explored. These values are as follows in Table 22 below.
‘Inexperienced’ Government Enterprise Community Franchise $ 15,659,306 $ 18,046,490 $ 15,609,586 $ 17,947,169 $ 15,457,607 $ 17,925,849 $ 14,771,094 $ 17,925,849 $ 13,655,452 $ 17,699,970 $ 13,510,235 $ 17,559,180 $ 13,447,948 $ 16,954,767 $ 13,376,987 $ 15,801,110 $ 12,675,315 $ 15,655,857 $ 11,860,598 $ 15,655,857 $ 11,624,875 $ 13,664,274 $ 11,367,333 $ 13,585,774 $ 13,559,356 $ 13,502,953 Average $ 13,584,695 $ 16,106,033 Table 22: Comparison of ‘Inexperienced’ Participants in both Government Enterprise regime and Community Franchise regime ‘final’ games It was hypothesised that the mean social welfare achieved by the two groups of participants would be equal. The alternative hypothesis was that participants of either regime could have a significantly different mean social welfare (i.e. a two-tailed test). The result was to reject the null hypothesis at a 5% level of significance, (two-tailed p- value of 0.0009). These results of the ‘Inexperienced’ are different from the ‘Experienced’ group and may indicate that there are operational differences that affect the outcomes of the two regimes in certain circumstances. It suggests that there may be something affecting the results of the Community Franchise game in at least the circumstances of variable experience but that this is not explicable from this limited experiment. The Community Franchise regime appears to favour an inexperienced participant in the context of this game. As well, almost intuitively, it appears that experience is much more important in a top-down overall planning exercise. 9.4 Genetic Algorithm (Evolver) Result from the ‘Final’ Experiment The algorithm had been run for the ‘beta’ series, (where Evolver had achieved a result that was approximately 7% above the level of the highest-scoring Government Enterprise regime game participant). It was rerun for the ‘final’ game and produced a highest found figure of $18,591,881, which was (not surprisingly) greater than that able to be achieved by humans in either the Community Franchise regime or the Government Enterprise games. It is noted that the capacity of a genetic algorithm running over 16 hours and testing a solution every two seconds is demonstrably superior to humans trying to obtain an ideal allocation of resources in twenty minutes. The higher scores of the algorithm are to be
Page 142 of 259
expected and were seen in the ‘beta’ game series where a similar disparity of human to computer performance has been noted. It is also noted that while this allocation is the highest that could be found using Evolver, it cannot guarantee that the Evolver discovered solution is the global optimal solution as the number of possible solutions is too large to individually check. In these circumstances, therefore, it is not appropriate to speak of optimal allocation, but rather of the highest found allocation. It is evident from Figure 7 that humans from both regimes can almost match that of the algorithm, even though humans had at most three attempts over 20 minutes each session to optimise the resource allocation and Evolver had many hours to do the same. The ‘final’ game human result in the Government Enterprise game was 96.3% of the Evolver score, compared to the 93% of the ‘beta’ game series. This suggests again the positive effect that practice, experience and familiarity seem to have on the results of humans. In a real-life situation with extended time frames, more resource allocation alternatives could be explored in both regimes than were possible in the confines of the experiment and it could be expected that more efficient arrangements are capable of being discovered by humans than within this experiment. The same would not necessarily be true of the algorithm which tends to plateau out reasonably early in its process. 9.5 Further Considerations of the Experiment Further value can be extracted from the experiment by an examination of the solutions devised by the algorithm compared to those discovered by humans. 9.5.1 Passenger Number Considerations It is noted that in the highest social welfare discovered by Evolver, the numbers of people travelling by public transport was less than the number of passengers in the highest producer surplus arrangements found by human participants in either game. The numbers of passengers being carried by the top 10 performers in the Community Franchise regime game is approximately 7% higher than in the Evolver solution. Again, the reasonable expectation of the computer being able to generate a more efficient solution is noted. However, a consequence of having lower social welfare appears to be that more people actually travel than in the most efficient arrangement. Humans must be doing this to the detriment of overall costs, e.g. number of vehicles and other infrastructure overheads, but is nevertheless it is noted to be a feature that may not be without positive economic consequences. 9.5.2 Highest found Resource Allocation The discovered resource allocation is shown on the attached screenshot of the Evolver solution in Fig 8 following.
Page 143 of 259
Figure 8 Actual resource allocation of the best solution found by Evolver It is noted that this particular allocation of resources by Evolver is almost identical to that found in the ‘beta’ game, targeting the then measure of social welfare. (The only exception was the use of an additional vehicle on Line 3). In this highest reported producer surplus result from Evolver, which has a quite unintuitive allocation of resources, three lines had no additional vehicles allocated to them. That set of lines without additional vehicles included Line 1 as well as Lines 4 and 5. Line 1 was the only one of these minimum vehicle allocation situations regularly discovered by human participants. Also, the traffic lights on Line 1 were not activated, thereby saving a significant investment which was another situation discovered by most high performing participants, but not by many of the other lesser performing participants. Another feature of this allocation is that the numbers of vehicles on certain lines are substantially higher than any that participants used in either the Government Enterprise or the Community Franchise regimes which had the effect of making certain heavily trafficked ‘trunk’ lines to carry the bulk of the passengers resulting in optimum economic efficiency. It also meant that the expensive intersection resources, the traffic lights and the transfer facility, on these two lines, did not have to be invoked as there was little passenger traffic on them. This has the effect of saving two very expensive items of expenditure, thus contributing to a higher producer surplus and then to higher social welfare. This compares with that allocation of the best human participant, as shown in Figure 9 below:
Page 144 of 259
Figure 9 Actual resource allocation of the best solution by humans in the Community Franchise regime game. This allocation of resources was very similar in many of the high performing participants in the Community Franchise regime game and the best in the Government Enterprise regime game. The resource allocation of the best Government Enterprise regime game is shown in Figure 10 below:
Figure 10 Actual resource allocation of the best solution by humans in the Government Enterprise regime game Good solutions that maximise social welfare are found that are generally independent of the form of the game in which they were sought. 9.6 Further Observations of Human Participants Although the ‘beta’ version of the Government Enterprise regime game was not used for evaluation of the hypotheses, the results provided important information in how
Page 145 of 259
participants undertook their role, especially when they attempted to plan the network. Observations of these ‘beta’ Government Enterprise regime games provided insights into that function as well as being a contrast to behaviour in the Community Franchise regime game where cooperation was required between participants. These observations supplement those of the ‘final’ Government Enterprise regime game where those earlier insights were confirmed. The various strategies adopted by participants in the two versions of the game, i.e. the Government Enterprise and Community Welfare games, are now considered. It was notable in all games that initial conditions dictated to some extent the outcomes of individual performance. This seems retrospectively obvious given that two participants had three lines and one had only two lines. However, there was variation between performances of individuals that were not related to those initial starting conditions. It was also the case that, as the participant did both versions of the game, any effects from this initial disparity were neutralised. 9.6.1 Community Franchise Regime Game Strategies The strategies adopted in the Community Franchise regime game were clearly directed towards maximising profit. No attempt was made by the Community Franchise regime participants to try and maximise overall efficiency, it not being in their perceived interests to do so: it was purely the profitability of their own set of franchised lines that mattered to them. This was particularly evident in those who had control of Line 1 which responded to additional resources in a very unintuitive way. There was a noticeable tendency to get rid of the redundant vehicles on Line 1 when this had a positive effect on the bottom line of the group in which Line 1 was included in both regime games, but this was not universally found by participants in either form of the game. . This showed that participants did not necessarily learn from their experience in the game that was played first, judging by their strategies in the second game. As the numerical analysis of the ‘beta’ game showed, the opposite was almost the case. There seemed to be a separation in the minds of these players that prevented some of them from using the lessons learnt from one game to the next. This applied to both orders of playing. However, the high performers recognised the need to engage in iterating the bus numbers towards the end of the session and were thus most likely to establish that there was no need to add vehicle numbers on Line 1. Several individual participants initially did not willingly accept that their objective was to maximise their own profit; some saw their objective was to ‘beat’ the other participants. It became evident that if in certain circumstances they could increase their own profit but, at the same time, increase that of the other participants by a greater amount, then these individuals were inclined to sacrifice their own profit to stay ahead of the others. Or at least, they considered restricting their own profit to hurt their perceived ‘competitors’. They contemplated the sacrifice of their own total profitability if it kept them ahead of the others. This never persisted for long and all participants eventually decided to play to maximise their own profit.
Page 146 of 259
In one particular case, this was decisively resolved by the participants themselves when the person who was reluctant to shed additional vehicles on Line1 and by doing so increase his own profit as well as that of his ‘competitors’, was finally convinced that he, and his shareholders, were better off in terms of profit if he did so. It took some time for that participant to internalize that it may be acceptable that by increasing his own profit he increased the profit of his competitors more. This was an example of one of the perverse unanticipated network effects where network performance often reacted in an unintuitive manner to any modification. (As well it was an indication of the complexity of even the simplified reality of networks such as the one in this model). The game was designed with the profits of all participants visible on the screen to others. This could be seen as a game design flaw but was included as, regardless of whether or not the exact figures of a competitor are known or not, the general financial performance of competitors in any field of commercial operations becomes very apparent to other industry participants. The knowledge of the exact figure for profit, having only a single screen for display purposes, was thus seen to be appropriate for the experiment. There was a willingness to negotiate with other participants for the development of intersections, both the traffic light enhancements and the line transfer facilities. There was almost always an exploration of the appropriate allocation of contribution from each line. When this was altered from, say, 50/50 to 70/30, the effect on the profit of each line was evident and a bargaining process ensued which usually resulted in a split that was perceived by both parties to be fair. This was rarely taken to an extreme position but often resulted in differing payments from each line affected. Several participants (numbers 51 and 54 were examples) were watching the Return on Investment (ROI) rather than the absolute amount of profit. Their behaviour was constant in both the Community Franchise regime and the Government Enterprise regime games. This caused them to fail to maximise their profit in both games. It also led them to adopt a very harsh stance in negotiations. In some cases, participant 54 convinced the competing line owner to pay the entire 100% of the cost of intersection improvements even though he benefited substantially from others’ investment. Contributing would have reduced his personal ROI. The other players eventually decided to contribute the whole amount in order to reap somewhat reduced rewards, even though it resulted in a higher profit for participant 50. All participants acknowledged this to be ‘freeloading’; although No 51 was adamant that he was protecting his ROI and that this was the true measure of profitability. He maintained that any excess capital should be returned to shareholders and not kept for reinvestment in the firm. Although other participants in another particular group which had a similarly inclined individual, when that participant, (no 54), was cajoled and it was argued that they were suffering because of No 54’s stance, there was no bending of this particular individual in
Page 147 of 259
the face of adherence to a higher principle.12 This principle concerning the efficient use of capital resulted in a position taken by only one or two other participants. A couple of others, who might have been aware of this point of view, adhered to the game objectives, which was to maximise individual profitability. The results from all the groups containing these types of contrarian individuals were included. Those who failed to meet the game requirement of maximising individual profit were not discarded. This was decided because they represented behaviour likely of some company directors and so should be included as representative of the many aspects of the functioning of the Community Franchise regime under corporate influence. 9.6.2 Government Enterprise Game Strategies The participants in the Government Enterprise games formed into two general groups of response types. There were those whose sole focus was on the figure for social welfare, the maximisation of which became the only indicator of success. This can be likened to a true ‘gamers’ approach. The other large group of participants focused instead on their own assumptions of what would generate the greatest social welfare and hence to provide the most effective allocation of resources. With the latter approach participants were keenly aware of the ratio of passengers to capacity and had a desire to ensure that there was not too much ‘excess’ capacity remaining on any particular line so that precious resources were not ‘wasted’ on this overcapacity. The extreme application of this strategy proved not to maximise social welfare. Also, there were those in the last grouping who sought to ensure that all lines were well serviced, even though it was found by most high performing participants that Line 1 did not require any more than the single vehicle it started with, any addition to that line of more vehicles, at mature stages of the game, reducing social welfare. Because of the focus on trying to manually plan the allocation of resources to that which seemed most logical and fair, the social welfare could be substantially below what was achieved by those who adopted a numerically driven approach to the allocation of resources. Grand strategies were a significant feature of many of the participants in the Government Enterprise game. These were generally applied after carefully noting the demand layouts and the connections of populations. Their strategies were updated during the game as the capacity of each line was absorbed by passengers. Those individuals adopted approaches which they thought would best solve the overall problem presented by demand. One particular participant thought that trunk routes around the perimeter would best solve the needs of passenger movement. More commonly, others thought that trunk lines down the two sets of crossing middle lines would be best. These strategies were sometimes pursued to the extent that they reduced overall efficiency.
12 It is noted that this pursuit of profit over financial efficiency is a regular issue in public companies where CEO’s and upper management can have grand investment plans that are held by some commentators to be more relevant to their own remuneration objectives or other less transparent reasons; an example of a misalignment of interests.
Page 148 of 259
The grand design approach contrasted with the other main set of participants who just focused on the figure for social welfare and experimented with everything else in a detached way. Those who concentrated on the social welfare figure and those who were experimental by inclination were the ones who generally got closest to the best solution found by Evolver. Some participants combined the two approaches, and after much consideration of the demand, they experimented with precise and graduated increments in the three variables, either up or down, and kept a close eye on the figure for social welfare. This was a successful strategy, but it is not clearly evident which part of this dual strategy was responsible for their success. One would suspect that the experimentation was crucial to the overall result and that the study of the demand provided a clearer understanding of what was happening as the network changed. There was a small group who did not engage with a process of design or experimentation to the extent of other participants. Their results were generally not as good as those others in the group. This appeared to be because their ways of working on these types of problems were not as inventive or persistent as others. This group of participants may not have been as engaged in the process of the game as those who had a naturally competitive and inquisitive nature. This group included both those who were classified as experienced transport planners, (i.e. the practitioners and the academics who teach those practitioners), and those who were not. This is consistent with observations of different approaches to the solution of problems experienced over a working lifetime in design-related industries. Some people are more suited to that task than others, whether they are actually engaged in a design industry or not. Having a significant number of participants meant that all personality types were represented over the whole spectrum of design ability and approach. 9.6.3 Participant Performance The most successful participants in the Government Enterprise regime game, i.e. those who came closest to the highest social welfare found by Evolver, were from a wide variety of backgrounds, including transport and academic personnel, but also including young adults with no academic credentials and older people with no background or expertise in transport. There was no noticeable improvement in performance where there was a professional background. The experiment was not testing transport planning skills as such and this is a possible explanation. The game was, however, to establish whether social welfare could be maximised by the singular pursuit of individual profit over a deliberate focus on the social welfare measure by central planning. It was a test of whether an effective network can evolve or if it requires planning, a deliberate test of central planning compared with economic evolution. There is no doubt that the skill of the professional in the circumstances of this simple model was no better than that which resulted from an undirected evolution arising from the pursuit of profit by a variety of people with different skill levels. These observations
Page 149 of 259
and the results themselves call into question the ability of individuals to plan even a simple network like that in the experiment, let alone the more intricate, extensive and mathematically complex networks that occur in real urban situations, networks with elaborate non-geometric layouts and the multiple options for the optimisation of network performance. This question is taken up in theoretical reviews, where the issue of the ways of ‘discovering’ solutions in situations of unbounded mathematical complexity is explored. See, for instance, numerous explanations of the development of all things from the natural to the commercial (Beinhocker, 2007, Dennett, 2017). 9.7 Discussion This section will deal with more general observations of the way that participants played the two versions of the game. There were clear and different strategies adopted by various participants. They could be grouped around certain preferences and could be observed in the two different game versions, the Government Enterprise regime and the Community Franchise regime. 9.7.1 Planning and Performance There was no overall planning that occurred in the Community Franchise regime game; there was no effort expended by any participant to encourage the other participants to do things for the good of everyone. In contrast, there was deliberate planning of the overall network occurring almost always in the Government Enterprise game. Some high performers did not engage in planning as such: their approach was more an experimental tinkering with the system with an eye exclusively on the social welfare amount with no thought being given to trying to ‘work out’ how to improve it. As has been seen, this tended to produce the best results in the Government Enterprise game, because the network effect was such that it was difficult to judge whether any alteration of resources would produce a better overall outcome. However, there was a noticeable tendency for most of the Government Enterprise participants to examine the demand display closely and to allow the demand and line layouts to influence their resource allocation. When this approach was combined with the experimental social welfare maximising approach, performance was not necessarily different from the pure social welfare maximising ‘gamers’. Some planned approaches involved the concentration of resources on perimeter lines or on central lines. There were many cases like this of misdirected theories being implemented, theories that did not have the desired increase in generating social welfare. These decisions were all taken after due consideration of the likely effects of such planned decisions which shows definite limitations in the ability of individuals to predict the consequences of adjustments to a complex system. Very few participants discovered the strategy shown by the best solution found by Evolver, which was to have multiple lines without any additional resources and to concentrate resources onto a few lines which tended to act as ‘trunk’ routes. This strategy gave marginal improvement that set the algorithm’s solution at just above the level achieved by humans playing the Community Franchise and Government Enterprise
Page 150 of 259
regime games. It was frequently found that Line 1 did not respond to having more resources applied to it, but it was never discovered by the participants that those other lines without additional resources would actually achieve higher social welfare. The genetic algorithm was able to consistently outperform humans, and it has been asked if decisions as to resource allocation would be better performed by computer rather than relying on the frailties of man? This observation is only relevant in that it was used to test the allocation of resources on a single arrangement of lines and was pitted against humans who had only twenty minutes to test their own assumptions. In real life, humans would accumulate much more experience and judgement and the model would never be able to fully represent the complexity of real life. An algorithm would be likely to give very exact answers to oversimplified questions that may not be appropriate for real life. This line of questioning also fails to acknowledge that the franchisee discretion in line creation is not being tested by the calculations of this model. It is the variation in network creation that appears to be the most important aspect of the design process that is not being ‘solved’ by the algorithm. There is also the inherent variation allowable in the line running costs that would influence the real-life performance of individual lines; this is not picked up in this very simple model. There may be an application for computers for these functions, but it was not tested here and assumptions about superior computer performance over humans cannot be said to be tested by this experiment. As in the Dolomites, there was no planning in the Community Franchise regime game; it was purely individuals seeking their own best advantage. The lack of planning in the Community Franchise regime game did not interfere with the capacity of that regime to produce social welfare results that on average were better than that of the Government Enterprise regime. Such a result is consistent with the observations of Patchell (2014) where he documents the relative performance of ‘Integrated’ (i.e. Government Enterprise regime) ski resorts in the US with the ‘community’ ski resorts in Japan. He notes that the capacity of the ‘community’ schemes to find solutions to common problems produced results at least as effective as the ‘Integrated’ resorts where the direct action of centralised planning was, on the surface, a more rational way to allocate resources. There was no advantage displayed by the Government Enterprise ‘Integrated’ regimes in his observations and that is consistent with the results of this experiment. 9.7.2 Regime Performance No human could match the performance of Evolver in generating social welfare. However, the differences between the performances of the two human groups playing the different regime games were not significant. Although Evolver was able to generate the most social welfare, the figures for the two regimes were both close to each other and not too far from the algorithm result. The two regime games played by humans were, on average, different to some degree (14%-19% less), but the top human performers were much closer (3% to 4% less). The human results are noted to be achieved in only a very small time period and not much should be read into the algorithm results. If humans
Page 151 of 259
had more time, they would almost certainly have done better, but the objective was not to try and achieve maximum results; it was to compare the two approaches on an equal footing, an objective achieved in the limited time allocated for each game version. Those high scoring Community Franchise regime games were achieved by a variety of participant groups. It was noticeable that those groups with quite average performers from the ‘beta’ version of the Government Enterprise game were capable of obtaining significantly higher results for social welfare in the Community Franchise regime game. There does not seem to be any consistency in the type of participant who achieved high group social welfare scores. Observation of the participants interacting with each other and the computer suggest that personality type is a possible explanation. This is outside the scope of this study, but it seems that those who take greater risks and who are more exploratory and curious make the discoveries that matter to overall system performance in both versions of the game. An interesting observation is that the profitability of the groups of lines in the Community Franchise regime game was substantially different from those same line groups in Evolver. The profit figures of the groups of lines used in the Community Franchise regime game and comparison of those individual profit figures with those produced in the highest-scoring Evolver game are shown in Figure 11 below.
