The Future for Freight |2005>
General
This report has been prepared by Port Jackson
Partners Limited with the analysis of the economic
benefits prepared by Access Economics. The
report is based on work commissioned by Pacific
National Ltd.
The ARA would like to thank Port Jackson
Partners and Access Economics for the
preparation of the report. The provision of data
and comments provided by Pacific National
Ltd, the National Transport Commission and the
Commonwealth Bureau of Transport and Regional
Economics is also gratefully acknowledged.
ARA Office
Unit 17, Level 3, National Circuit,
Barton ACT 2600
PO Box 4864, Kingston ACT 2604, Australia
Telephone 02 6270 4500
Facsimile 02 6273 5581
Website www.ara.net.au
Published by the
Australasian Railway Association Inc
© 2005 All Rights Reserved
Designed by GRi.D, Canberra
Printed by Pirion
Intro Para The Future for Freight—economic analysis of the cost of moving freight on the inter capital city corridors |2005>
The escalating freight task facing Australia poses a growing challenge to the transport industry and governments. Sound economics and informed discussion must be progressed to provide optimum solutions or the national economy will suffer.
THE FUTURE FOR FREIGHT > iii
message from the ceo
In the past rail has failed to present transparent and
accountable information on its role in the transport equation
and has suffered a lack of funding from governments as
a consequence.
This report seeks to provide an economic analysis of the
movement of freight between the mainland capital cities.
For the first time an open, transparent and independent
assessment has been made.
The report's key findings will have significant implications
for transport planning and policy in this country. Most
importantly the report highlights rail's fundamental efficient
cost advantage on all inter-capital city corridors (not just
the longest routes).
The obvious question the report raises is, if rail is
economically so sound, why is the modal share so low
on some corridors? This raises a number of issues about
the distorting effects of current government policy on
competition, economic regulation, access pricing, and
industry structure.
The purpose of this report is to inform and progress a much
needed debate about how we better plan for the future.
This will need to include a more sophisticated approach to
investment, and a recasting of transport policies.
The Australian rail industry is seeking a Council of Australian
Govrnments (COAG) sponsored microeconomic reform
agenda that encompasses inter capital city freight, regional
freight and passenger transport. Part of this agenda, the
need to address the pricing of infrastructure access, is
covered in this report.
Bryan Nye
Chief Executive Officer
THE FUTURE FOR FREIGHT > v
preface
This important question has remained unanswered
due to the absence of a comprehensive assessment of
the comparative economics of these different but often
competing forms of transport.
The work that forms the basis of this report was initially
commissioned by Pacific National and has been made
available to the industry through the Australasian Railway
Association. In undertaking this report Port Jackson
Partners has sought to provide a comprehensive,
transparent and balanced assessment of the relative
economics of rail and road. Importantly, this report focuses
on the underlying economic cost of transport and not the
commercial cost structure, which can be distorted by
pricing and charging systems that may not reflect true
economic costs. Where there were such distortions of
significance they have been explicitly discussed.
In conducting the analysis that underpins the conclusions
of this report we have drawn on the best available
information from both public and private sources. Generally
information has been sourced from key public authorities
(e.g., the National Transport Commission, the Bureau of
Transport and Regional Economics, etc) and where multiple
sources are available, they have been considered and one
selected (or a range used) as judged appropriate.
In some areas, most notably ‘above rail’ costs, there is
less public information available at the level of specificity
required and we have been fortunate to have been given
access to internal costing information from Pacific National
which, coming from the largest inter-capital rail operator,
we judged provided an accurate insight into above
rail economics. At times in this report we have had to
withhold some detailed analysis to preserve commercial
confidentiality.
Finally, we gratefully acknowledge input from the Bureau of
Transport and Regional Economics, the National Transport
Commission and the Network Economics Consulting Group
for their input and commentary in the course of this work,
and to Pacific National for making available non-public
cost information.
February 2005
This report has been prepared by Port Jackson Partners Limited to address a fundamentally important question in freight transport—what is the relative economic cost of rail and road in inter-capital freight in Australia, and what implications does this have for the future of the rail industry and for transport public policy?
