MD-PIT: A Multi-Disciplinary study of Pricing policies In Transport

1
MD-PITA Multi-Disciplinary study of Pricing
policies In Transport

• VU, UU, RuG, TUD

2
Goal

• A theoretical and empirical investigation of the
  direct and indirect effects of transport pricing
  policies from a multidisciplinary perspective
• Behavioural responses and their consequences
• Acceptability issues of various pricing and tax
  recycling schemes
• The evaluation includes the derivation and
  formulation of policy implications
• Attention for
• Network aspects
• Heterogeneity

3
Structure Participants
4
Relation with VEV goals

• Stimulating fundamental multidisciplinary
  research
• Taken very seriously most work is joint
• Concrete societal relevance
• Obvious through focus on pricing for
  sustainability and congestion management
• Eventually concrete policy recommendations
• Closest VEV-themes
• Paying for the use of infrastructure
• Efficient use of infrastructure

5
State of Affairs

• Monodisciplinary literature reviews and
  original studies finished / in progress
• Multidisciplinary work concentrates on qnnaires
• HH1 commuting/congestion behavioural effects
  and acceptability of congestion pricing
• Currently being analyzed
• HH2 Road users in general / broader issues
  in road pricing (other externalities, longer-run
  effects)
• Behavioural effects (SR LR) and acceptability
• Currently data collection
• QF Firms responses to pricing, and
  acceptability
• Spatial behaviour
• Fringe benefits etc.
• Currently in preparation

6
Value added multidisciplinary research

• Exploit comparative advantages
• Critical view on common habits within
  individual disciplines
• Cross-fertilization
• (More) complete coverage of the field
• Never a dull moment

7
International recognition

• A bit early to assess first original papers are
  now under review for various journals
• First empirical papers are now being written and
  to be submitted later
• Individual researchers have wide international
  networks

8
Dissemination / knowledge transfer

• Papers / conference presentations etc
• Special issue of Tijdschrift Vervoerwetenschap
• Special session at ERSA 2005
• To firms
• Via upcoming Transumo initiatives
• To governments
• Via recent special report on transport pricing
  for Dutch Ministry of Transport (NoMo)

9
Progress and Prospects

• Progress
• A bit slower than planned, but good prospects for
  timely finalization
• No evident threats at this moment
• Prospects
• Very rich datasets much to do
• Other (modeling) developments
• Future activities in Transumo

10
Scientific results and plans

• PhD students will take over this presentation
• Barry Dirk Taede Geertje Dusica
• They will present 6 slides each
• General description of their PhD research
• One specific study
• General description / motivation / goals /
• Method
• Results (two slides)
• Conclusion
• Including future research

11
Economic Perspective (VU)Barry Ubbels

• Monodisciplinary work transport pricing
  (literature review) and institutional economic
  aspects (modelling)
• Multidisciplinary behavioural responses to
  transport pricing acceptability and revenue use
  (literature reviews)
• Questionnaires behavioural responses, VOT, VOR
  etc., and acceptability

12
One Study Auctioning Concessions for Private
Roads

• Institutional aspects of pricing private versus
  public provision of infrastructure with
  regulation by means of auction
• Conceptual analysis using small theoretical model
  to determine welfare effects
• Simple two link network (tolled and untolled
  route) with capacity and toll free to choose by
  private bidders

13
Set up

• Single demand function and two average user cost
  functions (free flow, congestion (varying with
  capacity) and toll)
• To concentrate on performance of different types
  of indicators
• - Competitive bidding zero profit for winning
  firm
• - Identical bidders
• Objective of government is to maximize social
  surplus
• Auction firms objective as specified by
  indicator

14
Main Results

• Design of auction (indicator) major impact on
  relative performance
• Minimisation of travel cost and subsidy divided
  by total travel demand performs relatively best
• - very close to second-best
• - requires (as second-best) a positive subsidy
• Indicator should include travel times, toll
  level and lump-sum subsidy

15
Future Research

• Analyse data from first questionnaire (acceptance
  and behavioural models)
• Analysis data second questionnaire
• Institutional economic part modelling study of
  pricing by different levels of government

16
Behavioural responsesDirk van Amelsfort

• To determine the optimal price
• under different policy objectives

Need to determine the network effects of a road
pricing measure
Need to determine the behavioural responses of
road users
17
Stated choice experiment

• Aim of the data collection is to be able to
• Estimate choice models for departure time, route
  and (mode shifts) under road pricing
• Estimate values of time resulting from the choice
  models
• This taking into account
• Heterogeneity of road users
• Unreliability of travel time

18
Preferred choices
19
Some initial results

• Parameters are significant and right sign
• Value of time components point of discussion
• Different groups can be identified, mixed logit
  is significant heterogeneity is an issue

20
Conclusions/ further research

• Useable models can be estimated from the data
• Different approaches lead to different results
• More in-depth analysis of data is necessary
  together with research partners

21
Geographical part MD-PITTaede Tillema

• General goals
• To get insight into (geographical) accessibility
  effects of different pricing measures for
  households and firms
• To get insight into relocation intentions of
  households and firms under different pricing
  measures
• Now First exploration relocation intentions
  households