$10,000,000 $9,104,920 $7,289,066 $8,000,000 Humans $6,510,338 $6,016,585 $6,061,465 Evolver $6,000,000 $4,297,832 $4,000,000
$2,000,000
$- Group 1 Group 2 Group 3 Profits Figure 11 Comparison of Group Producer Surpluses for highest scoring Human Participant Line Group Profit Results with the Highest found Evolver Result These line groups were shown in Figure 6.6, where Group 1 consisted of Lines 1, 6 and 7; Group 2 consists of Lines 3, 4 and 8, and Group 3 consists of the last two Lines 2 and 5. It can be seen from an examination of the profits of each of these three groups that there is variation in results that would, in real life, be less likely to occur if these groups of lines were in private ownership. Individual private owners would not be expected to be interested in achieving greater overall social welfare if that meant less profit for the group of lines in their ownership. This may explain a portion of the greater social welfare achieved by Evolver. Human participants in the Community Franchise regime game were intent only on optimising their own situation. This compares to the Evolver allocation in the Government Enterprise game which has the effect of spreading the profits less evenly over the three
Page 152 of 259
groups in its quest to maximise total social welfare. The relative performance of each group of lines was not a consideration of Evolver in solving the Government Enterprise regime situation. Another effect of the Evolver approach as noted earlier was that there were fewer passengers transported than in the final equilibrium examples achieved by humans in both the Government Enterprise regime game and the Community Franchise regime game. So the maximisation of social welfare, as determined by this particular model of reality, is both somewhat unintuitive and means that fewer numbers of passengers are carried. 9.7.3 Complexity of the Model All solutions appeared to be unintuitive and the results unexpected by the participants: even with this simple model, the optimal allocation of resources was not obvious. Almost regardless of the starting theoretical point of view of the participant, their actual allocations of resources tended to shift when the reality of the social welfare figure in the Government Enterprise regime, or their own profit figure in the Community Franchise regime game, altered unexpectedly. The network effect produced some completely unexpected results. For example, those for Line 1, traffic lights or adding vehicles had little effect on social welfare or profits. These results were generally accepted and increases or decreases in resource allocation were seen to be rational and desirable. However, there was enormous variation in the effectiveness of the approaches, from those who were relentless in their pursuit of the highest figures and would experiment till all options were exhausted, to those whose curiosity or experimental temperament was more muted and who were satisfied with results that more exploratory experimental activity may have improved. It was obvious from observation of many games that results were path-dependent. So, if say expensive traffic lights or intersection transfer stations were installed too early, the participant’s score would not be increased. If, however, good cash flow had been built up by firstly increasing vehicle numbers and building up cash flow, and there was, therefore, the capacity to borrow efficiently, then a participant’s score would be increased, sometimes substantially. Not all participants seemed to appreciate what was happening and that such matters as the quantum of free cash flow were important in making decisions about items requiring large capital expenditure. In short, there was great variety in the choices and in the model responses, which were influenced by the timing of their decisions. This is also reflected in the fact that there are virtually no two results that are the same, despite a measure of perceived stability being reached by all participants. The experimental model is very basic. It has a truly minimum network, a simplified and idealised topology, a single-vehicle type, only a small number of intersection conditions, to name but a few variables that normally affect the efficiency of any transport system in a real city. However, even this simple model was not able to be easily solved by the participants. In fact, it was also difficult for the genetic algorithm to solve. Several runs of over 10-hour periods, where many thousands of potential solutions were evaluated, did not result in many of the same solutions. This is a reflection of the mathematical complexity of even this simple model of a network. This should cause some reflection on
Page 153 of 259
the assumptions made in all city modelling that determines the infrastructure investment made in Government Enterprise regimes. It was a necessary objective of the experimental game development that the model is kept relatively simple. Early versions of the game had to be substantially simplified to enable solutions to be ‘discovered’ by participants in a reasonable time frame. This allowed for resolution of the available options and the attainment of resource allocation ‘stability’ in a shorter period than a more complex arrangement. The simplicity of the model and the need to limit the number of options and variable parameters gave rise to a concern that it might have been too easy a problem for most people to solve and thus not provide any significant information on the performance of the two regimes. A model that was too simple would have failed to allow any distinction between the two regimes and thus not give any scope for testing the hypothesis. This concern was unfounded, as even with this very simple model, it was too complex for any participants to reach the optimal solution discovered by Evolver regardless of the regime version of the game. It is a reasonable expectation that participants with enough time would be able to achieve better outcomes, (i.e. higher profits and social welfare) if the model was more complex to better reflect reality. Suggestions for future research with this model, based on more complex and realistic models, are made in the Conclusions to this study in Chapter 12. 9.7.4 Analysis of Results by Order of Play and Age Further use was made of the ‘beta’ game series determining the suitability of participants. This data set provided some insight into other aspects of performance. The results were examined when grouped according to other criteria, i.e.:
• The order in which the groups played each version of the game. • The age brackets of participants, being below 30, between 30 and 60 and above 60 years of age. As discussed in Chapter 6, to overcome the inherent bias in playing one version of the game before the other, the orders of play were varied so that equal numbers of each order were played. This cancelled out any advantage derived from seeing the workings of the game before playing the other version. In this section, the results of both versions are analysed to establish if there was any substance to this potential bias, and if there was an advantage to maximising the scores in the second playing in whatever regime. This effect was analysed using the scores from the ‘beta’ series of results. The scores for each version of the ‘beta’ series of games are as follows in Table 23:
GE score when GE score when CF score when CF scores when played before CF played after CF played before GE placed after GE 1 $ 14,077,796 $ 13,737,693 $ 20,016,571 $ 19,544,806 2 $ 19,872,725 $ 19,771,159 $ 19,896,296 $ 19,957,696 3 $ 19,011,038 $ 18,890,122 $ 19,872,725 $ 19,943,654 4 $ 18,222,703 $ 17,460,036 $ 19,872,725 $ 19,923,148 5 $ 17,838,157 $ 16,821,573 $ 19,408,937 $ 19,835,594 6 $ 17,807,764 $ 16,821,573 $ 18,767,255 $ 19,833,475
Page 154 of 259
7 $ 16,642,283 $ 16,407,145 $ 17,843,559 $ 18,004,468 8 $ 16,394,050 $ 16,004,428 $ 17,699,970 $ 17,843,559 9 $ 15,607,067 $ 15,687,902 $ 17,626,648 $ 17,755,225 10 $ 15,095,803 $ 15,178,474 $ 17,626,648 $ 17,686,217 11 $ 14,790,836 $ 14,957,499 $ 15,627,052 $ 16,827,551 12 $ 14,192,717 $ 14,838,425 $ 15,548,552 $ 15,388,386 13 $ 13,709,133 $ 14,397,807 $ 15,497,583 $ 13,598,785 14 $ 12,610,843 $ 11,962,160 $ 15,481,066 $ 13,279,978 Averages $ 16,133,780 $ 15,924,000 $ 17,913,256 $ 17,815,896 Table 23: Social Welfare production from ‘beta’ games: Results due to Different Orders of Play A t-test with unequal means resulted in a p-value of 0.7943. With a null hypothesis that the scores would be equal irrespective of the order of play, the test failed to reject H0. The variation as observed between the two different games and their order of play was not expected but is not seen to be a real concern having been factored out of the experiment in any case. The three ages groups that were identified, i.e. below 30, between 31 and 60 and over 60 years, did not show much variation. The consolidated average results for these three age categories are in Table 24 below:
Age Category <30 31><60 >60 Average SW $ 15,958,211 $ 15,941,108 $ 15,831,559 Number of Participants 5 27 52 Table 24: Proxy Social Welfare production from ‘beta’ games: Results of Age Grouping All results are within 0.01% of each other. Age and experience did not have much apparent impact on the results. 9.8 Research Question 2: Competitor Cooperation The second research question was concerned with cooperation between competing participants. This question was stated in Chapter 6 to be: Are competing line owners able to cooperate so as to enhance network performance? This gave rise to Expected Outcome no 2, which stated: That in a game context, there will be sufficient cooperation to provide an acceptable allocation of resources compared to that which is achieved through central planning. Participant behaviour during play was observed. Transactions between the participants when cooperation to improve network were required to improve network performance was noted. Along with specific observations of behaviour relating to cooperation, noteworthy behaviour on general playing techniques was reported on. Peculiarities of behaviour were noted to assemble a picture that helped explain and illuminate the results.
Page 155 of 259
Participants cooperated in the Community Franchise regime game, even if it meant that one competitor may benefit more than the other. As long as there was a positive outcome for both parties, most negotiations were simple and productive. In virtually all games there was meaningful cooperation amongst competitors in negotiating outcomes for the development of intersections. Each would consider the effect on their own bottom line and then either accept or reject the offers of a split in the contribution to common infrastructure. This would vary from 100% being assumed by one of the parties, through a variety of proportional contribution, e.g. 60/40, or 30/70, but very often just a straight 50/50 split of the costs to both parties. Negotiations were achieved with a notable lack of rancour, but all conducted in an atmosphere of considerable competitiveness. Each of the participants, almost without fail, wanted to win in the sense of having a maximum score or a personal best for their own profit figures. Each participant tried as hard as any player in a game of Monopoly or Scrabble, to achieve their own selfish aims. However, this did not inhibit cooperation when there was a personal gain to be had. Some individuals drove harder bargains than others, and some were quite reserved in their bargaining manner. However, it was virtually unknown for people not to find some level of investment if it were found to be suitable for them. No one engaged with another party to do that party a favour: it was all on a commercial basis of selfish gain. This is consistent with expectations of normal commercial business behaviour, but it was helpful to see this working in the experiment as it served to underscore the extent of cooperation there is in competing businesses of all sorts, as long as cooperation does not put the participants at any disadvantage. It suggests that win/win situations will tend to be implemented. This is consistent with theories of group behaviour where one’s future welfare can be achieved by the generosity displayed towards fellow industry participants, whether competitors or not (Fehr and Gächter, 1998, Axelrod, 2006). In a network environment, it became obvious to participants that certain improvements were better for the whole network and could be advantageous to all players, even when they were not involved in a particular intersection negotiation or line capacity decision. Networks have this characteristic of seemingly perverse unrelated consequences resulting from actions in remote parts of the whole system (Laird et al., 2005). It was not lost on most participants that the best allocation of resources generally improved all operators’ profitability. This was not only a consequence of the intersection improvements; it was an observable effect when additional vehicles were purchased to operate on a particular line. This could have the consequence of increasing patronage overall and cause competitors to have increased profits at no expense to themselves. As games progressed, this network effect became quite evident and led to an awareness of how cooperation could maximise individual profits. However, it did not lead to any discernible desire to gratuitously increase the profits of other participants: each participant concentrated on their own bottom line firstly and would only contribute resources if it had some advantage for their own business.
Page 156 of 259
This result showing cooperation may have been different if participants had been in a laboratory setting, i.e. they were separated from each other and not able to directly communicate with other participants. The possibility of this is discounted as having any relevance to the experimental objectives. In real life, the different participants of negotiating parties are normally in direct contact with each other and can see, feel and read the psychological state of the various parties to any negotiation. This contact provides essential information for people to reach conclusive outcomes. A situation where people cannot interact in a normal commercial setting, as prevails in an economic laboratory, would be imposing a condition that would act against the attainment of realistic and believable outcomes. This consideration is thus a confirmation that the experiment as designed is an appropriate one for the testing of this research question. The behaviour observed confirmed that competing participants consistently cooperated when it was in their interests to do so, i.e. when their own profit could be enhanced. It is noted that there was not a single case observed where participants did not cooperate when it was in their interests to do so. These observations and the lack of any contrary evidence support the Expected Outcome no 2 that there will be cooperation arising from the incentive to maximise the profit of each participant. This offers considerable support to the expectation that competition between line owners would not adversely affect the need for cooperation to achieve better outcomes for society as a whole. 9.9 Summary This concludes this section of the study which has tested and failed to reject the first hypothesis stated in Chapter 7 that both regimes will deliver similar amounts of social welfare. As well, substantial support was provided for the Expected Outcome no 2, namely that in a game context, there will be sufficient cooperation between competing line owners to provide an acceptable allocation of resources compared to that which is achieved through central planning. This result does not reject the proposition that a Community Franchise regime could be used in place of a Government Enterprise regime without endangering social expectations. In the next chapter, a complementary subjective view of the overall question of this thesis, i.e. the applicability of a Community Franchise regime to urban public transport, as investigated through a survey of experts, will be reported on.
Page 157 of 259
Page 158 of 259
CHAPTER 10 EXPERT SURVEY 10.1 Chapter Content This chapter provides more detail of the expert survey methodology. It outlines the dimensions of the survey questions and the expertise of the respondents that it was appropriate to seek answers from. It then covers the questions that were included in the survey and the means used to obtain responses to those questions, both quantitative and qualitative. 10.2 Dimensions and Questions Questions target specific operations of the franchising regime that are different from those of normal commerce and government to identify problems that might result from its application to an urban situation. The range covered by the questions is generated from the consideration of the relevant dimensions of concern and interests (Okoli and Pawlowski, 2004). The questions asked are framed to encourage a full range of responses to draw out multiple views. The list of dimensions is provided below with a discussion of the reasons for inclusion as well as the type of respondent sought for this dimension and the questions that would then be asked of the panel. This extent of the six dimensions of inquiry identified is as follows: 10.2.1 Ownership and Direction Ownership of public services can be a contentious issue. There has been significant debate about the privatisation of public transport services and other traditional publicly owned services in NSW like power generation, lands title information systems, etc. The distinguishing feature of the Community Franchise regime is government ownership of the franchisor and private ownership by franchisees of operations. Such questions of whether such an arrangement is practical or has inherent problems detrimental to the social order have been addressed by such researchers as Castro and Maddock (1997 ch13). In the case of Australia Post, they found that politically sensitive pricing of the domestic postal service was influenced by the Australian Government as owners of Australia Post and that sufficient control was exerted to have priced in accordance with parliamentary expectations. This is similar for other services required to provide Universal Service Obligations reported on in other jurisdictions (Bel and Calzada, 2009). Ownership and direction are questions where the political impact of such an arrangement, as well as its practicality, has some importance. Particular insight may be given by local or state politicians who see the issues from the point of view of parliamentary or local government representatives. It will identify if, in the context of the situation in NSW, whether or not this issue would face political hurdles if it were to be implemented. Answers to questions on this topic will establish if the ownership model is sufficient to satisfy those who would be naturally against loss of government control. It is a question essentially of a political nature with management overtones to which appropriate
Page 159 of 259
answers can be provided by virtually any person in the social system. All experts would have a relevant opinion on this dimension. Questions will need to cover ownership and board control issues and three questions emerged to cover these concerns as follows: 1: A Community Franchise is a commercial entity completely owned, and with a board controlled, by Government. Public ownership of commercial entities already occurs with such companies as Australia Post and the Australian Rail Track Corporation. To what extent do you see problems associated with the ownership of a commercial franchise transport company that would set rates for passengers, collect all fares, distribute collected monies to franchisee operators, and generally operate in accordance with the objectives set by the company board? 2: To what extent would a politician, or his nominee, be an appropriate board member of a franchisor company? 3: What types of individuals would be more suitable than or should supplement politicians to represent the government on the board of such a company owned by government? 10.2.2 Budgets and Corporate Control The application of a Community Franchise regime has far-reaching implications for government budgets and corporate participation. In a Community Franchise, all public transport planning, assets and operations are removed from the government consideration and there is no requirement for the government to fund public transport unless deliberately subsidising fares. Such issues as the privatisation of government assets have been an ongoing point of debate for decades, especially since the mid- 1980’s when this process was experimented with on a global scale. A consistent view is that there is a need for the retention of government oversight and control of most industries, especially those with monopoly tendencies (Posner, 1978, Spence, 1975, Wolfstetter, 1996). An anticipated concern of any application of a Community Franchise will be as to what subsidies to operations or to which categories of citizens for state purposes are required. A Community Franchise has a particular effect on both private ownership and government control through its ownership and structure and questions here are sought to explore the effects on government budgets and the flow-on effects of such policy. The type of respondent required is, essentially, as for question 1 but will be particularly relevant for those in finance, those who examine the finances of infrastructure and transport operations particularly, e.g. large auditing firms, think tanks, public economic commentators, research institutions, etc. Two questions are used to address this issue, as follows: 4. In a Community Franchise, the government owned commercial entity that acts as the franchisor must operate profitably, setting prices for fares, the amounts to
Page 160 of 259
pay franchisees and its own remuneration as a percentage of gross turnover. To what extent would you expect the proposed regime to be able to control public transport spending compared to current arrangements? 5: Control by the franchisor over the franchisee operators is the way that franchise regimes operate. In this Community Franchise case, fares are collected by the franchisor. Payment by the franchisor to franchisee operators would depend on the compliant operations of each franchisee To what extent do you feel that the government owned franchisor has the capacity to look after the interests of the disadvantaged in the community and the electorate generally under a Community Franchise? 10.2.3 Network Planning and Operations In the Dolomites, the controlling franchisor does not engage in any network planning: all development proposals are by individual franchisee operator initiative which proposes the routes and mode and has subsequent control of the frequency of services over and above the proscribed minimum service frequency, levels of appointment to vehicles, etc. for each line that is proposed for inclusion in the network. This situation contrasts with what happens in conventional government enterprise systems where the controlling entity decides on the network layout, the modes and vehicles to be used, the timetabling and virtually every aspect of the service operation. One thing both regimes share is that the controlling entity sets and collects the fares. At most a private operator currently providing services for a Government Enterprise undertakes the task defined in detail by the controlling entity for a price which is generally determined by a tender process. Significantly, the contractors of government services do not influence the fares charged; theirs is a typical defined contractual arrangement. This concept of overall versus bottom-up asset allocation was explored in the computer game, where it was shown that there is no significant difference between the social welfare produced in either arrangement. However, it is a topic that generates much interest and it would be illuminating to see how a broad-based audience, including experts from business and the franchise industry itself, may regard the potential of this approach to asset allocation. This concept of top-down planning versus a bottom-up evolution is one that can be explored through the following question: 6: In the Dolomites, all proposals for network improvement come from individual franchisees which see a potential for a profitable line. Seeing a potential for a profitable venture, based on the rates being offered for the carriage of passengers, the franchisees respond with proposals for the network improvement. To what extent do you think that it is efficient for planning to be delegated to operators with only the profit motive and no regard for the overall efficiency of the network? Urban infrastructure is generally of a scale that government involvement could be thought to be essential for its planning, financing and operation. If a Community
Page 161 of 259