Executive summary 1
Chapter one—Introduction 9
1.1 Rail’s decline relative to road 9
1.2 Is inter-capital rail in terminal decline? 9
1.3 Rail can regain significant share, and by doing so will create significant value for the economy 10
1.4 Outline of this report 10
Chapter two—Efficient rail is lower cost than road on all inter-capital corridors 15
2.1 Above rail’s operating cost advantage 17
2.2 Above rail’s capital advantage 19
2.3 Rail’s advantage in infrastructure operating costs 19
2.4 Rail requires less infrastructure capital to meet forecast demand than road, at least in the short
to medium term 21
2.5 Allowing for the benefits of improved vertical co-ordination 23
2.6 Rail imposes significantly lower externality (indirect) costs 23
Chapter three—Constraints preventing rail reaching its natural economic potential 29
3.1 Inefficient North South below rail performance 29
3.2 Undercharging of heavy vehicles 31
3.3 Inconsistent access charging policies makes above rail investment unacceptably risky 36
3.4 Absence of a consistent approach to the assessment of road and rail capital funding 37
3.5 Negative impact of structural separation has not been properly overcome through alternative
vertical co-ordination mechanisms 38
Chapter four—Significant benefits will flow from lower cost rail growing modal share 45
4.1 Changes that have been modelled 45
4.2 Expected modal shares 48
4.3 Shifting to a lower cost transport mode drives significant economic benefits 50
Chapter five—Overcoming the constraints holding back inter-capital rail transport 61
5.1 Providing rail with a level playing field to facilitate efficient choice and appropriate investment 61
5.2 Rail needs to accelerate its internal industry reforms 62
5.3 Governments and the rail industry must together pursue these actions in structured and co-ordinated process 62
vi < AUSTRALASIAN RAILWAY ASSOCIATION
table of contents
Chapter six—Where to start—Immediate actions required to begin the reform process 67
6.1 The rail industry needs to deepen its knowledge in key areas so it can actively drive or participate
in the key public policy debates 67
6.2 Recognise that there is currently a favourable environment for change 67
6.3 While public policy changes are needed, there is much the industry can and should do today 68
Appendices 71
Appendix One—Derivation of Road and Rail Costs 73
1 Introduction 73
2 Derivation of ‘Above Road’ operating and capital costs 75
3 Derivation of ‘Above Rail’ operating and capital costs 77
4 Derivation of ‘Below Road’ operating costs 77
5 Derivation of ‘Below Rail’ operating costs 81
6 Derivation of ‘Below Road’ capital costs 82
7 Derivation of ‘Below Rail’ capital costs 84
8 Externalities 84
Appendix Two—Comparing International Road Costing Methodologies and Charging Regimes 89
1 Introduction 89
2 Why current cost allocation methods result in undercharging heavy vehicles 89
2.1 Overview of the three different cost allocation methodologies 89
2.2 Comparing outcomes from the international studies—current equity methodology results in
undercharging of heavy vehicles 93
3 Need to plan for a shift to mass distance charging 96
3.1 Comparing different heavy vehicle charging regimes 96
3.2 Emerging international experience with mass distance charging 96
4 Implications 96
Appendix Three—National economic benefits of cost savings on inter-capital rail freight 99
Appendix Four—Bibliograpy 115
Appendix Five—Glossary of Terms 116
THE FUTURE FOR FREIGHT > vii
table of contents
viii < AUSTRALASIAN RAILWAY ASSOCIATION
Exhibit 1 Direct Savings from Improving Rail's Costs and Modal Share 2
Exhibit 2 Three Key areas of Reform 2
Exhibit 3 Trends in Modal Share—Road versus Rail 5
Exhibit 4 Economic Cost Comparison—Road Versus Rail Post Rail Reform 11
Exhibit 5 Total Cost Comparison 16
Exhibit 6 Above Road/Rail Comparisons—Operating Costs 16
Exhibit 7 Above Road/Rail Comparison—Capital Costs 18
Exhibit 8 Below Road/Rail Comparison—Operating Costs 18
