EU task force on rail infrastructure charging

1
EU task force on rail infrastructure charging
Summary findings on best practice in marginal
cost pricing

• John Thomas, Head of Regulatory Economics
• 23 October 2002

2
Objective
To describe best practice in the implementation
of the rail infrastructure charging principles as
set out in Directive 2001/14
3
The contributing Member States

• Austria
• Finland
• France
• Germany

• Italy
• Portugal
• Sweden
• UK

4
Directive 2001/14/EC Main Charging
Provisions

• Charge to be set at the cost directly incurred as
  a result of operating the train service
• a charge to reflect scarcity of capacity may be
  included
• a charge for the effect of environmental costs
  may be levied
• charges may be averaged over a reasonable spread
  of train services and times.
• Specific exceptions to the above principles.

5
Marginal Wear and Tear Costs Best Practice

• A diversity of charging approaches exist across
  Member States
• The charging group identified two alternative
  approaches as best practice, implemented in a
  number of Member States, for deriving marginal
  wear and tear costs
• The econometric approach (Sweden, Finland,
  Austria)
• The usage cost charging approach (Britain)

6
The Econometric model

• Calculates marginal costs using historic
  engineering data and costs built up on an
  activity-by-activity basis.
• Total infrastructure maintenance/renewal cost is
  established as a function of a number of
  variables (track age, length, quality, etc.).
  Resulting equation manipulated to derive marginal
  costs
• There are differences of approach between
  Finland, Sweden and Austria
• data sets used and
• renewals sometimes excluded.

7
Finland

• The cost model used in Finland was derived from
  the type of function developed by Johansson and
  Nilsson (1998) as follows
• Cit g(Yit, Uit, zit, ?it)
• Cit maintenance/renewal costs
• Yit track length
• Uit utilisation level
• zit quality vector (no. of switches, age, etc)
• ?it error term

8
Estimation Results Coefficients for Utilisation
(gross tons)
9
Results of different studies
10
Observations on the econometric approach

• Very detailed, well specified and ability to
  derive a highly disaggregate charging system
  leading to appropriate price signals.
• In practice, a network-wide marginal cost is
  applied, with disaggregation generally between
  freight and passenger traffic only.
• Consistency of approach.
• No assessment of impact of different vehicle
  characteristics on damage to the infrastructure.

11
The usage charging model

• Used in Britain
• Top down model
• total anticipated renewals and maintenance
  expenditure
• multiplied by percentage variabilities by
  infrastructure type
• equals aggregate variable cost.
• Uses basic operating (vehicle) statistics to
  calculate equivalent gross tonne mile (or
  relative damage factors)

12
The usage charging model (2)

• There is no geographical or asset type
  disaggregation in usage charges.
• The usage charges derived from the top-down model
  are therefore network-wide averages but are
  disaggregated by type of vehicle.
• Research to develop an alternative bottom-up
  engineering model to estimate marginal costs
  deemed not appropriate at this time.

13
Marginal cost

• The two approaches produce a final marginal cost
  (or proxy for marginal cost) for using the
  network.
• Marginal costs
• UK 2.7 /kGTKM (freight)
• Finland 1.2 /kGTKM (2000 study)
• Austria 0.55 /kGTKM
• Sweden 0.3 /kGTKM (1998 study)
• Large differences, even between countries using
  the same approach.

14
Comparison between Britain, Sweden, Finland and
Austria

• Scope Maintenance and renewal
• Definitions UK includes all infrastructure
  assets
• Unit costs labour etc.
• Derivation of budgets
• Output risk
• Internal charges
• Technical issues (existing condition of assets)
• Data issues

15
Conclusions on Marginal Wear and Tear Costs

• The working group concluded that it is possible
  to calculate the marginal cost of rail
  infrastructure use and it is possible at a
  practical level to adopt a marginal cost pricing
  approach.
• There may be difficulties in terms of data
  collection, data processing and establishing
  appropriate billing systems but these are not
  insurmountable.
• A degree of consistency is important, for example
  in the inclusion of renewals costs as well as
  maintenance costs.

16
Other elements of marginal social cost (1)
Charging for environmental externalities

• Emission costs and charges
• Swedish charge based on nitrogen oxide emissions
  from diesel traction
• Finnish charges are based on health impacts,
  materials damage, ecological effects and climate
  change.
• Accident costs and charges
• Swedish charge based on total external accident
  costs allocated to train kilometres and different
  for freight and passenger
• Finland distinguish between internal and
  external costs and then allocate appropriately.

17
Congestion and Scarcity pricing
Other elements of marginal social cost (2)

• Congestion costs impact of congestion / delay
  on other services / users. Levied in Britain.
• Scarcity pricing market clearing, pricing
  mechanism. Reflects the value placed on train
  paths
• Currently there is no best practice in scarcity
  pricing.
• There needs to be further research in this area.