Franchise were to be applied to urban situations, then the planning, financing and operations of infrastructure shifts to the private sector. It is thus important that private industry is not only able to able to finance, plan and undertake these projects but then to operate them in ways that benefit the public which is the ultimate beneficiary and on whose behalf it is being constructed. A further question that emerges from this dimension is as follows: 7: The largest urban infrastructure projects (like Sydney’s new metro system) would become the responsibility of privately-owned franchisee companies to plan, finance, construct and operate. To what extent do you think that privately owned franchisee companies would be capable of planning, funding, constructing infrastructure and then operating their lines of urban public transport? 10.2.4 Control and Maintenance of Standards A concern of any government is the control that its administration has over external operators undertaking work on its behalf. The controls able to be used in a Community Franchise differ from those of direct contracting of, say, bus services in a traditional government designed and run network. They differ also from complete privatisation of a service due to the continued involvement of the government owned franchisor as an active regulator in the Community Franchise regime structure. Questions of control over the providers of the transport service emerge that can be addressed by the following two questions: 8: In a Community Franchise, franchisee behaviour is controlled by the government owned franchisor which sets rules of operation and controls the cashflow to the franchisees. To what extent do you think that the structure of a Community Franchise would allow effective government oversight of the public transport system. The type of respondent is essentially the same as for question 1-5 but requires skills in and knowledge of large-scale infrastructure, its financing, ownership and operation. Such attributes may be found in government organisations, think tanks and research organisations looking at infrastructure as well as large scale industry and those professional organisations that serve them. 10.2.5 Fares Prices of newly privatised industries can fluctuate from previous levels and would be a concern to government and voters. Government, financial professions and research organisations are the types of organisations that would be well placed to have relevant opinions on this issue. The question that emerges from this dimension can be expressed as: 9: In a Community Franchise the government owned franchisor must set the level of fares by balancing the need to encourage investment (i.e. keep fares high) with the desire to keep fares at politically acceptable levels, (i.e. keep fares low). What
Page 162 of 259
would, in your opinion, be the likely impact on current fares levels by the implementation of a Community Franchise? 10.2.6 Alignments of Interest and Regulatory Capture The question of where peoples’ interests lie in any social structure is important to most people looking for social justice and economic efficiency. It was observed in the Dolomites that there appeared to be an appropriate alignment of interests between the various participants in the ski lifting industry and the issue suggests the need for opinions from a range of experts. All of those in previous sections would be well placed to comment on this dimension. The question that emerges from this dimension is whether it would have a positive effect on alignments of interest if applied to Sydney. which can be questioned as follows: 10: In a Community Franchise, the government owned franchisor must operate profitably and must set fares at levels that encourage investment by private franchisees. The franchisor’s only income is from a percentage of ticket sales. The only income of franchisees is being paid for each passenger/km. The only option for passengers is to switch routes or mode to avoid a franchisee operator that is not satisfactory. To what extent do you think that the interests of the passenger, the franchisor regulator and the franchisee provider, coincide? It is of concern to society if a particular structure can be exploited by sectional interests to the detriment of social equity. A question to explore this dimension can be expressed as: 11: In a Community Franchise there might be a temptation for the government- controlled board to favour particular franchisees for some future reward that is not in the public interest (known as regulatory capture). For example, if the franchisee regulator was to have relationships with franchisee operators, there may be pressure or inducements to change rates to favour a particular franchisee at the expense of another operating franchisees of different modes. (Note that all operators receive the same payment/passenger/km for any particular mode). To what extent do you think that the interests of the franchisor and the operating franchisees may overlap and create conditions for regulatory capture? 12: To what extent might this be different in an existing government enterprise arrangement. The same set of respondents as for the previous question would be suitable for these. 10.2.7 Overall Feasibility and Critical Features This is a question of the applicability of the particular franchising regime, as has been defined in this study, to the public transport task. This overarching question of whether this regime is applicable to urban public transport is addressed by seeking a response from all participants to the question as follows:
Page 163 of 259
13: Acknowledging that this regime has operated for over forty years in the fixed line ski lifting markets of Europe, what is your assessment as to this business regime being able to function in an urban public transport setting? To allow getting an opinion about any conceivable factor that might, even remotely, be detrimental to the application of a Community Franchise, the following open-ended question suitable to all participants is provided: 14: Are there any features of this business regime that in your opinion would rule a Community Franchise out of consideration for its application to an urban public transport setting? 10.3 Respondents The types of respondents sought were experts in the identified dimensions, as discussed in Section 7.4.5 and potential respondents were identified from each dimension. These people had backgrounds in transport, finance, franchising, government and researchers from universities and specialist think tanks. Snowballing from an initial group helped identify additional respondents. Multiple organisations were identified, and many were contacted to establish if management would support one or more of their staff undertaking the survey. Seeking additional candidate organisations was only from the organisation or its departmental head and was not sought from the intended or actual respondent. The head of the appropriate section in these organisations was contacted and it was requested that a qualified, economically literate member of their team be nominated and made available to answer the series of questions on the topic. Only those organisations that agreed to participate in the survey were included in the results. No participants were sought from organisations that did not want to be formally involved. Contact details of the individual nominated, including direct email connection, were obtained from the responsible officer providing the opportunity. The email address was used to provide the questions and support material to those individual respondents. This process resulted in letters being sent to 16 individuals, 13 of whom responded with a valid survey. The final list of those who accepted the invitation to respond is listed in Table 25 Dimensions and Participants in the next chapter. 10.4 Mechanism for Data Collection The questions were sent to each respondent as a Word document attached to an email. The respondent was requested to open the document in Word and then to fill out the form in that program. A response to each question was sought in two ways, one as a scaled measure and the other a free form commentary of the scaled response in an expandable text box following the scale. This mechanism for the scaled response was in the form of an invitation to shift an indicator of their response on a scale as follows:
Page 164 of 259
1 2 3 4 5 6 7 8 9 10
Then the opportunity to embellish, reinforce, explain or amplify this response was provided in a text box as follows: Please provide features and reasons in the text box below (The opinion of the Expert placed here…..)
The results of answers to the questions is documented in Chapter 11, and the explanatory document sent to participants with this list of questions is included in Appendix C. When completed in the Word document the respondent was requested to return to the researcher electronically in a PDF format as follows: When completed, please return a copy to the sender, using the following procedure from within the Word program:
• File • Share • email • Send as PDF Send to: [email protected] 10.5 Research Question Three Summary This chapter provided the detail of the methodology for the survey of experts, outlined the dimensions of the inquiry covered, the questions posed, the rationale of expert selection and the means of collecting data. The results of the survey follow in the next chapter. These provide the multidimensional expert view of the applicability of the Community Franchise business regime to the urban situation of Sydney and identifies the potential benefits and problem areas of such an application.
Page 165 of 259
Page 166 of 259
CHAPTER 11 RESULTS OF EXPERT SURVEY 11.1 Chapter Contents This chapter provides an analysis of the numerical and written opinions of those who participated in the expert survey, together with a discussion of the key issues raised by survey respondents. Of the sixteen experts approached, thirteen provided valid responses. Their written contributions have been analysed in NVivo and the numerical results of scaled values summarised in Excel. In the analysis below, each question is stated followed by the numeric value of the sliding scale qualitative results shown for all participants. The Average and the Standard Deviation of those scores are listed. Following each numerical score, there is a brief discussion interpreting both the numerical and the written notes provided by the experts. The full responses are included in Appendix D containing all questions and comments from participants. The results were provided to each participant with an invitation to provide their feedback, and the responses from the only two who did so are discussed. Then the numerical data is presented with a discussion on the implications of those figures. Following these results, there is a discussion as to the major concerns identified in respondent comments. The chapter concludes with a summary of the major findings. 11.2 Participants Sixteen approaches were made to potential participants, of which thirteen provided valid responses. These participants cover the five dimensions to be examined and are listed in Table 1 below. These show colour coding that will be used throughout this analysis.
Dimension Name Expertise and Background 1 Franchising Jenny Buchan Professor/franchising UNSW 2 Trevor Daley Franchisee Owner/Chartered Accountant 3 Matt Wheatly Head Australian Franchisee Association 4 Government Tim Rogers Head NSW State Department of Local Government (rtd) 5 Peter Warrington Sydney City Council (public transport specialist) 6 Roy Bishop NSW Parliamentary Advisor 7 Finance David Naylor Senior Partner Australia wide Accountants ChanNaylor 8 Graeme Troy Specialist Chartered Accountant small business advisor 9 Planning Paul Forward Head of Infrastructure Australia 10 Rebecca McPhee Senior executive TfNSW 11 Ian Muncaster Senior construction executive Lend Lease (rtd) 12 Operations Stephen Rowe Founder Bus Ways (international bus operator) 13 Wijnand Veeneman Professor/transport Delph Technical University Table 25 Dimensions and Participants Some participants had expertise in more than one category, but they are classified as to their principal strength. The range of expertise enlisted was sufficient to provide between two and three participants in each dimensional category. 11.3 Written Results All raw results (provided in appendix D) are organised firstly in separate items for each question, and then by respondent, with separate items for each participant. The scaled responses are followed by averages and standard deviations. The means show the
Page 167 of 259
average of all participant scores to each question, and the standard deviations show the extent to which there is consensus or disagreement amongst the participants on each question. A summary of the numerical means and standard deviations of group responses is shown in Appendix E. The results and commentary now follow, grouped into the areas of inquiry. Each question is taken in turn and shown graphically with the scaled response provided, colour coded for participant dimensional grouping. The colour and size coding is shown by the following graphic technique of Figure 12:
Figure 12 Coding of Participant Responses by Expressed Confidence Level Each graphical representation is sized to match the degree of confidence that the participant reported for the answers provided. Each respondent’s numerical choice for a question is located on the graph with the scaled representation of their degree of confidence in their response. This representation allows all information concerning the scaled response and the degree of confidence to be evident for each question. The mean of all participant choices is displayed with the vertical dotted line to show the average score of the entire participant group. The standard deviation provides a measure of group consensus.
Page 168 of 259
11.3.1 Ownership and Direction
11.3.1.1 Question 1 You may be aware that the Australian Government is the sole owner of the corporate entities Australia Post and the Australian Rail Railtrack Corporation. To what extent do you see problems associated with the ownership of a commercial franchise public transport company that would set rates for passengers, collect all fares, distribute collected monies to franchisee operators, and generally operate in accordance with the objectives set by the company board? (1 none - 10 many)
11.3.1.2 Q1 Interpretation Overall, the average score to Q1 of 5.5 indicates a neutral stance to the concept of government ownership, but, with the standard deviation of 3, a pronounced divergence of views. While the scores varied widely, there were very few written comments that indicated doubt as to the government’s capacity to own and operate commercial entities, just that there may be problems with it. The comment from the Finance (red), Government (green) and Operator (purple) perspective were generally positive and there was the least support from those in Planning (yellow) and Franchising (blue). One comment of note was from the Dutch academic with knowledge of the government-owned commercial transport operators in that jurisdiction, who commented that “..this seems a good basis for the securing of the wider public good”.
11.3.1.3 Question 2 To what extent would a politician, or his nominee, be an appropriate board member of a Government owned commercial franchise company?
(1 good - 10 bad)
Page 169 of 259
11.3.1.4 Q2 Interpretation There was great variability of response to Q2. There was a quite solid average of 6.5 against such a proposition, with the extremes of a 1 and multiple values of 10 both represented. There was much variation between groups in Q1, with a large standard deviation. However, this difference in response to Q2 was not reflected in the category of respondent: all colours are represented on both sides of this divide. Board membership is shown by the responses to be a contentious issue and a different question from government ownership where there was more unanimity between dimensions. The written comments to Q2 indicated that some respondents were entirely comfortable with the concept of political representation on the board of the commercial franchisor, but those in infrastructure provision and regulation thought that politicians would either not be suitable or it should operate independently of politics. From some of those with direct experience with government and with statutory authorities, there was an apparent acceptance that politicians may be suitable board members with a little expressed reservation with only a minority view from this perspective to the contrary. Senior government personnel expressed the sentiment, likely held by others, that a politician would allow politics to interfere with judgement.
11.3.1.5 Question 3 What other types of individuals would be more suitable than, or should supplement politicians to represent the government on the board of a Government owned commercial franchise company? Please use text box below There was no scaled response for this question. The comments are in Appendix D.
11.3.1.6 Q3 Interpretation There was general written agreement as to board membership beyond politicians. Overall, there was strong support for a diversified board membership with appropriate skills. As to who would be a suitable board member the suggestions included expert transport specialists, strategic thinkers, those with a finance background, town planning experts, commercial lawyers and the respected businessperson. Membership of the board, except for the political component, was an issue on which there appeared to be a broad agreement from all participants. 11.3.2 Budgets and Control
11.3.2.1 Question 4 In a Community Franchise, the government owned commercial entity acting as the franchisor must operate profitably, setting fare levels, the amounts to pay franchisees and receiving its own remuneration as a percentage of gross turnover. To what extent would you expect a Community Franchise to be able to control public transport spending compared to current arrangements
(1 very well – 10 very poorly)
Page 170 of 259
11.3.2.2 Q4 Interpretation There was again great variation in responses to this question on financial control of spending. There was a very slightly positive average (5.4) and the Financial group was the only one with a consistent (positive) group response. All other dimensional groups were spread over the complete spectrum with representation on both sides of the divide. In written comments, on the negative side, there was the hard-line response from the Planning dimension and the possibly realistic response of Government, Finance, Operations and Franchise dimensions which would promote a less optimistic view of public servants and their spending controls. An interesting feature was the more positive responses of some of the Government dimension (public servants themselves), local council employees with public transport responsibilities and academics. All of these participants would have some experience in these matters, and there was no real consensus in this question even within dimensions.
11.3.2.3 Question 5 Control by the franchisor over the franchisee operators’ operations is a feature of franchise regimes operations. Fares are collected by the franchisor. Payment by the franchisor to franchisee operators would depend on the compliant operations of each franchisee. To what extent do you feel that a government owned franchisor in an urban application of a Community Franchise regime has the capacity to look after the interests of the disadvantaged in the community and the electorate generally?
(1 very well – 10 not at all)
11.3.2.4 Q5 Interpretation The mean of 6.2 for the Q5 results came from a large range of opinion and was in overall terms clearly negative, i.e. that the government owned franchisor would not be able to control franchisees and look after the interests of the disadvantaged. There was a limited but quite solid positive opinion at the opposite end of the scale, generating a high standard deviation score of 2.7. The positive end had the support of Franchise,
Page 171 of 259
Government and Finance dimensions, but all these dimensions were represented in the negative responses as well. (Three notable readings in a minority position against the proposition was from a senior public servant, parliamentary advisor and state statutory body head who arguably might have most insight to the situation). The written responses to this question suggest that the experts view the type of administrative structure for the franchisor and its incentives and capacity to control costs is likely to be a controversial question. This suggests it will be considered of great importance for any potential application of this regime. 11.3.3 Network Planning and Operations
11.3.3.1 Question 6 All network-planning in the Dolomites is undertaken by franchisee operators who research, plan and then propose the lines on which the franchisee feels has a sustainable profitable future. If applied to an urban area, decisions as to the routes of new lines and alterations to routes and modes of existing lines will be made by the private franchisee operators responding to their perceptions of passenger demand. To what extent if a Community Franchise was applied to a city do you think that the network planned in this manner would operate efficiently and meet passenger demand?
(1 very well – 10 not at all)
11.3.3.2 Q6 Interpretation The Q6 mean of 4.8 suggests an overall acceptance of the idea of an evolution of a workable network, as opposed to a centrally directed one but, from the range of opinion represented, there is not a clear conviction from some participants that it would work. Those from the Operational dimension were clearly opposed to the proposition and those from the Finance dimension most in favour. The Government dimension, as well as the Planning dimension, was split, but the negative opinion of the Planning dimension was a thinly held one and the positive view was solidly held. The Franchising dimension held a favourable position with again, the slightly dissenting voice being held with less confidence.
11.3.3.3 Question 7 All urban infrastructure projects (from the largest like a metro line to small lines of just a few buses without even a terminal) would become the responsibility of privately-owned franchisee companies to plan, finance as well as operate. To what extent do you think
Page 172 of 259
that privately owned companies would be capable of planning, funding, constructing infrastructure and then operating their lines of urban public transport?
(1 very well – 10 not at all)
11.3.3.4 Q7 Interpretation In support of the numerical score, the written comments indicate that most respondents thought the private sector was well capable of undertaking the business of organising, financing and running public transport, with an overall mean of 4.6. It was notable that one in the Planning dimension did not have high regard for the sector being able to plan, fund, construct then operate transport infrastructure, a view shared by several franchise experts. This is an interesting comment as it would be organisations from this grouping who currently, and potentially in an application of the regime, would be the ones to undertake the supply of infrastructure and its subsequent oversight, installation and operation. In contrast, others in the Planning dimension were solidly in agreement. There was a slightly wider range of confidence than other categories of inquiry. Some from the infrastructure, accounting and franchisee backgrounds indicating that they felt unqualified or uncertain of their capacity to contribute to this question. 11.3.4 Control and Maintenance of Standards
11.3.4.1 Question 8 In a Community Franchise, the only control over privately owned franchisee operator behaviour is by the government owned franchisor setting the rules of operation for franchisee operators, policing them as the franchisor and controlling the cashflow to the franchisees. To what extent do you think that the structure of a Community Franchise would enable the government owned franchisor to exercise proper control over franchisee operator standards and behaviour?
(1 not at all – 10 very likely)
Page 173 of 259
11.3.4.2 Q8 Interpretation While there were several notable sceptics to this proposition, the average score was on a neutral setting with a moderate standard deviation. The Finance dimension did not see it as a problem, but those from the Operations dimension saw it in a negative light. However, several respondents well versed in franchisee obligations, and thus providing a relevant perspective, suggested in their written comments that the government was well able to provide the control and to maintain the standards as the franchisor. The most noticeable voices thinking that government would not have control included both parliamentary and businesspeople from the Government and Operations dimensions. One franchisee made particular mention of the control inherent in the way that cash flowed through the government in the first instance and only on to the franchisees for compliant performance. However, this feature of government franchisor control of the cash flow was identified as a distinct negative by one from the Franchise dimension in academia. This control over cash flow by franchisors was thought to place franchisees in a vulnerable position open to exploitation. This would be of obvious interest to potential franchisees and to the board members of the franchisor, who are entirely dependent on franchisees for delivery of the service. There was a wider range of confidence scores (from 1 to 10) which was a slightly unusual occurrence, the extreme outliers in this instance being from franchisees and government officials. Respondents in most other questions had a neutral to high score on the measure of confidence. 11.3.5 Fares
11.3.5.1 Question 9 In a Community Franchise the government owned franchisor must set the level of fares by balancing the need to encourage investment (i.e. keep fares high) with the desire to keep fares at politically acceptable levels and to stimulate demand, (i.e. low). What, in your opinion, would be the likely effect on fares levels be in the application to an urban area of a Community Franchise? Would fares be likely to rise, to fall, or to stay the same? (1 will fall dramatically, to 10 will rise dramatically).