Exhibit 9 Below Rail Operating Cost Reduction Potential 20
Exhibit 10 Derivation of ‘Below Rail’ Operating Costs—RIC and ARTC Accounts 20
Exhibit 11 Capital Required for Growth—Road versus Rail 22
Exhibit 12 Potential benefi ts from Vertical Coordination 22
Exhibit 13 Cost of ‘Externalities’—Rural Areas 24
Exhibit 14 Total Cost Comparison—Pre RIC Cost Reduction 24
Exhibit 15 Mechanisms to Improve Relative Road/Rail Access Pricing 30
Exhibit 16 Methodologies for Calculating Road Usage Costs 30
Exhibit 17 Pay As You Go (PayGo) Road Allocation 32
Exhibit 18 International Comparison of Road Marginal Costs 32
Exhibit 19 Impact of Changes to the Current Cost Allocation Methodology 34
Exhibit 20 Access Regime Comparison—Road versus Rail 34
Exhibit 21 Comparison of Access Pricing Policy Regimes 35
Exhibit 22 Rail’s Regulatory Regime Allows for large Access Price Increases 37
Exhibit 23 Four Types of ‘Vertical Market Failure’ need to be addressed 40
Exhibit 24 Impact of ARTC Track Investment on Service Characteristics 46
Exhibit 25 BTRE Forecast Rail Volumes by Corridor 49
Exhibit 26 Relationship Between Price Discounts and Market Share 51
Exhibit 27 The Volume Shift to Rail—Rail Reform Versus Business as Usual 51
Exhibit 28 Growth in Freight Tasks—Road and Rail 52
Exhibit 29 Comparison of PJPL and BTRE Forecast Rail Volumes Across all Corridors 52
Exhibit 30 Trends in US Rail productivity: 1964-Present 53
Exhibit 31 Direct Savings from Improving Rail’s Costs and Modal Share 53
exhibits list
THE FUTURE FOR FREIGHT > ix
Exhibit 32 Broader Impact on the Economy (2014 preferred case) 55
Exhibit 33 Possible Conservative Assumptions 55
Exhibit 34 Sensitivity of Value Forecast to Model Characteristics 57
Exhibit 35 Three Key Areas of Reform 57
Exhibit 36 Further Reform is Timely 63
Appendices
Exhibit A1.1 Total Cost Reduction—Post RIC Reduction 74
Exhibit A1.2 Total Cost Comparison—Pre RIC Cost Reduction 74
Exhibit A1.3 Derivation of Above Road Operating Costs 76
Exhibit A1.4 Derivation of Above Rail Operating Costs 76
Exhibit A1.5 Derivation of Below Road Operating Costs 78
Exhibit A1.6 Comparison of Below Road Cost Allocation Methodologies 80
Exhibit A1.7 Below Road—Comparison of Cost Estimates 80
Exhibit A1.8 Below Rail Operating Cost Comparison 81
Exhibit A1.9 Derivation of Below Road Capital Costs 83
Exhibit A1.10 Road Traffi c Flows by Major Corridor 83
Exhibit A1.11 Calculating Below Rail Capital Requirements 85
Exhibit A1.12 Externality Assumptions—Road 86
Exhibit A1.13 Externality Assumptions—Rail 86
Exhibit A1.14 Cost of Road and Rail Externalities—Rural Areas 87
Exhibit A2.1 Methodologies for Calculating Road Use Costs 88
Exhibit A2.2 ‘Equity’ or ‘Club’ Approach—Basic Formula 91
Exhibit A2.3 Overview of the ‘Equity’ or ‘Club’ Approach 91
Exhibit A2.4 Costs Allocated by the ‘Indirect’ Approach—The Newbery Theorem 92
Exhibit A2.5 Comparison of Marginal Costs Derived From International Studies 92
Exhibit A2.6 UK Cost Allocation Methodology Versus The Australian NRTC Methodology 94
Exhibit A2.7 Comparison of NRTC and BTRE Allocations with Martin’s Econometric Findings 94
Exhibit A2.8 Mass-Distance Charging—Views from the Transport Economists 97
Exhibit A2.9 ‘Avoidable’ Road Wear Costs and Charges—6 Axle Articulated Truck 97
Exhibit A2.10 NRTC Average Hypothecated Fuel Charge and Avoidable Road Wear Costs 98
Exhibit A2.11 Summary of European Mass-Distance Charging Initiatives 99
exhibits list
THE FUTURE FOR FREIGHT > 1
executive summary
This conclusion stands in direct contrast to the general
perception that the use of rail for inter-capital freight
transport is in long-term decline. While rail’s declining
trend in modal share supports this observation, the
conclusion that this is somehow a natural outcome based
on competing technologies, while attractive in its simplicity,
is wrong. Rather, this decline has been due in large part
to its past legacy of fragmented public ownership and
inconsistent transport public policy.