22
Exploration relocation HH introduction

• Kilometer charge 30 eurocent/kilometer
• 2 types of questions
• Direct question relocation choice
• Relocation choices as option amongst other
  options
• Aims
• To get insight into the influence of the type of
  question on relocation choices households
• To get insight into explaining variables of
  household relocation due to a kilometer charge

23
Exploration relocation HH methodology

• 2 types of analyses
• Kendalls tau
• Ordered probit analysis
• Results ordered probit analysis

24
Exploration relocation HH results (1)

• Type of question influences the derived possible
  location changes
• Relocation possibility is lower for
• People who are older
• People with a higher income
• People who get a traffic cost compensation
• Positive relation between distance and relocation
  change
• Important explaining variables current
  probability of moving/changing job within 2 years

25
Exploration relocation HH results (2)

• Differences (in significant explaining variables)
  between probability changing residential location
  and work location
• Work location days in congestion, gender,
  (travel cost compensation)
• Home location moving house within 2 yrs,
  province, constraints starting time work
• Probability work location change seems gt
  probability moving house

26
Further research

• More elaborate research relocation intentions
• More pricing measures and pricing levels
• Not only household but also firm relocations
• Accessibility effects due to road pricing
• Households
• Simulation study accessibility effects
• Empirical data perceived accessibility change
• Firms
• Empirical data perceived accessibility change

27
Geertje SchuitemaUniversity of
GroningenDepartment of Psychology

• General description of PhD-project
• Factors that affect the effectiveness and
  acceptability of transport pricing policies
• Individual factors (based on social psychological
  theories)
• Features of policies (among which revenu use)

28
Questionnaire

• 515 respondents evaluated one scenario on
  kmcharge
• 7 RU types were systematically varied between
  respondents
• The scenario was judged on acceptability
  (7-point-scales 1 not, 7 very

29
7 types of revenu use
General public funds
Improve road system in general
Improve PT
Reduce costs car use, improve infrastructure
Abolish road taxes
Improve build roads
Reduce fuel taxes
30
Results acceptability of RU types
public funds
abolish road taxes
infrastructure
PT
reduce fuel taxes
31
Conclusion

• Revenue use is an important determinant of
  acceptability of pricing policies
• Other (future) research questions
• Which other features of policies affect
  acceptability of transport pricing?
• How do various features of policies affect the
  effectiveness of transport pricing
• Which individual factors affect the acceptability
  and effectiveness of transport pricing?

32
Bi-level optimal toll design problem in dynamic
networksDusica Joksimovic
PhD thesis in Transportation Modeling

• Authors
• PhD candidate ir. Dusica Joksimovic
• Promoter Prof. Dr. P.H.L. Bovy
• Daily supervisor Dr. M.C.J. Bliemer
• TU-Delft, Faculty of Civil Engineering and
  Geosciences, Transportation and Planning Section

33
Bi-level optimal toll design problem in dynamic
traffic networks
General description of the PhD research

• The aim to provide a theoretical and practical
  assessment of effects of applying road pricing
  policies in dynamic traffic networks with
  different user classes
• The focus in on time-varying (dynamic) pricing

34
Optimal Toll Design Problem with Joint Route
and Departure Time Choice
One specific research study

• General description we consider the problem of
  determining the optimal prices applying
  second-best tolling.
• Motivation to address difficulties in modeling
  and complexity in optimizing the objective
  function
• Departure time is included showing that travelers
  will change their departure times (travel earlier
  or later) and routes.

Paper will be presented at TRB Annual Meeting
2005, Washington D.C.
35
Optimal Toll Design Problem with Joint Route
and Departure Time Choice
Methods applied

• The problem is formulated as bi-level programming
  problem and formulated as Mathematical Program
  with Equilibrium Constraints (MPEC)
• Secondbest tolling scenarios are applied
• Joint route choice and departure time choice is
  modeled
• Iterative procedure (grid-search method) is
  applied to determine optimal tolls
• Up to now, small hypothetical networks are used

36
Optimal Toll Design Problem with Joint Route
and Departure Time Choice
Results framework
37
Optimal Toll Design Problem with Joint Route
and Departure Time Choice
Results
Total travel time
6500
6000
5500
5000
4500
4000
3500
3000
0
20
40
High tolls
60
5
10
15
20
80
25
30
Low tolls
38
Conclusion and further research

• This PhD addresses the complicated problem of
  determining the optimal toll design in dynamic
  traffic networks
• State-of-the art, conceptual framework,
  mathematical model are developed as well as
  iterative solution procedure applied to small
  networks
• Travelers are permitted to change their departure
  times in response to toll information
• Pricing lead to savings in the total travel time
  (compared to the no-toll situation)
• Future research global optimization methods to
  find a global optimum of the road-pricing problem

39
Overall Conclusion

• Time is too short for a full presentation
• Further information
• www.feweb.vu.nl/md-pit
• offers downloadable papers