11.3.5.2 Q9 Interpretation This quite high score, low deviation and healthy confidence of response relevance suggest a strong feeling across dimensional groups that fares would rise. The lowest scores, i.e. those who were not so clear that fares would rise, came from the Operations
Page 174 of 259
and Finance dimensions. Those thinking it was most likely to rise came from the Franchising and Planning dimensions. 11.3.6 Alignment of Interests
11.3.6.1 Question 10 In the Community Franchise in the Dolomites, the government owned franchisor must operate profitably and must set fares at levels that encourage investment by private franchisees. The franchisor’s only income is from a percentage of ticket sales. The only income of franchisees is being paid for each passenger/km. The only option for passengers is to switch routes or mode to avoid a franchisee operator that is not satisfactory. To what extent do you think that the interests of the passenger, the franchisor regulator and the franchisee provider, coincide?
(1 closely - 10 not at all)
11.3.6.2 Q10 Interpretation There was only a single respondent who thought that the regime would have an inappropriate alignment of interests between parties; all respondents, accept this one Government member, were either neutral or positive with a positive average of 4.5. The most positive responses were from the Franchising and Operations groups and the negative outlier was from a state parliamentary situation in the Government dimension. There was a slightly greater variation in confidence which might suggest that respondents were not as sure of this question as compared to others. This is countered by the relatively low standard deviation.
11.3.6.3 Q11 Regulatory Capture In a Community Franchise there might be a temptation for the (government-controlled) franchisor company to favour particular franchisees for some future reward that is not in the public interest. For example, if the franchisee regulator was to have relationships with franchisee operators, there may be pressure or inducements to change rates to favour a particular franchisee at the expense of say other operating franchisees of different modes. (Note that all operators receive the same payment/passenger/km for any particular mode). To what extent do you think that the interests of the franchisor
Page 175 of 259
and the operating franchisees may overlap and create conditions for regulatory capture?
(1 not at all - 10 a great danger)
11.3.6.4 Q11 Interpretation The mean for Q11 on whether regulatory capture was a potential problem, suggests that overall the view is that there is not necessarily a danger of capture in the regime structure, but the great range of opinion suggests that it may be a controversial issue. The grouping of dimensional responses shows the Franchising and Government dimensions being more sceptical than all the other dimensions, but this was not reflected in the written responses which, generally, were more nuanced. There was the observation from the Government dimension that it is an issue in the current arrangements also and from the Planning dimension that it is a factor that can be minimised by proper regime design.
11.3.6.5 Question 12
To what extent is this different from existing government enterprise arrangements
11.3.6.6 Q12 Interpretation This question featured a very large deviation so that there was a great diversity of opinion, but the mean was slightly positive. The responses to Q12 suggests that regulatory capture is a current issue in existing arrangements so is not a new or unique feature of this regime. Getting these regime design features correct was noted to be a “core requirement”. There was a very solid level of confidence from most respondents, with only one outlier (in Franchising), but interestingly none giving a 10, or even a 9, as occurred in several other questions.
Page 176 of 259
11.3.7 Overall Feasibility and Critical Features
11.3.7.1 Question 13 What is your assessment as to this Community Franchise business regime being able to function in an urban public transport setting?
(1 very well - 10 not at all)
11.3.7.2 Q13 Interpretation The average reading of 5.2 for its overall feasibility suggests expert opinion was that on balance a Community Franchise was seen that it may work in an urban setting. Operational respondents were split with a 2 and a 9, but Franchising and Finance were generally comfortable with the concept. However, nearly half of all responses, were only mildly convinced, suggesting a difficulty in making a call on this complex question. Those on the positive side appeared to be more varied and less convinced of their position than those on the negative side with three solid responses at high levels of disagreement. There were several suggestions from government and academic participants in their comments that trials out of Sydney were the appropriate way of assessment of this question, because of uncertainty and a need to ‘iron out bugs.’ Several, from the Operator and Financial dimensions, saw no problems in implementation but no one from the Government dimension saw that it would not work. All but three respondents were of a positive opinion, and those on the negative side also had others of the same dimension who were positive about the concept. Those in the Finance and Government dimensions were all positive That there was a moderately large deviation to these scores with many apparently ‘sitting on the fence’, suggesting a difficulty in predicting the outcomes of such complicated questions without any hard evidence for the respondents to consider.
11.3.7.3 Question 14 What features of this business regime, if any, would make a Community Franchise unsuitable for its application to urban public transport? There were no numerical scores requested in question Q14, but the average for the level of confidence was 6.5, with a standard deviation of 1.5 which is quite small compared to other questions.
Page 177 of 259
11.3.7.4 Q14 Interpretation There were several areas of concern identified. The principal ones, generally in the words of the participants, were:
• the disruption of current departmental reorganizations, • the continuation of the support for special groups like the elderly, school children, etc. so that these groups were not full fare paying, • the challenge of correctly pricing fares, • infrastructure funding by commercial entities as opposed to government, • the cost of capital to private firms as compared to government borrowing • the risk to be assumed by private companies in areas of commerce with the uncertainty of large-scale infrastructure, e.g. tunnelling, underground stations. • failure of franchisees with the resulting suspension of service to the areas it served • concern for newer areas of the city that were not sufficiently populated to financially support public transport services, including uneconomic routes that would not be serviced • the lack of cross-subsidization of routes as occurs in area systems now • the possible privatization of the system These concerns will be discussed in Section 11.6 to identify the issues that would affect any application of the regime to an urban area. 11.4 Dimensional Group Figures All figures for all questions were analysed and are shown in Appendix E. These provide the averages and standard deviations for each question and then for each dimensional group. Overall these show that there is less deviation between dimensional groups, being in the 0.8 to 2.2 range than there is for the whole sample of individuals, but the importance of this is not asserted. These are small samples with which to make these measurements and accordingly are not purported to be indicative of group characteristics. However, they are presented as they are available. This indicates that there was some consistency within dimensions, more than across dimensions, again with reservations about the small numbers involved. The greatest divergence of opinion between dimensional groupings as indicated by the numerical scores is shown in Q1 Ownership and Q11 Regulatory Capture. In these questions, those with franchising expertise and those with financial business expertise had least to agree about. The least divergence between dimensional groups by numerical scores is shown in Q2 Politicians as Board Members, Q10, Alignments of Interest and Q13 Overall Feasibility, where there was more consistency between all varieties of expertise. The financial/accounting and operational experts appeared to be less concerned with Ownership than the Franchising and Government groups, whereas, those in the Planning dimension showed less concern over Regulatory Capture than those in Government and Franchising.
Page 178 of 259
11.5 Feedback on Results from Participants The results as reported in section 11.4 were sent to all respondents for their information and to provide them with the opportunity to comment on the results. A period of two weeks was given for any response and one written and one verbal contribution was made. It is common for there to be a falloff in participation from the original participants in most surveys of this kind; this was no exception. A participant from the Finance dimension was surprised at the lack of enthusiasm for the concept of a Community Franchise. One from the Government dimensions was not surprised by this apparent scepticism but, if it were to be tried, wanted to emphasise that politicians would not be suitable board members, (bringing their politics to influence commercial decisions). He also commented on the wide range of opinion provided by the group and how the opinions would be swayed by their personal experiences in whatever line of work they had or were pursuing. In summary, there was little in the way of significant viewpoints to be considered over and above those provided in the original survey. Their comments are included in Appendix F 11.6 Discussion of Major Respondent Issues The principal concerns of respondents can be grouped into five major areas. This section highlights those issues which have been identified that may be of importance if a government were to consider the implementation of a Community Franchise to an urban situation and to identify those critical to its implementation. 11.6.1 Ownership and Direction Significant caution was expressed about private ownership of an urban public transport service, and support for the notion that government ownership is important for this control. This caution was expressed multiple times by those from the public sector and academics. Some of the concern was for if and/or when privatisation of the services would be contemplated by a government sometime in the future. This potential to sell off the businesses and thus loose government control appeared to be the issue of concern for these cautious views. 11.6.2 Budgets and Corporate Control A written comment showed that there was an appreciation of the situation that franchisees would feel that the government owner of the franchisor was able to withhold payments for breaching of franchise standards. This was seen by most to confer sufficient power to the government to control franchisee behaviour. There was no substantive objection to this, but an emphasis on the need to ensure appropriate regulations and organisational structures in the franchisor. There was a comment on the vulnerability of the franchisee in the situation of complete dependence on the franchisor for cash flow, with some speculation as to what might happen if the service was privatised. It is noted by some that the franchisor has almost complete control over pricing and choice of the franchisee, but only indirect control over
Page 179 of 259
the service provision. It highlights a logical consequence of this situation that there is a complete mutual dependency of franchisee and franchisor and that the arrangement can only survive if there is a high level of trust. 11.6.3 Fares Overall, the experts declared that fares would at best stay neutral or rise. The profit motive of both franchisor and franchisees was seen to tend to keep prices up rather than down from current levels. There was also recognition that fares will be determined in part on the capacity of the network to attract new patronage. There was a concern from an academic that the link between costing and pricing would ‘become loose’. If so, it correctly interprets the flexibility inherent in the ways that the franchisor can manipulate incentives to both passenger and service provider. It also identifies where a Mohring effect can materialise if desired. Others from the Finance and with commercial backgrounds pointed out the need for franchisees to generate a sufficient return on the capital invested, which for the reasons noted above, is not totally dependent on fares as these can be subsidised to ensure that a well-run franchisee will have sufficient return to meet this concern. 11.6.4 Network Planning and Operations There was the recognition that the profit motive would ensure that some routes would be favoured over others, but little discussion as to the likely location over the urban area of this demand. Some comments indicated an assumption that commercial decisions would favour more affluent areas. However, there were few examples of comments connecting the issues of subsidies and demand and thus commercial intent. One from the Operations dimension suggested the need for the specification of minimum service levels to ensure a certain provision that passengers could rely upon. This is, as has been noted in the Literature Review, Section 4.6, a built-in feature of franchising, where the service levels of time and frequency are specified, and the franchisor has the opportunity and the responsibility to adjust franchisee payment rates to influence the supply of service capacity. Several participants from the Government dimension asks why a franchisee would propose a route that would not deliver a profit or would be reluctant to service unprofitable routes with social importance. Others from the Planning and Finance dimensions assumed that the profit motive would be motivating franchisees, but that this would not guarantee network coverage to all areas. These comments all point to the need to fully address the issue of subsidies or special arrangements for these areas of the city. 11.6.5 Treatment of Developing Areas of low demand Several participants from the Government dimension spoke of the difficulties of establishing public transport services in newly established areas where demand would not justify the extent of servicing in older areas. In the current system, services, if provided in new areas, may run with less fare recovery than in more established
Page 180 of 259
suburbs. This is an issue of great importance and again, points to the need for government consideration and possibly a special rate for areas defined as ‘new’ or ‘underprivileged’. 11.6.6 Alignments of Interest and Regulatory Capture The very wide responses to these questions suggest difficulty in predicting outcomes to these concerns. Parallels were drawn by government managers with current arrangements in subcontracting and the ultimate regulation of the Corporations Act. This lack of certainty suggests the need for any application of the regime to be closely watched in its initial form, possibly though the suggested pilot scheme in a regional city. Such concerns could be evaluated at that stage, and existing regulatory mechanisms such as the Independent Commission against Corruption, (ICAC) would have a role in this evaluation and adjustment of controls to suit a broader implementation. Importantly, there were several from government and academia who saw this as a danger present in the current arrangements and was accordingly no different. There were no comments that indicated that a Community Franchise would be more vulnerable to these effects compared to existing arrangements. 11.6.7 Pilot program It was evident to several participants from the Government and Franchising dimensions hat a pilot program or trial area would be a necessary preliminary action before the implementation of a Sydney wide application of a Community Franchise. The Hunter or Illawarra regions were suggested as potential candidate regions, because of their isolation from Sydney and presumably their size. It is noted that several isolated cities, for example, Wagga Wagga or Coffs Harbour or other larger regional cities, could meet that criteria of isolation from Sydney and size to pre-test any implementation, although density may limit the applicability of those. A pilot program would also address a concern from the Government dimension about the functioning of the wholesale electricity market, which has some similarities to the Community Franchise regime. The concern expressed was about the lack of responsiveness to high market rates for power that have not had the anticipated attraction of suppliers to the industry. A trial would provide some indication of the responsiveness of franchisee transport operators and establish if the relatively low entry costs of a transport provider, compared to power generation, could be relied upon to provide the necessary services. It would test the idea that the franchisor in a Community Franchise would be prohibited from undertaking service provision itself and would have to rely on the franchisees to do so. A trial could enable the costs of attraction to be realistically estimated to see if the current budgetary settings of TfNSW would be sufficient to attract the necessary number of franchisee suppliers to a major market like Sydney. 11.6.8 Asset Value Realisation It was noted by some experts that the individual lines and the infrastructure associated with them have a value that can be realised when sold by the government to franchisees
Page 181 of 259
at the commencement of regime operations. For a government interested in privatisation of public assets, a Community Franchise was identified as providing a mechanism for the realisation of value. 11.7 Summary The expert survey was used to gather varied opinion from experienced representatives of the five identified dimensions of the topic, as to the concerns that a government would have if the Community Franchise business regime was to be applied to urban public transport in Sydney. Thirteen out of sixteen potential respondents provided their opinions on fourteen questions covering all identified dimensions of research interest. Two participants provided further comment on the completed survey. The survey responses showed a high degree of variability. As well, the range of responses to the quantitative values where scaled values were requested, was notable: there was very little consensus on most of the subject topics.
• There was the least agreement on matters of ownership, board direction and regulatory capture. • There was most agreement on matters of fare levels, network planning and the alignment of interests in a Community Franchise. • Those from the dimensions of Finance and Operations tended to have more cohesive views, and those in Franchising, Government and Planning tended to have greater ranges of opinion. There was broad acceptance of an evolving network from the bottom up commercial pressure of franchisees responding to economic forces. Whilst the experience of Dolomiti SuperSki would support that this is a viable way to allow a commercially responsive network to develop, it appears to be a contentious question and some of those with direct experience in transport planning in Australia, and in franchising indicated that they doubt it would be effective. This division of opinion calls for additional observations and research. This indicates a large range of potential outcomes that experts can foresee in any application of the regime. Were the application of a Community Franchise regime to an urban setting ever to be contemplated, there is the identified need to address certain questions. These include:
• fare levels and subsidies with the possible use of the mechanisms within the regime to allow fares to be set independently of the rates on offer to franchisees; • the need to ensure unprofitable areas were serviced, an obvious focus of the subsidy potential inherent in the loose linkage of the fares to franchisee payments and the capacity to have different ratings of areas for franchisee payments. • the need in regime design to allow the necessary government control to be exercised. This expressly applies to government ownership and control of the franchisor.
Page 182 of 259
• The need to establish if the market for transport services could be sufficiently developed to provide the depth and quantity of franchisees to undertake the services. • the need to ensure that there were appropriate controls to avoid exploitation of franchisee operators and • that controls should ensure that developing and lower socioeconomic areas of the city should be well serviced. In the next and final chapter, the results from all aspects of this study will be reviewed, the extent to which research questions are answered and hypotheses met will be summarised and overall conclusions will be drawn.
Page 183 of 259
Page 184 of 259
CHAPTER 12 CONCLUSIONS 12.1 Chapter Contents This chapter will review the process by which this study has been conducted and then evaluate the extent to which the research objectives defined in Chapter 6, have been achieved. It will then provide a summary of the main findings, before making recommendations for future research. The chapter concludes with a review of the contributions that this research has made to the understanding of business models of public transport. 12.2 Review of Study Process This thesis was undertaken to understand and document a public transport business regime, known to its operators in the ski fields of the Dolomites in Italy, as a Community Franchise. A case study established the workings of this regime and this was examined, documented and compared with other more conventional regimes of ski lifting businesses. An algorithm was used to define the regime and this definition was applied to urban public transport to document a tangible idea for review. Issues identified from this hypothetical application were scrutinized through a computer experiment and an expert survey to obtain quantitative and qualitative measures of its relevance to urban public transport. 12.3 Answers to Research Questions 12.3.1 First and Second Research Questions: Quantitative Evaluation The computer experiment simulated the condition where a government agency would be attempting to maximise social welfare, and this was contrasted with the situation where a franchisee undertaking the same function would be focussed on maximising its own profitability to the potential detriment of social welfare. This addressed the first research question which read: “What is the impact on social welfare of a group of profit-maximising franchisees compared to a single purpose centralised organisation seeking to maximise this measure?” The Hypothesis no 1 for this question was: The Null Hypothesis is that both regimes will deliver similar amounts of social welfare (SW), i.e.:
H0: Mean GE (SW) = Mean CF (SW)
H1: Mean GE (SW) ≠ Mean CF (SW) The measure of social welfare in both situations was shown to be statistical similar and the null hypothesis could not be rejected. This second research question was brought about by a need to examine a potential problem identified by commentators at the ITLS. This commentary identified a risk that a
Page 185 of 259
lack of cooperation between line owning franchisees as commercial rivals provided the potential for there to be a reduction of social welfare. This research question read: “Are competing line owners able to cooperate so as to enhance network performance?” The expected outcome of this question was that: That in a game context, there will be sufficient cooperation to provide an acceptable allocation of resources compared to that which is achieved through central planning. There were no instances identified in the experimental computer game of selfish behaviour between competing franchisees that jeopardised levels of social welfare. All win/win situations resulted in cooperation between the competing franchisees. All situations where only one of the competing franchisees would benefit from a particular action were able to be concluded by that party alone without detriment to the overall functioning of the network. This was not an unexpected outcome, given the considerable research into behavioural economics and psychology. There was no statistically significant reduction in social welfare identified due to the processes inherent in a Community Franchise, and there was no selfish behaviour between competing franchisees identified that caused this business regime to have reduced social welfare. Accordingly, both these first two research questions in this quantitative examination were resolved with the null hypothesis and the expected outcomes being maintained. 12.3.2 Third Research Question: Expert Evaluation The third research question was addressed by a survey of experts. This read: What aspects of a Community Franchise regime are transferable or could be adapted to urban public transport? The expected outcome of this survey was: The hypothetical application of a Community Franchise regime to urban public transport can be demonstrated to have a reasonable chance of success. The opinions and numeric responses of a wide range of experts representing the differing dimensions of the consideration were notably varied. There was no consensus as to the viability of the proposed regime to apply to urban public transport. However, to balance those experts who did not think the regime was transferable, some experts considered the possibility of the regime being appropriate for use in Sydney. The overall mean of all opinion was that the regime may be adapted to an urban application, and accordingly, the expected outcome of the regime having “….a reasonable chance of success” in its application to Sydney was not refuted. 12.4 Limitations of this Research Some progress towards answering the questions raised in this study have been made. But it is perhaps a realistic assessment that the true applicability of a Community
Page 186 of 259
Franchise business regime to urban public transport is beyond the capacity of a thesis to determine. There are many possibilities if this study could be extended in certain ways. More in-depth interviews in a new case study with Dolomiti SuperSki could assist in broadening the picture gained in this thesis. Future case studies could also include ski fields in the US and Japan as well to gather a wider view from other social traditions. A more extensive computer game could be derived from the one used in this study and played with additional subjects. This could provide more robust data to examine the question of resource allocation methods. An expanded expert survey with more people recruited from a wider body of research, commerce and industry would provide more statistically significant results. The fact remains that Dolomiti SuperSki has successfully operated a public transport business using this business regime for over 45 years. As a ski field, it has quite significant differences in outcomes to its European peers and especially compared to those in overseas markets, in the economical delivery of the product and largest choice through the highest number of lifts services per unit of ski run length. This research has at least raised the awareness of this method of delivering a transport service to academia and possibly the public and has gone some way towards its evaluation. It is evident that newer business regimes of transport provision, such as Uber, have unexpected and significant effects on traditional ways of organising the business of public transport. There is much more research that can be done in the case of a Community Franchise. It would appear to be a subject with more to reveal as a business regime with differences to current practice and with some promise of benefit to consumers, government and providers. 12.5 Further Research Many questions were raised during the progress of this study that have not been able to be addressed. The principal areas are identified as follows. 12.5.1 Field Study of Community Franchise Regimes A necessary step in understanding a Community Franchise regime is to study actual examples in the field to take into account consider all the complex interactions of multiple influences. Hence, it would be useful to have other researchers undertake observations of Dolomiti SuperSki as well as any other examples of a transport system that can be found to be organising the transport of the public under a similar franchise arrangement, e.g. by Uber and the other ride-sharing companies. This extends to the need to study any examples of franchise regimes in other industries owned by government if these can be identified, as this is the “Community” in the Community Franchise regime title which fundamentally changes the dynamics of interested parties and the outcomes of operations because of the particular alignments of interests. At this stage, it seems that fact-finding in the manner of case studies is still a priority as a single example of a phenomenon is not a large sample to go on. There is thus the need to undertake studies that can observe and record the actual workings of examples of
Page 187 of 259
Community Franchise business regimes in real life. Cooperatives and similar community based economic enterprises may be fertile areas of research. 12.5.2 Laboratory Experiments A laboratory experiment is of value assisting the observations of the real world and providing a more detailed understanding of observed complex behaviour. There are several ways of extending the computer game to extract far more than was possible with this necessarily limited experimental setup. It would be of great interest to have this game repeated by others and more formal and strict interaction of the experimenter and the participants instituted through use of computer laboratory setups or other devices so that there is less possibility of experimenter influence over participant behaviour. It would provide great insights if the Community Franchise regime game could be played with more than just the three participants who were involved in the running of this experiment. Having more participants would provide much more information on the efficiency and pattern emergence effects between the two regimes. Do four participants give more or less social welfare than three? Do five, six, seven or eight participants provide an increase that is more than that recorded with the three participant games played in this experiment? Or do they provide a decrease in social welfare? The Excel spreadsheet is set up so that this is an easily achievable reconfiguration, without need to adjust any of the parameters or the MNL model determining passenger allocations to lines. The game in its initial prototype form was significantly more complex than the one used, with additional variables for vehicle type, loading arrangements, potential to introduce penalties for changes of mind, amongst other complications. The game had to be simplified to the point that it was able to be ‘solved’ by ordinary people and transport professionals in a reasonable time period, i.e. about 20 minutes and the complicating variables were ‘switched off’ in the program used for the experiment. This meant the experiment lost some of its richness and the capacity to more completely test the evolutionary credentials of the Community Franchise regime. Further experiments could use the switched off features of this game to reintroduce this greater level of complexity. Variations to the game can be imagined that would test the influence of initial conditions, an aspect that could address issues of fairness and inequality, as well as network development. Other variations to this game could include using passenger-km rather than passenger boardings. This would simulate conditions in say Sydney, where the distance travelled is factored into the Opal card payments compared to say Stockholm where ticketing is for a defined period and one is free to travel any distance in that time. In short, there is much that this simple model of an artificial city can be used to further explore this problem that results from a lack of field situations to observe. 12.5.3 Testing a Real-Life Application of a Community Franchise Regime There are testable conditions that can be identified if a Community Franchise regime was ever to be applied to urban public transport, something suggested by several experts. Some of the questions that could be tested include:
Page 188 of 259
• Would local government be the level at which city-wide public transport regimes would be owned and operated? • Would the layouts and modal choices of public transport networks be appropriate for actual passenger demand? • Would the network typology remain as a branching tree focused on the CBD? • Would such a regime cater adequately for the lesser socio-economic areas of the city? • Would infrastructure respond to the economic status and standards of different areas of the city? • Would franchisees enter the property market to generate demand? (as happens in Japan and Hong Kong) • Would fare levels rise or fall? • Could the franchisor adequately satisfy the obligations of the government towards the socially disadvantaged? • Would car usage decline as a proportion of overall trip numbers? • Would there be an improvement in the profitability of the public transport sector? These are all testable propositions that could be used to directly compare the Community Franchise regime with other ways of providing urban public transport. 12.6 Contribution to Knowledge This study has sought to document and explore the business regime suspected to have application in fields beyond its current employment in the vertical transport in the ski fields of the Italian Dolomites. There are several areas in which this study has made some concrete contribution. These include: 12.6.1 Documentation of the Community Franchise regime Although there are a considerable number of variations to the principal business models of urban public transport operating in the world, the Community Franchise regime as it operates in Italy today has not been previously documented to occur. The operations of this regime clearly have quite different implications for the alignments of interest of the various component groupings of involved parties and the incentives to its participants that operate in that environment compared to other regimes. These differences have now been reported and published; a definition of its workings recorded as an algorithm. 12.6.2 Definition of Design A definition of design that considers the three essential elements of the process, namely initiation of variation, the evaluation of alternatives and the selection between these alternatives, was proposed. This definition is consistent with current evolutionary theory, with many areas of human endeavour and in this thesis has enabled a relevant comparison to be made between the normal centralised network design of most public transport systems and the evolving undirected process observed in the Dolomites.
Page 189 of 259
12.6.3 Effects of Regime on Social Welfare The experimental game provided a way of testing the generation of social welfare. It explored the proposition, that was not disproven, that bottom-up methods of resource allocation can be as effective as centralised planning in the generation of social welfare. 12.6.4 Applicability of Regime to Sydney The expert survey provided a fresh view of the issues and showed how experts are divided in their opinions, giving a limited, but overall slightly positive support for the feasibility of an application of a Community Franchise to the public transport to Sydney. Some individual experts saw that there are distinct possibilities were it to be tried and this makes the concept worthy of serious investigation. Others saw considerable cautionary elements providing a valuable framework for practical evaluation. 12.7 Final Word The study has identified differences and some potential benefits that a Community Franchise regime might have over conventional regimes of public transport, including local ownership, independent entrepreneurial activity and total control by the government-owned franchisor over levels of pricing, service timing, frequency and safety along with other relevant standards. The observations of the Dolomite operations suggest that in an application of the regime to an urban setting, franchisee initiative may encourage an increase in the number of lines with subsequent increased connectivity with resultant network effects. As well there is the possibility of capital returns to government. This return of capital is achieved with the retention by government of responsibility and power over the commercial processes that control fares, operating times and service frequencies. Such possibilities must be at least of interest to those wishing to improve on the relatively minor share that public transport currently has in the movement of people around Sydney and to those wanting to influence the significant cost it has to the NSW state budget.
Page 190 of 259
Page 191 of 259
Page 192 of 259
APPENDIX A – CASE STUDY ETHICS APPROVALS
continued below….
Figure 0-1: Image of Completed Ethics Approval https://irma.sydney.edu.au/production/content/rqf
Page 193 of 259
Page 194 of 259
APPENDIX B EXPERT SURVEY ETHICS APPROVAL
continued below…..
Page 195 of 259
Page 196 of 259
APPENDIX C EXPERT SURVEY AS SENT TO PARTICIPANTS
Page 197 of 259
Page 198 of 259 Consideration of the Application of a Community Franchise Business Regime to Urban Public Transport
This survey investigates the application to a city environment of an atypical business model of public transport. You are asked to provide your personal view of this business model by responding to the questions that follow this brief explanation of the regime and the way that it would apply to an urban setting. This hypothetical construct was developed with the city of Sydney as the example but could apply to any urban area in the developed world.
The business regime, from which this hypothetical application for a modern city is derived is that used by Dolmiti SuperSki to operate its network of fixed line ski lifting in the Dolomites in Italy. The basic arrangement is a franchise business model, with a government owned and controlled commercial franchisor and independent, separate commercial franchisees of the individual lines of transport. Franchisees have property rights over lines in their possession with the exclusive right to provide the transport function on their lines. However, franchisees must maintain operational standards of times and frequency of operations as set by the franchisor. The operators of this largest of all ski fields in the world, described the business regime as a Community Franchise.
A fundamental difference from the public transport business regime operating in most urban areas of the first world is that the resulting network is not planned by a government entity. In the traditional government enterprise, the government owner plans and operates all aspects of the service and only contacts out to private operators the right to run portions of the overall network which it judges to be economically beneficial. Contracts for these operational services are typically only for a limited time period, often five to ten years.
In the Community Franchise regime, the franchisee operators of the component parts of the system own their lines and do not have this time limited contract restriction. The franchisee plans all aspects of their lines of transport, have property rights over them and are free to buy and sell these assets. The government franchisor owns the brand, sets all standards, markets the service, sells all tickets and collects all fares. The franchisor considers franchisee proposals and has the capacity to include or exclude those seeking to participate in the system. They only pay franchisees which operate to the standards which the franchisor sets.
The franchisee initiates a proposed line, making all decisions as to mode and route, and provides all capital and operational expertise when its proposal for a particular line is accepted. The individual franchisees are paid by the franchisor an agreed fee for the carriage of each passenger on a per distance basis. Payment is standardised for each mode and applies equally to all franchisees. This fee is paid only for the number of passengers carried, not for the provision of the service.
In the Community Franchise model, a popular line will attract more passengers and the owner of that line will have a greater income than a line that has fewer passengers. This provides a clear incentive for franchisees to choose routes that will be in demand and then to provide the most attractive service on that line. Any owner has the opportunity to improve their line by introducing better infrastructure or by increasing the number of services over the minimum to cater for greater passenger demand. The franchisee benefits by any increased income generated by their initiative. The economies and improvements incorporated into any particular line operation that result in reduced costs or increased passenger numbers will only benefit that franchisee. This is true both of invested capital and of running costs thus providing a motivation for franchisees to establish the most economical arrangements for their lines. In the Dolomites, considerable competition exists between lines, both in the original conception of the line’s layout, mode, route, etc. but also in the attractiveness of the line and the efficiency with which it is run. It is expected that this competition would occur in an application of a Community Franchise business regime to urban public transport. A summary of this regime is as follows:
Regime Background Urban Application
The Community Franchise business regime of ski lifting in The Community Franchise business regime of public transport the Dolomites in Italy successfully operating since 1974. applied to an urban area.
• A traditional commercial • Owned by the local governments of franchise business regime the city.(In Sydney as a Regional • Fully owned/controlled by the Organisation of Councils (ROC)) communities of the region • A traditional commercial franchise • The franchisor: Dolomiti business regime SuperSki, markets the skiing and the Dolomites, • The franchisor: markets the services, sells smart card sells electronic smart card tickets to patrons. tickets electronically to passenger, pays franchisees • The franchisees: Commercial lift owners; take as on a per passenger/distance basis, considers many skiers as possible on the lifts that they own. proposals for new lines/routes from franchisees. • Only paid by franchisor on passenger numbers. • The franchisees own one or more individual lines of • Financing and constructing of the transport transport over which they have exclusive property and infrastructure by the private franchisee operators operational rights on the routes those lines take. The Franchisor The Small/Medium Franchisee
What is this government organisation and what does it Who carries out the basic public transport function? do?” • Lines/routes of transport exclusively • A commercial Proprietary owned by individuals and companies. Limited company (like Australia Post • Capable of financing and running fully owned by the Australian buses and ferry lines. Government) • Will propose lines/routes of transport • Controlled by a board of that they consider financially viable. director answerable to the elected Councillors. • When their proposed lines are accepted by the • Does not undertake transport function directly but franchisor they plan, finance, construct any acts as the franchisor. infrastructure and then operate the buses, ferries or • Engages franchisees who propose and carry out whatever mode of transport that they decide they will the transport function on various lines of public use on their particular line. transport. • Franchisees optimise their operations to attract • It is the regulator of standards for safety, passenger and minimise operating costs. Additional frequency and hours of operation. Only pays passenger numbers increase their income. Any compliant franchisees. savings to operating costs are retained by the franchisee. The Large Franchisee Fares
What kind of organisation could provide the large scale How are fares set? infrastructure and operate a dedicated line of transport? • The Franchisor sets the fare levels • Large, publicly listed and is overseen by the government owner, companies with expertise in who is responsive to the electorate infrastructure. • Fare levels will be set as low as • Capable of assessing complex possible to encourage patronage and give the city investments and planning for their cheap public transport. However…. implementation. • The franchisor seeks to attract franchisee interest in • An organisation able to raise large amounts of investing in new lines and will maximise the fares by capital this measure as this is the principle payment made to • An organisation able to operate complex franchisees. infrastructure and comply with the standards and rules of the franchisor. Cash Flow Network Planning
How is financial incentive created? What is the mechanism of network design?
• Fares collected electronically • Passenger congestion signals an by Franchisor (like Opal, Oyster Card) opportunity for a franchisee to profit from • Electronic recording of the satisfaction of obvious demand. passenger activity on each separate • Proposals for new and improved vehicle and line. lines of transport are made by individual franchisee owners and entrepreneurs. • Direct payment by Franchisor to each separate • Passenger use of lines provides the selection Franchisee for passenger/distance on their line. mechanism to determine which new and existing lines • No other payment to Franchisees survive and/or thrive • Payment/passenger is at a level to cover costs of • Evolution of network by bottom up entrepreneurial capital employed as well as running cost. initiative rather than top down central planning.
Page 200 of 259
Do you have any Queries on the functioning and structure of the Community Franchise? Should you have any questions about the structure of the regime or about any other aspect of this hypothetical application to an urban area, please send your query to: [email protected] The researcher on this project will answer the question and circulate the question and answer to all participants without revealing identities.
Questions for Expert Panel Consideration Your involvement in this survey will test the assumptions used in creating the hypothetical application of a Community Franchise regime, illustrated on the preceding two pages. Please consider each of the questions and provide a response in the text box for each question. Where a scale response is requested, please shift the marker to the chosen point. Pleases open and complete the survey within Word, using the scaled response and the expanding text boxes to accommodate your responses. When completed please return the survey as a PDF document using the Save and Send option at the end of the survey, again within Word. You may keep the Summary sheet (page 2) open when you address each question.
Personal Expertise Assessment You are being asked to answer questions about aspects of public transport business regimes that may not be in your personal area of expertise. To assist in assessing the implications of different areas of expertise, you are asked to provide a measure of the confidence you have in your own responses. This will be made at the end of each section of questions with a sliding scale to provide your assessment based on having minimal confidence at 1 to significant confidence at 10.
Times for Responses This should not take you more than one hour to complete.
Questions one to fourteen follow over the page.
Page 201 of 259
Ownership and Direction
Privatisation and responsibility for network planning can be contentious topics. There has been significant debate about the privatisation of public transport services and other traditional publicly owned services like power generation, lands title information systems, etc. The distinguishing feature of this application of a Community Franchise regime is ownership of the franchisor. A Community Franchise is a commercial entity completely owned, and with a board controlled, by Government. This company would set levels of and collect all fares, distribute collected monies to franchisee operators, and generally operate in accordance with the objectives set by the company board. and receipt of all fare income by government, but with private initiation, ownership and funding of public transport operations by private franchisees. Please indicate your response to the following questions.
1 You may be aware that the Australian Government is the sole owner of the corporate entities Australia Post and the Australian Rail Railtrack Corporation.
To what extent do you see problems associated with the ownership of a commercial franchise public transport company that would set rates for passengers, collect all fares, distribute collected monies to franchisee operators, and generally operate in accordance with the objectives set by the company board? (1 none - 10 many)
1 2 3 4 5 6 7 8 9 10
Please provide reasons in the text box below.
2 To what extent would a politician, or his nominee, be an appropriate board member of a Government owned commercial franchise company?
(1 good - 10 bad)
1 2 3 4 5 6 7 8 9 10
Please provide reasons in the text box below.
3 What other types of individuals would be more suitable than, or should supplement politicians to represent the government on the board of a Government owned
commercial franchise company? Please use text box below
Page 202 of 259
Please provide the level of confidence you have in your responses to this section (1 low – 10 high).
1 2 3 4 5 6 7 8 9 10
Budget
4 In a Community Franchise, the government owned commercial entity acting as the franchisor must operate profitably, setting fare levels, the amounts to pay franchisees and receiving its own
remuneration as a percentage of gross turnover. To what extent would you expect a Community Franchise to be able to control public transport spending compared to current arrangements
(1 very well – 10 very poorly)
1 2 3 4 5 6 7 8 9 10
Please provide reasons in the text box below.
5 Control by the franchisor over the franchisee operators’ operations is a feature of franchise regimes operations. Fares are collected by the franchisor. Payment by the franchisor to franchisee operators would depend on the compliant operations of each franchisee. To what extent do you feel that a government owned franchisor in an urban application of a Community Franchise regime has the capacity to look after the interests of the disadvantaged in the community and the electorate generally? (1 very well – 10 not at all)
1 2 3 4 5 6 7 8 9 10
Please provide reasons in the text box below.
Page 203 of 259
Please provide the level of confidence you have in your responses to the budgetary section (1 low – 10 high).
1 2 3 4 5 6 7 8 9 10
Network Planning and Operations
6 All network-planning in the Dolomites is undertaken by franchisee operators who research, plan and then propose the lines on which the franchisee feels has a sustainable profitable future. If
applied to an urban area, decisions as to the routes of new lines and alterations to routes and modes of existing lines will be made by the private franchisee operators responding to their perceptions of passenger demand. To what extent if a Community Franchise was applied to a city do you think that the network planned in this manner would operate efficiently and meet passenger demand? (1 very well – 10 not at all)
1 2 3 4 5 6 7 8 9 10
Please provide reasons in the text box below.
7 All urban infrastructure projects (from the largest like a metro line to small lines of just a few buses without even a terminal) would become the responsibility of privately-owned franchisee companies
to plan, finance as well as operate. To what extent do you think that privately owned companies would be capable of planning, funding, constructing infrastructure and then operating their lines of urban public transport? (1 very well – 10 not at all)
1 2 3 4 5 6 7 8 9 10
Page 204 of 259
Please provide reasons in the text box below.
Please provide the level of confidence you have in your responses to this Network Planning and Operations section (1 low – 10 high).
1 2 3 4 5 6 7 8 9 10
Control and Maintenance of Standards
8 In a Community Franchise, the only control over privately owned franchisee operator behaviour is by the government owned franchisor setting the rules of operation for franchisee operators,
policing them as the franchisor and controlling the cashflow to the franchisees. To what extent do you think that the structure of a Community Franchise would enable the government owned franchisor to exercise proper control over franchisee operator standards and behaviour?
(1 not at all – 10 very likely)
1 2 3 4 5 6 7 8 9 10
Please provide reasons in the text box below.
Please provide the level of confidence you have in your responses to this Control and Maintenance of Standards section (1 minimal – 10 significant).
1 2 3 4 5 6 7 8 9 10
Page 205 of 259
Fares
9 In a Community Franchise the government owned franchisor must set the level of fares by balancing the need to encourage investment (i.e. keep fares high) with the desire to keep fares at
politically acceptable levels and to stimulate demand, (i.e. low). What, in your opinion, would be the likely effect on fares levels be in the application to an urban area of a Community Franchise? Would fares be likely to rise, to fall, or to stay the same? (1 will fall dramatically, to 10 will rise dramatically).
1 2 3 4 5 6 7 8 9 10
Please provide reasons in the text box below.
Please provide the level of confidence you have in your responses to this section on Fares (1 low – 10 high).
1 2 3 4 5 6 7 8 9 10
Alignments of Interests
10 In the Community Franchise in the Dolomites, the government owned franchisor must operate profitably and must set fares at levels that encourage investment by private franchisees. The
franchisor’s only income is from a percentage of ticket sales. The only income of franchisees is being paid for each passenger/km. The only option for passengers is to switch routes or mode to avoid a franchisee operator that is not satisfactory. To what extent do you think that the interests of the passenger, the franchisor regulator and the franchisee provider, coincide?
(1 closely - 10 not at all)
1 2 3 4 5 6 7 8 9 10
Page 206 of 259
Please provide reasons in the text box below.
Please provide the level of confidence you have in your responses to this Alignment of Interests section (1 low – 10 high).
1 2 3 4 5 6 7 8 9 10
Regulatory Capture
11 In a Community Franchise there might be a temptation for the (government-controlled) franchisor company to favour particular franchisees for some future reward that is not in the public interest.
For example, if the franchisee regulator was to have relationships with franchisee operators, there may be pressure or inducements to change rates to favour a particular franchisee at the expense of say other operating franchisees of different modes. (Note that all operators receive the same payment/passenger/km for any particular mode). To what extent do you think that the interests of the franchisor and the operating franchisees may overlap and create conditions for regulatory capture?
(1 not at all - 10 a great danger)
1 2 3 4 5 6 7 8 9 10
12 To what extent is this different from existing government enterprise arrangements
1 2 3 4 5 6 7 8 9 10
Page 207 of 259
Please provide reasons in the text box below.
Please provide the level of confidence you have in your responses to this Regulatory Capture section (1 low – 10 high).
1 2 3 4 5 6 7 8 9 10
Overall Feasibility
13 What is your assessment as to this Community Franchise business regime being able to function in an urban public transport setting?
(1 very well - 10 not at all)
1 2 3 4 5 6 7 8 9 10
Please provide reasons in the text box below.
Critical Features
14 What features of this business regime, if any, would make a Community Franchise unsuitable for its application to urban public transport?
Please provide features and reasons in the text box below.
Please provide the level of confidence you have in your responses to this Overall Feasibility and Critical Features section (1 low – 10 high).
1 2 3 4 5 6 7 8 9 10
Page 208 of 259
Overall Personal Expertise Assessment
Your background and expertise are important to the analysis of this survey. Can you please move the indicator to best describes your background?
Government Finance Transport Business Franchisor Academic Other Planning Operations Research Administration
If you fit into more than one category, or if you describe yourself as “Other” please provide a brief indication of your professional/business expertise in the text box below? You might advise of your professional and academic background, training or current role, in fact any matter that sheds light on the perspective from which you view this topic.
Further Involvement
These results will be consolidated without revealing the identities of any participant. You will be asked for a final comment on this consolidated response.
Ethics Consent and Submission You consent to being part of this survey by returning this document.
If you are ready to submit your responses, please follow the directions as follows from within this document in the Word program:
• File • Share • Email • Send as PDF • Send to: [email protected]
With thanks for your participation from the ITLS, Business School, University of Sydney
Page 209 of 259
APPENDIX D RESPONSES TO EXPERT SURVEY
Question 1 You may be aware that the Australian Government is the sole owner of the corporate entities Australia Post and the Australian Rail Railtrack Corporation. To what extent do you see problems associated with the ownership of a commercial franchise public transport company that would set rates for passengers, collect all fares, distribute collected monies to franchisee operators, and generally operate in accordance with the objectives set by the company board? (1 none - 10 many)
Participant Response
Jenny Buchan Numerous Australia Post franchisees are in dire straits financially
The model proposed with the franchisor collecting fares then paying franchisees makes franchisees 100% dependant on the franchisor for cashflow. The franchisee becomes a free line of credit for the franchisor. This might be OK if the government is franchisor but if/when the franchisor is privatised the franchisees become very financially vulnerable. This is like the model adopted by Kleenmaid – hopelessly insolvent – and others.
Paul Forward Depends on the complexity of the operating environment regarding issues to be managed. The more complex the environment the more problems to be managed.
Trevor Daley The existing similar franchise arrangements have been reasonably successful. The reservations are limited to pressure on service levels as a result of potential pressure from non-commercial price setting.
David Naylor Far less red tape. More efficient process Peter Warrington As long as objectives align with agreed government policies, this model could work. Roy Bishop The further away from direct control, the further away from direct accountability – as with things like the SCG Trust. Not sure how this would be very different from current situation where Transport for NSW sets rules and contracts operators by tender.
Community transport providers were oases of responsive transport carried out by organisations made of people who want to make sure that people with disability, limited mobility or unable to use mainstream transport could get around. They were based in their local communities and funds went on services not fancy stuff and administration.
Graeme Troy No significant concern. Change always involves risk and unforeseen issues.
Rebecca McPhee My understanding of the description above is that the Franchisor sets rates and collects fares, and may accept new routes proposed by Franchisee. As such I believe that it is essential that this is a Government-controlled entity (as a commercial business would not have the same remit to consider social
outcomes). This is made even more important give that public transport in Sydney (as the case study) is heavily subsidised by the taxpayer
This doesn’t mean that the Franchisor couldn’t have commercial drivers eg an independent Board, but should ultimately be subject to Ministerial direction.
Stephen Rowe The focus of the franchisor does not seem to extend to any input into the planning task. This could result in substantial service differences in different regions / routes depending on the franchisees strategies. Or are there service levels dictated by the franchisor?
Wijnand Veeneman Obviously, the link between costing and pricing will become loose. That has an effect on the development of operation as the pricing scheme will not allow for more intricate pricing schemes that could drive new service types.
Matt Wheatley Maintaining infrastructure and maintenance on vehicles may be of concern. The franchise model is a very complex relationship especially at termination and can be best described as a ‘divorce’ of which ownership of assets, debts and goodwill can be quite messy.
Ian Muncaster The commercial risks are huge with the franchisor setting the fare levels. I don’t think any franchisees would accept this as level of fares together with forecasts of travel numbers are a fundamental to the business model. This would be highest for rail because of significant capital cost and less for bus & ferry. I imagine that franchisee would make a proposal based on a certain fare level & perhaps them taking risk on travel numbers. Fare levels would be subject to increase based on a formula included. It would be an unacceptable risk to have this determined by the franchisor board. Typical on other infrastructure deals eg poles and wires & road tunnels the franchisee makes a one off capital offer. With road tunnels I believe also based on a minimum numbers of cars with some compensation if numbers fell below this.
What model was used for the airport line? Very convenient but a lot of consumer backlash for high cost of travel as compared with the rest of the network!
Huge D & C risks associated with tunnels and underground stations. Who provides funding to the franchisor? Banks , private investors,, superfunds and what guarantees they require based on their assessment of risk.
There is also political risk and an example was the proposed extension of the rail from Bondi Junction to Bondi. The government of the day cancelled this because of protests led by the actor Michael Caton. Their argument was all the potential riff raff coming from the western suburbs portrayed as petty criminals, drug users etc. Tim Rogers Effectively the government controls this now so you are changing the model not establishing a new set of controls.
Page 211 of 259
Question 2 To what extent would a politician, or his nominee, be an appropriate board member of a Government owned commercial franchise company?
Participant Response
Jenny Buchan Entirely depends on his/ her prior expertise, as with all board members.
Paul Forward Board members need to be independent of government but aware of the government’s policy agenda and the regulatory framework that the company operates in.
Trevor Daley As long as equal voting rights – government needs to be aware of issues and it may help in obtaining government support if required.
David Naylor Prefer a person with experience vs. a bureaucrat Peter Warrington Should operate independent of politics (but in accordance with government policy). Roy Bishop If you have politicians running things, it will be done with political intent.
Graeme Troy Some infrastructure industry expertise should be a requirement. Should be an expert nominee who can independently represent/liaise with political masters.
Rebecca McPhee Per above, I see it essential that the Minister has power to direct the Franchisor.
NB should not sit on the board of a Franchisee – as these organisations would profit from government expenditure
Stephen Rowe Politicians or there nominees would be as appropriate as anyone else, as long as they are reasonably intelligent. Obviously the make-up of the board would be reasonably varied and balanced
Wijnand Veeneman The CFC is representing the public side of things. We see (in European systems) that a balance between public requirements and commercial operational reality is required, but this seems a good basis for the securing of the wider public good.
Matt Wheatley The Government should be on the Board
Ian Muncaster Definitely not politicians! 70% are lawyers and or political apparatchiks with little experience of life in the business world. Look at their current performance. 6 years and still not able to get a coherent energy and emissions plan in place Tim Rogers A political appointment always brings an expectation of responsiveness. A nominee may generate less expectations.
Page 212 of 259
Question 3 What other types of individuals would be more suitable than, or should supplement politicians to represent the government on the board of a Government owned commercial franchise company? Please use text box below
Participant Response Jenny Buchan Basically the board should identify skills needed then go looking for people with those skills, not just the old boys network that is a current feature of some Australian public company boards. People with requisite expertise in transport, government, and franchising for a start. Paul Forward Board needs to be diversified with representatives selected from people with a strong commercial, transport, land use planning, economic, customer service, community and environmental background. Board needs to have a gender balance. Trevor Daley Specialists in transport are imperative. Strategic thinking practitioners with knowledge of population trends. Company Director’s with finance backgrounds and large corporation experience. Representatives of the franchise network. David Naylor Someone with experience and background in the industry Peter Warrington Business experts. Community representatives. Senior government officials. Roy Bishop If governments appoint them, won’t they be ‘mates’ or ‘yes’ people? Look at what they’ve done with Local Health District boards… Graeme Troy Nominee- as in question 2 answer Rebecca McPhee No comment
Stephen Rowe Individuals could be government bureaucrats, town planners, local
council staff, transport planners, or even operator representatives.
Wijnand Veeneman Someone with operational experience in public transport, but not related to the franchisees. Someone representing a travellers’ perspective.
Matt Wheatley Commercial Lawyer, Accountant who has worked in franchising, and possibly an ex-director of a master franchise that operated in Australia. Ian Muncaster Only people with proven business experience. Preferably directors of large companies Tim Rogers People with a commercial/economic, transport or planning expertise.
Page 213 of 259
Question 4 In a Community Franchise, the government owned commercial entity acting as the franchisor must operate profitably, setting fare levels, the amounts to pay franchisees and receiving its own remuneration as a percentage of gross turnover. To what extent would you expect a Community Franchise to be able to control public transport spending compared to current arrangements
Participant Response
Jenny Buchan Controllable to the extent that they kept closely in contact with each franchisee and genuinely understood the present, evolving and future needs of the community.
Paul Forward Unlikely that the Franchise would be able to raise sufficient capital from the fare box to fund infrastructure works on the network. The Franchise may even struggle to fund operating cost without a government subsidy. Treasury needs to be involved in expenditure decisions and
budget allocations.
Trevor Daley A top down targeted setting of pricing may provide a focus for franchisees to control costs. There may be opportunity to set coal face remuneration packages that reward efficiency.
David Naylor Based on commercial terms Peter Warrington Unclear how would operate differently (in reality) from, say, a Sydney bus regional operation Roy Bishop By definition public transport relies on government subsidy. Current ideology for government is to lower taxes and farm things out so they are not responsible. Anything the franchisee can do will be limited by budget allocations and political decisions about protecting the backs of ministers and the re-election of governments. Look at Murray/Darling water allocations!
Graeme Troy Public transport spending would be significantly reduced. Do not see the need to ‘must operate profitably’. Targeted subsidies may be appropriate. However, such could be applied to fare reductions/refunds for specific passengers that does not impact payments to franchisees.
Rebecca McPhee Given the current extremely high level of subsidy for public transport routes in Sydney I don’t see it as feasible that the franchisor would operate profitably.
This may be possible in other cities (eg Hong Kong, London) where public transport operations are characterised by:
- Higher density populations and therefore higher average loads
- Higher fares per km
- Greater ancillary revenues eg from property development
Page 214 of 259
A profit motive would result in significantly fewer routes being operated in Sydney.
Moving expenditure decision-making closer to the customer may result in better control of spending, however given the two of the key contributors to patronage growth would still be held by the Franchisor (fares and marketing), the Franchisee wouldn’t have full control.
Stephen Rowe Potentially better control as they Community Franchise would be able to negotiate change rather than relying on a tender regime to implement disruptive change and all the weaknesses in such a model.
Wijnand Veeneman Not sure what is meant with current arrangements. A reflective note, the incentive is to the franchisee to raise their price per pkm or trip. The incentive to the franchisor is to keep that low, for as long as the public is pushing that. The possibilities for objective analysis of real
costs and effective pricing are somewhat limited in this model. So, a tug of war on costs and on the above issue is to be expected. Again, not sure how that relates to “the current arrangements”.
Matt Wheatley No comment
Ian Muncaster Very poorly. There appears to be two parties both wanting to make a profit.
If the Franchisor profit is based on gross turnover and the franchisee keeps all the profit from fare levels what benefit is this to the community
Tim Rogers Because it sets the base service levels, if a service operates the risk is low. It only pays out a proportion of the fare and is able to set the level of that fare. In practice it should be OK although there are theoretical limits.
Question 5 Control by the franchisor over the franchisee operators’ operations is a feature of franchise regimes operations. Fares are collected by the franchisor. Payment by the franchisor to franchisee operators would depend on the compliant operations of each franchisee. To what extent do you feel that a government owned franchisor in an urban application of a Community Franchise regime has the capacity to look after the interests of the disadvantaged in the community and the electorate generally? Participant Response Jenny Buchan Depends on the politicians and the heads of government departments – some would be good and others terrible. Reimbursement for transporting the disadvantaged (unemployed, physically/ mentally disabled) would need to receive a strong $ weighting to incentivise franchisees to make provision for them.
Page 215 of 259
Paul Forward The level of subsidy should be a government decision, with input from the franchisor. Government should be responsible for determining any social service obligations affecting the franchisee operators. Trevor Daley It may be more difficult to become aware of special needs in specific areas and to reflect potential higher costs in setting pricing. After establishment of an individual franchise arrangement it may be difficult to the contract to reflect changed circumstances. David Naylor Need commercial practical experience Peter Warrington As per 4 Roy Bishop They will be focussed on renewing their franchise and ticking boxes so they look good. Graeme Troy Can subsidise fares where and when appropriate without impacting payment so payees. Subsidisation should be for disabled persons and not to ‘aged’ persons who do not require subsidies. Rebecca McPhee Franchisee would be incentivised to cut poorly used local routes – these typically offer infrequent, indirect journeys and poor journey times. As a result they tend to be used by more ‘transport disadvantaged’ but would be the first cut in this model Stephen Rowe Depends on a number of things; the level of funding that the franchisor receives; the willingness of the franchisee to implement new services and actively plan the services they operate; The power of the franchisor to suggest new services; the willingness of the franchisee to cross subsidise services to gain a better long term outcome.
Wijnand Veeneman Obviously, this is subject to the legal ability of the franchisor to set demands to the operations of the franchisee. And the tendency of the franchisor to avoid risks. They should be able to do it, whether they apply this ability with restrained is to be seen.
Matt Wheatley Both this question and the previous question are highly reliant on the profitability of the franchisee(s). Ian Muncaster How well is the inspection by the franchisor of the franchisees operations actually going to be carried out. Compliance or no payment will be a big incentive for the franchisee to comply. The reality is the government has an extremely poor record of monitoring compliance.eg Building regulations, ABRA & ASIC regulations as applied to the banking industry. Irrigators taking water from the Murray without any metering etc Tim Rogers The fixed fee franchise arrangement seems to suit where a base level of operations will provide sufficient income to meet its costs. What happens in developing areas without transport links and full occupancy or areas of depressed income? What powers might the franchisor have to deal with startups in developing areas or dealing with concessions?
Page 216 of 259
Question 6 All network-planning in the Dolomites is undertaken by franchisee operators who research, plan and then propose the lines on which the franchisee feels has a sustainable profitable future. If applied to an urban area, decisions as to the routes of new lines and alterations to routes and modes of existing lines will be made by the private franchisee operators responding to their perceptions of passenger demand. To what extent if a Community Franchise was applied to a city do you think that the network planned in this manner would operate efficiently and meet passenger demand? Participant Response Jenny Buchan Ideally some consistency would be introduced into the system so that where franchisees’ territories intersected there would be consistency of service in time, route and capacity. So, franchisees would need to communicate well with each other. Presumably the franchisor would have a monitoring role here. Paul Forward Franchisee operators will most likely be motivated by maximising profits, however, under this regime some members of the community will not receive an adequate transport service without support from government. Trevor Daley The Dolomite model supports a specific focused demand trigger, to my understanding ski fields. The broader demand triggers in an urban context may be more difficult to anticipate. I assume that there will be a level of reliance on links to the operations of other franchisees and individual planning may be undermined failure to respond by other adjoining operators. David Naylor Appears more commercial Peter Warrington It would create some better routes than current processes, but may miss some community service style routes that are important but not profitable. Roy Bishop Once you have a tram, train or metro line there is not much you can do about moving it. There is a long historical framework on which current day decisions need to be made – including where people live in order to be able to get access to transport. What happens in new growth areas where it is not profitable (yet)? What are places like Waterloo where 75% of the population are on pensions/benefits? Not profitable = not worth servicing. Graeme Troy May be conflicting interests in use of routes. Rebecca McPhee Network would be more efficient but perhaps not as well meet the needs of transport disadvantaged Stephen Rowe Dependant on the skill of the franchisee. A minimum service level strategy may be appropriate, or an approval to operate new services and change current services based on the franchiser knowledge and skills.
Wijnand Veeneman Not well with the smaller franchisees providing bus services, as the incentive to over provide services for competitive purposes with other franchisees is much greater as costs of entry are lower (without the need for extensive infrastructure as in the Dolomites).
Matt Wheatley Franchisees will chase the highest demand for sales. On one side, some areas will be fantastic, on the other, there will be areas (most likely
Page 217 of 259
regional) where franchisees will be less optimistic in investing unless there are subsidies or concessions from the franchisor. Ian Muncaster If the perceptions of passenger demand match future reality then this would operate. The question who is prepared to take this risk. There is significant cost and undertaking this research apart from capital cost to set up. OK only for bus and ferry. Importantly who provides the infrastructure and service for those for the remote and potentially non profitable routes. This then fall back to government. Tim Rogers The needs of passengers on well patronised routes are likely to be met well but I have doubts that those in more marginal situations would be as good. Why would a franchisee propose a route it can't make a profit on within a short period?
Question 7 All urban infrastructure projects (from the largest like a metro line to small lines of just a few buses without even a terminal) would become the responsibility of privately-owned franchisee companies to plan, finance as well as operate. To what extent do you think that privately owned companies would be capable of planning, funding, constructing infrastructure and then operating their lines of urban public transport? Participant Response Jenny Buchan Depends on their fundamental competency. There is a big difference between a large company that is used to dealing with planning authorities and government, and has access to funding, and a start-up. Paul Forward Transport projects need to be planned and constructed to meet the needs of the city and its current and future inhabitants. Private operators are unlikely to represent the long-term community interests and to make the correct long-term decisions. Trevor Daley Large operations will most probably have inhouse resources or finance to employ specialists for the planning function. The issues for the smallest franchisees will be less complex and easy to plan and finance. The response by mid-sized operations may be influenced by a lack of access to appropriate resources. David Naylor Depends on management as to success Peter Warrington Government should plan railways as they have a 100-year lifetime, beyond any franchise. Roy Bishop See above! There is also a sorry history of private operators handing infrastructure back when it is nearing the end of its useful life, ready for taxpayers to again fund renovation.
Page 218 of 259
Graeme Troy Standards of comfort, safety and disabled person requirements would need to be set and met. Cash flow considerations: daily payments to franchisees, appropriate depreciation rates on capital expenditures (‘financial’ rather than ‘effective life’ based on. This may need to be some lobbying to improve tax depreciation rates and thus avoid tax generated cash flow problems. Sale of business by franchisee should be at a value determined by an independent expert. All assets to have a value assigned. This will ensure new franchisee does not overpay and has appreciated tax depreciation claims going forward. Rebecca McPhee There is evidence that privately-owned companies are capable of these activities. However, risk sharing between state and private sector would need to be considered carefully so as to best allocate to party most able to bear risk. Stephen Rowe Many franchisees would embrace this approach. Obviously revenue certainty would impact on the franchisees ability to finance infrastructure.
Wijnand Veeneman They would be capable (as we have seen in the Netherlands), but coordinative economies are lost. Here we see a move back from leaving these aspects to operators to bringing it back to the authority.
Matt Wheatley This will come down to the vetting process of franchisees. Getting the funds will not be the issue, operational capabilities would be of concern though, as well as compliance as mentioned earlier. Ian Muncaster Private companies have the capacity to plan finance and operate. There are current examples in Sydney with NW metro, airport line light rail to Randwick and some bus routes in the inner being taken over by private operators. The latter not planned new routes. The question is what capacity and preparedness to take risk would private companies have for ALL urban infrastructure. Tim Rogers There are large players in the transport industry capable of doing this but without government backing the cost may be higher for finance and the difficulties of property acquisition may impact the cost and capacity to deliver the best outcome. These may affect the ability to generate a service for the acceptable price.
Question 8 In a Community Franchise, the only control over privately owned franchisee operator behaviour is by the government owned franchisor setting the rules of operation for franchisee operators, policing them as the franchisor and controlling the cashflow to the franchisees. To what extent do you think that the structure of a Community Franchise would enable the government owned franchisor to exercise proper control over franchisee operator standards and behaviour? Participant Response Jenny Buchan Government owned franchisor could achieve this but they’d need excellent systems and the right people. Paul Forward Governments need to establish a strong regulatory regime, but enable the operator sufficient freedom to deliver for its customers.
Page 219 of 259
Trevor Daley Compliance with policy with the ‘stick” of withholding cash would be effective. The problem is ultimate use of this sanction may result in reduction in service levels. Policing would add an additional level of costs for recovery in prices. David Naylor Unsure not enough infomation Peter Warrington No strong opinion/knowledge base Roy Bishop Once you have the franchise, you have the franchise. How long has it taken governments to take control back from public hospitals being operated by private businesses even though they don’t meet standards, service levels or quality controls. Graeme Troy Depends on wording in franchise agreement and legal interpretations of such wording. Rebecca McPhee Regulation and key performance indicators regimes can be very effective were key customer outcomes are well known and measured Stephen Rowe If the rules are set and they are broken repeatedly by the franchisee then the ability to withhold funding or even tender the franchisees business would be a very powerful incentive for franchisees.
Wijnand Veeneman Again, it depends on the level of operational conditioning the legal system is allowing the franchisor. If it is limited to providing access to the market and setting price levels, it will be rather limited. With more leeway, it would be better, but could stifle entrepreneurial freedom.
Matt Wheatley I’m not quite sure that the Government does a great job ensuring franchisors are compliant overall, let alone trying to police their own franchisees. It will be a completely different relationship as opposed to employees who have a job to do, whereas an attitude of us versus them develops which occurs often in franchising. Ian Muncaster See my response to the previous question. The government has shown a very poor track record in this regard Tim Rogers For services which are economic this should work OK, assuming that there is some set of penalties built into the system. The question is what happens if the line becomes unprofitable?
Question 9 In a Community Franchise the government owned franchisor must set the level of fares by balancing the need to encourage investment (i.e. keep fares high) with the desire to keep fares at politically acceptable levels and to stimulate demand, (i.e. low). What, in your opinion, would be the likely effect on fares levels be in the application to an urban area of a Community Franchise? Would fares be likely to rise, to fall, or to stay the same? (1 will fall dramatically, to 10 will rise dramatically). Participant Response Jenny Buchan Likely to rise in areas where not many people use public transport – unless people could be encouraged to convert in large numbers away from cars and into public transport. May fall in high use areas, but no incentive for government to reduce fares.
Page 220 of 259
Paul Forward The operator needs to be able to generate sufficient return on the capital invested. Trevor Daley I think understanding is that current transport is subsidised in some areas. Unless the franchisor provides cross subsidies by using profits from high use areas low use prices should rise to allow recovery of fixed costs. David Naylor Driven by market forces Peter Warrington Would probably vary route by route. Ipart considers similar issues when setting fares now. Roy Bishop I’d expect them to possibly drop in the short term but rise overall as operators settle in. Don’t forget they will need to market/promote/sell themselves as well as run the services. Graeme Troy Difficult to say. Depends on current or pre-franchise subsidisation levels. In a franchise operation subsidies can be paid to passengers rather than reducing fares charged. Both objectives can then be achieved. Rebecca McPhee Cost recover in Sydney is currently very low (c. 25% across public transport modes). For the franchisor to operate profitably fares would need to rise significantly Stephen Rowe Stay the same, but obviously dependant on the franchisor’s desire to recover funds from the farebox. All else being equal fares would not change.
Wijnand Veeneman This depends on so many factors discussed above. Agency given to the franchisor, representation of perspectives at the franchisor, etc.
Matt Wheatley Due to there being an additional stakeholder (the franchisee), of who now also needs to be accommodated for, I would assume the prices would rise even though costs to the Government decrease with less capital investment. Ian Muncaster The reality is that fares must rise as both the Franchisor & Franchise want to make a profit Tim Rogers I doubt that the potential entrants will do it for less (other than to gain larger area) and will use the excuse of risk despite the opportunity to gain from increased patronage.
Page 221 of 259
Question 10 In the Community Franchise in the Dolomites, the government owned franchisor must operate profitably and must set fares at levels that encourage investment by private franchisees. The franchisor’s only income is from a percentage of ticket sales. The only income of franchisees is being paid for each passenger/km. The only option for passengers is to switch routes or mode to avoid a franchisee operator that is not satisfactory. To what extent do you think that the interests of the passenger, the franchisor regulator and the franchisee provider, coincide? Participant Response Jenny Buchan Interests coincide very strongly on the surface but only coincide to a limited point. Both the franchisor and the franchisee need to set the expectations of all their stakeholders at correct points. Each stakeholder has levers it can/ can not pull to get what it thinks it is entitled to. It will depend on the level of commitment each has to its statutory stakeholders – so, if the franchisee is a big company it will have to calibrate its service to the public – a contract-based level of service, with the expectations of its shareholders – a higher level of delivery that is ultimately regulated under the Corporations Act.
Paul Forward The franchisor is likely to have more of a long rem interest than the shorter-term interests of the franchisee resulting in different motives and behaviour. The level of expertise will also have an impact on the outcomes delivered. Trevor Daley Revenue for both is bums on seats. Keep the customer happy. David Naylor The market will determine sale prices and demand and supply Peter Warrington There will be areas of alignment, and areas where the social access needs of the community for access do not align with the interests of the operator. Roy Bishop Passengers don’t really have any choice if there is one provider and one route. Graeme Troy Just of matter of balance with can be achieved by setting appropriate parameters, regulations and dispute settling procedures. Rebecca McPhee Interests align for majority of passengers. However, some marginal customers may be disadvatnaged
Stephen Rowe Tying a franchisee’s revenue to a passenger/km rate means that the incentive is to carry passengers and plan effective and efficient bus routes within the guidelines given. If the payment is tied to km/ hours and bus numbers only then the incentive is to maximise km / hours / and bus numbers but not to carry more passengers effectively and efficiently.
Wijnand Veeneman Mostly aligned, but obviously, pricing is the misalligner.
Matt Wheatley There are way too many conflicts of interest at play here. The passengers will be the ones who need to compromise. Franchisees want to make a profit. Franchisor will need to keep both the passengers and the franchisees happy. Ian Muncaster I have said 5 as it is balancing act. Switching of modes or routes in reality is not practical or desirable. This question is somewhat negated as it is possible to purchase a super pass that takes in all lifts. The
Page 222 of 259
individual will assess the cost benefit of this decision ie super pass vs limited area pass Tim Rogers On well patronised and desirable routes coincidence will be high but on less well patronised routes or unserviced areas it will be lower. By definition the higher numbers will align but those that miss out will be noisy.
Questions 11 In a Community Franchise there might be a temptation for the and (government-controlled) franchisor company to favour particular Question 12 franchisees for some future reward that is not in the public interest. For example, if the franchisee regulator was to have relationships with franchisee operators, there may be pressure or inducements to change rates to favour a particular franchisee at the expense of say other operating franchisees of different modes. (Note that all operators receive the same payment/passenger/km for any particular mode). To what extent do you think that the interests of the franchisor and the operating franchisees may overlap and create conditions for regulatory capture? To what extent is this different from existing government enterprise arrangements Participant Response
Jenny Buchan Because reliable public transport is so important it must be free from self-serving political bias. Things like Mike Baird immediately stepping into a high paid bank role on quitting politics are appalling but it would be even worse if important public services/ infrastructure was compromised by political pressure or inducements.
Paul Forward While regulatory capture is a risk, it can be minimised if the Board and the regulator are separated and have the necessary level of expertise. Trevor Daley The difference is moving assessment of customer satisfaction and efficiency closer to day to day operations. There will be greater flexibility to respond to local conditions. Remuneration can be set to reflect local conditions and opportunity may exist to allow incentives at base salary levels. David Naylor Government would have more control and legislate against collusion Peter Warrington Capture is an issue with service planning and current regulation. Roy Bishop Who controls the purse strings calls the shots. Graeme Troy Will depend on the make up of the Board in Question 3. Strong independent ethical characters are essential, as in an ICAC. Rebecca McPhee This isn’t different from current outsourced operations. Strict controls in pace to prevent favouritism.
Page 223 of 259
Stephen Rowe There is always the risk of regulator capture regardless of the strategies used by Government to provide services.
Wijnand Veeneman 11: there is always a need for aligned interests between franchisees and franchisor. Creating those is the core of any model. The financing structure should align the interests of the franchisee with that of the franchisor. Capture is when there is a skewed information situation
that drives the franchisor to overvalue the interests/position of the franchisee in the view of the franchisor. This means there is a need for the franchisor to have their own expertise at hand.
12: again, not sure what to compare it with.
Matt Wheatley I’m not quite sure that the Government does a great job ensuring franchisors are compliant overall, let alone trying to police their own franchisees. It will be a completely different relationship as opposed to employees who have a job to do, whereas an attitude of us versus them develops which occurs often in franchising.
Ian Muncaster Completely different business models. The Dolomites model is for a very specific situation which I suspect has grown organically over time. The model provides benefits for potentially many disparate groups in each valley to come under one uniform umbrella whilst also allowing the Franchisor to make a profit. In many ways future planning is reasonably predictable based on real data from previous seasons. I have great reservations as to whether this can applied to transport planning in Sydney. Under current enterprise agreements each situation will be assessed under a tight risk profile vs potential reward..
Risk may be reduced by a one of capital payment rather than payment of ongoing fees.
Typically the operator would want to control fare prices. A parallel is in the leasing of space in shopping centres which is set by area, location and a fee based on turnover. In this situation the lessee sets his own price for the goods they sell Tim Rogers Assuming this is for general transport services the franchisor is dependent on getting the system to transport passengers. If the only means to do this is price and the approval of routes it is open to capture by operators who are prepared to design and offer routes which will meet its objectives.
Page 224 of 259
Question 13 What is your assessment as to this Community Franchise business regime being able to function in an urban public transport setting?
Participant Response Jenny Buchan Not sure, best thing would be to run a pilot where the government owns but franchisor and franchisee to test it and iron out the bugs before selling the franchise parts. Get a genuine commitment from the government that the franchisor aspect would be properly run and would remain in public hands.
Paul Forward There would need to be a demonstrated long term commitment from government and the private sector to the approach. The model would need to have support from both sides of politics. Trevor Daley The model may not be suitable for blanket city coverage but may be efficient for higher volume routes. David Naylor unsure Peter Warrington Should be trialled in a number of locations/routes. Possibly the Hunter or Illawarra might be suitable given the self containment. Roy Bishop Seems to me that we’d all spend a massive amount of time effort and money restructuring this system all over again. This means years of chaos and service disruption with no guarantee it will work again. Though that could mean you get rid of bureaucrats that prevent improvements too as they’d be applying for their jobs again or moving to a corporate. The dismembering of DADHC when disability services were contracted out is a case study. Graeme Troy It is not going to happen all at once. Needs to commence and be observed and modified over time as issues arise. It is a good idea that is proven in the Dolomites. Just needs to be applied/adapted to urban infrastructure. Rebecca McPhee Depends on the City Stephen Rowe No comment
Wijnand Veeneman It resembles a mix of several European models. There are huge devils in details.
Matt Wheatley Clear expectations of deliverables for both parties and tight compliance checks coupled with exit plans for the franchisor/franchisee would improve the prospect of this model however the most important factor will be profitability and return on investment for the franchisees.
Unhappy franchisees will be a larger mess than what could ever be imagined in a nonfranchised model. Ian Muncaster See my previous response Tim Rogers I can see this working in specific areas, particularly where there are established communities and there may be an incentive to provide additional services. It may need more control in developing areas where there are not those patterns.
Page 225 of 259
Question 14 What features of this business regime, if any, would make a Community Franchise unsuitable for its application to urban public transport? Participant Response
Jenny Buchan The need to provide services to geographic areas and unpopular times (eg last at night) where/ when public transport is vital but currently not reliably well patronised.
The need to provide differential fares for school children, the elderly and others in a category that was not full fare paying.
The challenge of providing transport to major events – like the Olympic Games venues – who will do this? How would the revenue be split between extra contractors and the regular franchisees?
Paul Forward Needs to have a long term perspective with a long term contract in place with a long-term commitment of the franchisee with ownership constant over the contract period. Difficult to ask for at the set up phase. Trevor Daley The challenge of assessing pricing. Potential for price increases on individual routes due to reduction in cross subsidisation. Potential disruption to service through franchisee financial failure. Payback horizons for government are influenced by political implications. Capital invested in infrastructure by Governments can be longer than commercially justified. Franchisees most probably will have shorter payback requirements. Governments can borrow at much lower rates to fund infostructure higher borrowing costs will add to ticket prices. David Naylor unsure Peter Warrington Emphasis on serving profitable routes. Would need to be accompanied by a mobility as a service response to maintain access for people no longer served by buses. Roy Bishop I think the answers throughout the survey cover this. It does not seem to me that this is a suitable model for public transport. Given the disruption underway in transport and the promise of new ways of doing things using emerging technologies (data, carshare, Ubers, self-driving vehicles etc), it would seem to me that these services will be pushed into other shapes without having to undergo this sort of wholesale reorganisation. See https://www.cityofsydney.nsw.gov.au/council/publications- updates/global-issues-ideas-and-conversations/city-talks-design/data- driven-design Graeme Troy No objection to features. Assessment of ‘applying franchisees’ essential – technical and financial capabilities.
Rebecca McPhee Very low cost recover, low density, highly subsidised operations, few competing modes / transport alternatives
Stephen Rowe I can’t see any features that would make it unsuitable, but it is reliant on the Franchisor’s ability – something that would be similar to the current tendering model used in Sydney.
Page 226 of 259
Wijnand Three main things brought up: Veeneman - What is the agency given to the franchisor?
- How is the direction of the franchisor set up, independent, public, or more towards the operational and commercial side?
- There is a real difference between building and operating skilifts from running busses, with higher investment costs, lower gains of strategic market play.
Matt Wheatley No comment
Ian Muncaster See my previous response Tim Rogers The single price and franchisee initiated proposal features seem to interact to potentially discourage services to less desirable routes.
Page 227 of 259
APPENDIX E NUMERIC DATA FROM EXPERT SURVEY
Results of Scaled Questions for Expert Survey
Differences to normal GE GE tonormal Differences
Ownership Problems Ownership
Respondent Number Respondent Q1 Membership Board as Q2Politician Q4Budget Control Q5Franchisor Planning Q6Network Capability Company Q7Private of Maintenance and Q8Control Standards Levels Q9Fare Interests of Q10 Alignment Capture Q11 Regulatory Q12 arrangements Feasibility Overall Q13 1 10 9 5 3 5 7 5 5 5 10 8 5 2 4 3 5 8 6 4 7 8 2 3 6 5 3 6 10 6 8 5 8 3 8 5 10 1 7 4 3 8 3 9 3 5 5 5 5 5 7 5 5 2 5 3 2 4 5 3 6 4 4 9 3 6 3 10 5 5 5 2 5 5 5 9 1 4 7 9 6 2 3 3 2 4 9 8 7 3 3 8 3 5 8 8 6 8 6 6 4 8 7 5 9 10 10 10 7 3 2 3 9 5 2 9 9 10 9 1 2 8 2 2 8 10 5 2 1 5 11 6 8 8 2 7 9 6 5 5 3 5 5 12 3 2 8 8 7 3 9 5 1 3 2 9 13 4 8 5 9 7 3 8 5.5 5 5 5.5 2 Mean 5.5 6.5 5.4 6.2 4.8 4.6 5.5 6.7 4.5 5.5 5.0 5.2 Standard Deviation 2.9 3.0 2.4 2.6 1.7 2.5 1.9 1.8 1.6 2.9 2.9 2.0
Page 228 of 259
Averages and Standard Deviations in accordance with Group Dimensions
Company Capability Company
Sectional
Q4 Budget Q4 Control Q5Franchisor Planning Network Q6 Private Q7 of Maintenance and Control Q8 Levels Fare Q9 Interests of Alignment Q10 Capture Regulatory Q11 GE normal to Differences Q12 Feasibility Overall Q13
Q2 Politician as Board Board as Q2Politician Membership Standards arrangements Averages Problems Ownership Q1 Questions 1 2 4 5 6 7 8 9 10 11 12 13 av franchising 6.7 7.3 5.3 6.3 5.3 6.3 5.0 7.0 4.0 7.7 5.0 5.7 av Government 5.0 7.0 5.0 5.3 4.7 4.0 5.0 6.7 5.7 8.0 3.7 4.0 av Fin/Bus 2.5 6.5 3.0 5.5 3.5 5.0 4.0 5.5 4.5 4.5 8.0 4.0 av Planning 8.3 6.3 6.7 5.7 4.0 4.3 5.7 8.0 5.0 2.3 5.0 6.3 av Transport 3.5 5.0 6.5 8.5 7.0 3.0 8.5 5.3 3.0 4.0 3.8 5.5 Confidence av Franchising 9.0 7.3 7.0 7.7 5.7 7.3 5.3 6.3 av Government 5.5 6.0 7.7 3.7 6.3 7.3 7.3 5.7 av Fin/Bus 6.0 6.0 6.5 6.0 6.0 6.0 6.0 6.0 av Planning 8.0 7.7 5.7 7.3 7.3 6.7 7.3 6.3 av Transport 6.0 8.5 8.5 8.5 9.5 9.0 7.5 8.5 Standard Deviations of Group Av. 2.1 0.8 1.4 1.3 1.2 1.0 1.2 1.1 1.0 1.5 1.7 1.0 1.4 0.9 1.0 2.2 1.6 0.9 0.9 1.0
Page 229 of 259
APPENDIX F SUMMARY OF RESPONDENT COMMENTS ON EXPERT SURVEY
Feedback On consideration of the reported responses of all experts. Participant Response
Graeme Troy The wide spectrum of opinion is noted. I am surprised at the level of seemingly strong opposition by some respondents to the concept of a "community franchise business model". To my mind such negativity is irrational. There will certainly be minor issues with application of the model, but such matters will be refined and improved over time once the system has become operational.
Tim Rogers* Noted the large range of opinion on most topics.
Saw politicians as very biased and poor board members of a government owned franchisor; they tend to put politics first. Note need for independent people who have both government and business understanding. Would like to see public servants with a role.
Comfortable with the role of government ownership of commercial ventures. This occurs now, and all shares are in the name of the minister and departmental head; not unusual and can function well.
Saw parallels with the electricity industry, where the ‘market’ is supposed to provide competitive supply of wholesale power. This was not functioning and would need to be addressed if a Community Franchise was to operate in urban public transport.
* Record of a conversation held on the 8th November 2019
Page 230 of 259
Page 231 of 259
REFERENCES
ACCONCIA, A., CORSETTI, G. & SIMONELLI, S. 2014. Mafia and Public Spending: Evidence on the Fiscal Multiplier from a Quasi-experiment. American Economic Review, 104, 2185-2209. ALEXANDER, C. 1964. A city is not a tree. 1964. AXELROD, R. M. 2006. The evolution of cooperation, Basic books. AYRES, I. & BRAITHWAITE, J. 1991. Tripartism: Regulatory capture and empowerment. Law & Social Inquiry, 16, 435-496. BAGWELL, P. 2002. The transport revolution 1770-1985, Routledge. BARKER, T. C., ROBBINS, M. & EXECUTIVE, L. T. 1975. A History of London Transport: passenger travel and the development of the metropolis, Allen and Unwin for the London Transport Executive. BASSO, L. J. & JARA-DÍAZ, S. R. 2010. The case for subsidisation of urban public transport and the Mohring effect. Journal of Transport Economics and Policy (JTEP), 44, 365-372. BEINHOCKER, E. D. 2007. The origin of wealth : evolution, complexity, and the radical remaking of economics, London, Random House Business. BEL, G. & CALZADA, J. 2009. Privatization and Universal Service Obligations. Journal of Institutional and Theoretical Economics (JITE) / Zeitschrift für die gesamte Staatswissenschaft, 165, 650-669. BENARIE, M. 1988. Delphi-and Delphilike approaches with special regard to environmental standard setting. Technological Forecasting and Social Change, 33, 149-158. BERITELLI, P., BIEGER, T. & LAESSER, C. 2007. Destination Governance: Using Corporate Governance Theories as a Foundation for Effective Destination Management. Journal of Travel Research, 46, 96-107. BLIEMER, M. C., MULLEY, C. & MOUTOU, C. J. 2016. Handbook on transport and urban planning in the developed world, Edward Elgar Publishing. CASTRO, M. & MADDOCK, R. 1997. The Universal Service Obligation for Post: Some Australian Calculations. Managing Change in the Postal and Delivery Industries. Springer. CERVERO, R. 2000. Informal transport in the developing world, UN-HABITAT. CHAN, C., FORWOOD, D., ROPER, H. & SAYERS, C. 2009. Public infrastructure financing-an international perspective. CLYDESDALE, G. 2007. Ski development and strategy. Tourism and Hospitality Planning & Development, 4, 1-23. COOKE, N. J. & SHOPE, S. M. 2004. Designing a synthetic task environment. Scaled worlds: Development, validation, and application, 263-278. CRESWELL, J. W. 2012. Qualitative inquiry and research design: Choosing among five approaches, Sage. DAL BÓ, E. 2006. Regulatory capture: A review. Oxford Review of Economic Policy, 22, 203-225. DEMSETZ, H. 1968. Why regulate utilities? The Journal of Law & Economics, 11, 55- 65. DENNETT, D. 2017. From Bacteria to Bach and Back. The Evolution of Minds. NY: Notron and Company. DIETZ, A. S., KEEBLER, J. R., LYONS, R., SALAS, E. & RAMESH, V. Developing Unmanned Aerial System Training An Event-Based Approach. Proceedings of
Page 232 of 259
the Human Factors and Ergonomics Society Annual Meeting, 2013. SAGE Publications, 1259-1262. EMERSON, D., MULLEY, C. & BLIEMER, M. C. J. 2015a. Market Initiative in Public Transport in Europe Recent Developments. thredbo 2015. Chile. EMERSON, D., MULLEY, C. & BLIEMER, M. C. J. 2015b. A theoretical analysis of business models for urban public transport systems, with comparative reference to the case of the Individual Line Ownership regime. The 14th International Conference on Competition and Ownership in Land Passenger Transport. Santiago Chile. EMERSON, D., MULLEY, C. & BLIEMER, M. C. J. 2016. A theoretical analysis of business models for urban public transport systems, with comparative reference to a Community Franchise involving Individual Line Ownership. Research in Transportation Economics, 59, 368-378. EMERSON, D. & PAROLIN, B. 2011. Time Cost Measurement of Travel in Sydney and Implications for Public Transport Patronage Potential. EVANS, A. & HOLDER, A. 1985. Bus Deregulation and Game Theory. Policy & Politics, 13, 85-92. EVANS, A. W. 1991. Are urban bus services natural monopolies? Transportation, 18, 131-150. FEHR, E. & GÄCHTER, S. 1998. Reciprocity and economics: The economic implications of homo reciprocans. European economic review, 42, 845-859. FISCHBACHER, U. 2007. z-Tree: Zurich toolbox for ready-made economic experiments. Experimental economics, 10, 171-178. FRANCH, M., MARTINI, U., LUIGI NOVI INVERARDI, P. & BUFFA, F. 2006. From reconstruction and analysis of tourist profiles to some suggestions to destination management - An empirical research in the Dolomites area. Tourism Review, 61, 30-37. FRANCH, M., MARTINI, U. & TOMMASINI, D. Mass-Ski tourism and environmental exploitation in the Dolomites: Some considerations regarding the tourist development model. 2003. International Scientific Conference «Sustainable Tourism Development and the Environment». FRANCHISE COUNCIL OF AUSTRALIA. 2016. Available: http://www.franchise.org.au/ [Accessed]. FRANKENA, M. W. 1983. The efficiency of public transport objectives and subsidy formulas. Journal of Transport Economics and Policy, 67-76. FRANKOPAN, P. 2015. The silk roads: A new history of the world, Bloomsbury Publishing. FRIJTERS, P. & FOSTER, G. 2013. An economic theory of greed, love, groups, and networks, Cambridge University Press. FUKUI, K. 1992. Japanese National Railways Privatization Study. GHAURI, P. N. & GRØNHAUG, K. 2005. Research methods in business studies: A practical guide, Pearson Education. GIULI, M. & MASELLI, I. 2015. UBER: Innovation or déja vu? CEPS Commentary, 25 February 2015. GLAISTER, S. & BEESLEY, M. 1991. Bidding for tendered bus routes in London. Transportation Planning and Technology, 15, 349-366. GOEL, V. & PIROLLI, P. 1989. Motivating the Notion of Generic Design within Information Processing Theory: The design problem space. . AI Magazine, 10, 18- 36.
Page 233 of 259
GUNAYDIN, H. M. 2006. The Delphi method. Optimization Group, accessed at http://www. iyte. edu. tr/~ muratgunaydin/delphi. htm, 16. GWILLIAM, K. 2008. Bus transport: Is there a regulatory cycle? Transportation Research Part A: Policy and Practice, 42, 1183-1194. HENSHER, D. A. & STANLEY, J. 2003. Performance-based quality contracts in bus service provision. Transportation Research Part A: Policy and Practice, 37, 519- 538. HENSHER, D. A., YVRANDE‐BILLON, A., MACÁRIO, R., PRESTON, J., WHITE, P., TYSON, B., VAN DE VELDE, D., VAN WEE, B., GUILHERME DE ARAGÃO, J. J. & DOS SANTOS, E. M. 2007a. Delivering value for money to government through efficient and effective public transit service continuity: some thoughts. HENSHER, D. A., YVRANDE‐BILLON, A., MACÁRIO, R., PRESTON, J., WHITE, P., TYSON, B., VAN DE VELDE, D. M., VAN WEE, B., GUILHERME DE ARAGÃO, J. J., DOS SANTOS, E. M., ORRICO FILHO, R. D. & HENSHER, D. 2007b. Delivering Value for Money to Government through Efficient and Effective Public Transit Service Continuity: Some Thoughts. Transport Reviews, 27, 411-448. HIRSCHHORN, F., VEENEMAN, W. & VAN DE VELDE, D. 2017. Experts’ inventory and rating of core performance indicators and organisational features in public transport: a global Delphi survey. HOFFMAN, R. C. & PREBLE, J. F. 2004. Global franchising: current status and future challenges. Journal of Services Marketing, 18, 101-113. HSU, C.-C. & SANDFORD, B. A. 2007. The Delphi technique: making sense of consensus. Practical assessment, research & evaluation, 12, 1-8. JOHNSON, D. & NASH, C. 2012. Competition and the provision of rail passenger services: A simulation exercise. Journal of Rail Transport Planning & Management, 2, 14-22. KASAI, Y. 2003. Japanese National Railways; It's Breakup and Privatisation. KETS DE VRIES & MANFRED F.R. The Psychopath in the C Suite: Redifining the SOB INSEAD Working Paper No. 2012/119/EFE. [Online]. [Accessed 12th April, 2015]. KOSOFF, M. 2014. Uber drivers speak: We are making a lot less money than Uber is telling people [Online]. Business Insider Australia Available: http://www.businessinsider.com.au/uber-drivers-say-theyre-making-less-than- minimum-wage-2014-10 [Accessed 16th June 2016 2016]. LAIRD, J. J., NELLTHORP, J. & MACKIE, P. J. 2005. Network effects and total economic impact in transport appraisal. Transport policy, 12, 537-544. LAYTON, A. P., ROBINSON, T. J. & TUCKER, I. B. 2011. Economics for today, Cengage Learning. LINSTONE, H. A. & TUROFF, M. 2011. Delphi: A brief look backward and forward. Technological Forecasting and Social Change, 78, 1712-1719. LOEB, M. & MAGAT, W. A. 1979. A decentralized method for utility regulation. The Journal of Law & Economics, 22, 399-404. MACKIE, P., PRESTON, J. & NASH, C. 1995. Bus deregulation: ten years on. Transport Reviews, 15, 229-251. MARSHALL, N., DOLLERY, B. & WITHERBY, A. 2003. Regional organisations of councils (ROCS): The emergence of network governance in metropolitan and rural Australia? Australasian Journal of Regional Studies, The, 9, 169.
Page 234 of 259
MARTIN, E., LYON, D. R. & SCHREIBER, B. T. Designing synthetic tasks for human factors research: An application to uninhabited air vehicles. Proceedings of the Human Factors and Ergonomics Society Annual Meeting, 1998. SAGE Publications, 123-127. MEES, P. & DODSON, J. 2011. Public Transport Network Planning in Australia: Assessing current practice in Australia’s five largest cities. Research Paper, 34. MINKLER, A. P. 1992. Why Firms Franchise: A Search Cost Theory. Journal of Institutional and Theoretical Economics (JITE) / Zeitschrift für die gesamte Staatswissenschaft, 148, 240-259. MITCHELL, M. 1998. An introduction to genetic algorithms, MIT press. MOHRING, H. 1972. Optimization and scale economies in urban bus transportation. The American Economic Review, 62, 591-604. MORETTI, E. 2010. Local multipliers. The American Economic Review, 100, 373-377. NASH, C. A. 1978. Management objectives, fares and service levels in bus transport. Journal of Transport Economics and Policy, 70-85. NIELSEN, G., NELSON , J. D., MULLEY, C., TEGNÉR, G., LIND, G. & LANGE, T. 2005. HiTrans Best practice guide 2. In: GOVERNMENTS UK AND SCANDINAVIA (ed.). NISHIMURA, T. 1986. Franchising in Japan: The Business and Legal Environment. Int'l Bus. Law., 14, 312. NORTON, S. W. 1988. An empirical look at franchising as an organizational form. Journal of Business, 197-218. OFFICE OF RAIL AND ROAD. 2017. About Rail, Road, Info for the Public.. [Online]. UK. Available: http://orr.gov.uk/ [Accessed 2017-04-08]. OKOLI, C. & PAWLOWSKI, S. D. 2004. The Delphi method as a research tool: an example, design considerations and applications. Information & management, 42, 15-29. PATCHELL, J. 2014. Evolution of Collective Action in a Ski Resort Facing Declining Domestic Demand and Inbound Opportunity: Hakuba's Dynamic Downsizing. Environment and Planning A, 46, 2752-2768. PETERSON, A. & DANT, R. P. 1990. Perceived advantages of the franchise option from the franchisee perspective: Empirical insights from a service franchise. Journal of Small Business Management, 28, 46-61. PLATT, B. S. & MTENGA, P. 2007. Parametric optimization of steel floor system cost using evolver. Computer Aided Optimum Design in Engineering X, 91, 119. POSNER, R. A. 1978. Natural monopoly and its regulation. J. Reprints Antitrust L. & Econ., 9, 767. PRESTON, J. & VAN DE VELDE, D. Workshop 7. Market initiative: regulatory design, implementation and performance. Thredbo 14: Conference on Competition and Ownership in Public Transport 2015 Santiago Chile. RUBIN, P. H. 1978. The Theory of the Firm and the Structure of the Franchise Contract. The Journal of Law & Economics, 21, 223-233. 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, 215-225. SCHNEIDER, A. 2015. Uber Takes the Passing Lane: Disruptive Competition and Taxi- Livery Service Regulations. Elements, 11. SHARKEY, W. W. 1979. Decentralized Method for Utility Regulation: A Comment. JL & Econ., 22, 405.
Page 235 of 259
SHARPE, S. A. 1990. Asymmetric information, bank lending, and implicit contracts: A stylized model of customer relationships. The journal of finance, 45, 1069-1087. SHIRES, J. D. & DE JONG, G. C. 2009. An international meta-analysis of values of travel time savings. Evaluation and program planning, 32, 315-325. SKULMOSKI, G. J., HARTMAN, F. T. & KRAHN, J. 2007. The Delphi method for graduate research. Journal of Information Technology Education: Research, 6, 1- 21. SPENCE, A. M. 1975. Monopoly, quality, and regulation. The Bell Journal of Economics, 417-429. STANWORTH, J. & CURRAN, J. 1999. Colas, burgers, shakes, and shirkers: Towards a sociological model of franchising in the market economy. Journal of Business Venturing, 14, 323-344. STANWORTH, J., PURDY, D., PRICE, S. & ZAFIRIS, N. 1998. Franchise Versus Conventional Small Business Failure Rates in the US and UK: More Similarities than Differences. International Small Business Journal, 16, 56-69. SUNDARARAJAN, A. 2014. What Airbnb gets about culture that Uber doesn’t. Harvard Business Review. TETLOCK, P. & GARDNER, D. 2016. Superforecasting: The art and science of prediction, Random House. TIRACHINI, A., HENSHER, D. A. & ROSE, J. M. 2013. Crowding in public transport systems: Effects on users, operation and implications for the estimation of demand. Transportation research part A: policy and practice, 53, 36-52. TRANSPORT FOR NSW 2012. The Long Term Transport Plan. In: NSW, T. F. (ed.). Sydney. TRANSPORT FOR NSW. 2016. About Transport for NSW [Online]. Available: http://www.transport.nsw.gov.au/about [Accessed]. UK GOVERNMENT. 2000. Transport Act 2000 [Online]. Available: http://www.legislation.gov.uk/ukpga/2000/38/contents [Accessed]. VAN DE VELDE, D. 2000. Dutch and Japanese railway reforms and exchanges. Japan Railway & Transport Review, 24, 10-16. VAN DE VELDE, D. Regulation and competition in the European land transport industry: some recent evolutions. Thredbo 8 conference, Rio de Janeiro, 2003. VAN DE VELDE, D. 2013. Market Initiative in Public Transport in Europe: Recent Developments. VAN DE VELDE, D. & BECK, A. 2010. Workshop report–Beyond competitive tendering. Research in Transportation Economics, 29, 145-151. VAN DE VELDE, D. & PRESTON, J. 2013. Workshop 3B: Governance, ownership and competition issues in deregulated (free market) public transport: Lessons that can be learnt from developed and developing economies. Research in Transportation Economics, 39, 202-207. VAN DE VELDE, D., VEENEMAN, W. & SCHIPHOLT, L. L. 2008. Competitive tendering in The Netherlands: Central planning vs. functional specifications. Transportation Research Part A: Policy and Practice, 42, 1152-1162. VAN DE VELDE, D. M. 1999. Organisational forms and entrepreneurship in public transport: classifying organisational forms. Transport policy, 6, 147-157. VEENEMAN, W. W. 2002. Mind the Gap: Bridging theories and practice for the organisation of metropolitan public transport, Delft University Press. VISCUSI, W. K., HARRINGTON, J. E. & VERNON, J. M. 2005. Economics of regulation and antitrust. MIT Press Books, 1. WARDMAN, M. 2004. Public transport values of time. Transport Policy, 11, 363-377.
Page 236 of 259
WELLONS, S. 2012. The Devil in the Boardroom: Corporate Psychopaths and Their Impact on Business. PURE Insights, 1, 9. WHITE, P. 2006. Competition in public transport in Great Britain. WHITE, P. 2013. An Assessment of the Competition Commission Report and Subsequent Outcomes. 13th International Conference on Competition and Ownership in Land Passenger Transport. Oxford. WHITE, P. R. 1997. What conclusions can be drawn about bus deregulation in Britain? Transport Reviews, 17, 1-16. WILKINS, G. 2016. Dumped Uber driver pleads for explanation [Online]. Fairfax Publishing. Available: http://www.smh.com.au/business/consumer- affairs/dumped-uber-driver-pleads-for-explanation-20160519-goz0dl.html [Accessed 16th June 2016 2016]. WILLIAMS, M. 1981. The economic justification for local bus transport subsidies. International Journal of Transport Economics/Rivista internazionale di economia dei trasporti, 79-88. WILLIAMSON, O. E. 1976. Franchise bidding for natural monopolies-in general and with respect to CATV. The Bell Journal of Economics, 73-104. WOLFSTETTER, E. 1996. Regulation of Monopoly, Citeseer.
Page 237 of 259