Specifically, based on the first comprehensive review of
long-haul land transport economics, this report draws three
important conclusions:
1. In relation to inter-capital city freight, ‘effi cient rail’ is the
lowest cost land transport mode and consequently should
capture a far higher modal share than is observed currently.
2. A number of important changes are needed to achieve
this outcome:
> A ‘level playing fi eld’ between rail and road transport
for infrastructure charging and investment is needed to
ensure effi cient choices are made between transport
modes and to enable investments to be made with
certainty
> Rail needs to accelerate its internal industry reforms;
specifi cally:
- ARTC must ensure it quickly captures the expected
operational cost savings by bringing NSW track under
its management
- Above rail operators must overcome their legacy of
poor customer service
- Track owners and train operators must quickly
achieve improved vertical coordination.
3. Governments and the rail industry must together pursue
these actions in a structured and co-ordinated process.
This report draws a clear conclusion that rail is the most cost effective mode of transport for inter-capital containerised freight movements. Because of this fact, rail can be expected to increase its modal share on all inter-capital corridors. Rail will therefore play an increasingly significantly role in the nation’s inter-capital freight transport system.
EW NS Average
*After RIC cost reductions, volume increases and cost reductions from improved vertical coordination and productivity improvements
Source: BTRE; ARTC; Pacific National; Port Jackson Partners analysis
EXHIBIT 1: DIRECT SAVINGS FROM IMPROVING RAIL’S COSTS AND MODAL SHARE
Total cost savings = $26/'000 ntk
Volume shifted to rail =
Benefit from modal shift:
14.0b ntk
Average annual benefit = $370m
Value created = $5.2b NPV
Benefit on existing volumes:
Incremental cost savings = $8/'000 ntk
Existing task = 16.5b ntk
Average annual benefit = $127m
Value created = $1.8b NPV
Total value created = $7.0b NPV
EW NS Average
Rail's cost advantage over Road
$ per '000 ntkToday Estimated*
Rail's share of freight task
PercentToday Estimated*
Outcome for the economy
TodayEffect of changes discussed in this report
32
182628
18
72%
32%50%
EW NS Average
59%
16%
35%
EW NS Average
-7
EXHIBIT 2: THREE KEY AREAS OF REFORM
1. Rail needs a level playing field with road transport to ensure efficient choices are made between transport modes and to enable investments to be made with certainty. This requires consistent:
- Access usage charging methodologies
- Capital recovery policy
- Investment decision making criteria
2. Rail industry needs to accelerate internal reforms
- Reduce NSW track costs to efficient levels
- Innovate customer service offering
- Improve vertical coordination
3. A framework for Governments and the rail industry together to pursue a structured and co-ordinated process to achieve the above is required.
2 < AUSTRALASIAN RAILWAY ASSOCIATION
executive summary
Rail as the lowest cost freight transport mode
When operating the inter-capital city rail freight services
at the normally expected levels of efficiency, ‘efficient rail’
should provide a significantly lower cost freight transport
system than road on all corridors; thirty percent lower cost
on the North South corridor, and fifty percent on the East
West corridor. This conclusion is based on a ‘bottom up’,
corridor-by-corridor examination of above and below rail
and road operating and capital costs on a like-for-like basis.
Rail’s costs are already significantly below road’s on the
East West (EW) corridor. On the North South (NS), rail will
(over the next three years) become significantly lower cost
than road when NSW track operating and maintenance
costs are reduced to the Australian Rail Track Corporation’s
(ARTC) targeted levels. This, combined with small but
important improvements to the way above and below rail
operators work together, will result in ‘efficient rail’ costs
being thirty and fifty percent lower than road on the North
South and East West corridors respectively.
Rail will therefore provide major economic
benefits for transport users, and for the Australian
economy
When rail’s cost advantage over road is multiplied by the
significant achievable volume gains, there should be annual
direct cost savings to the Australian economy in the order
of $370m (Exhibit 1). The analysis shows that, on average
across all corridors, inter-capital ‘efficient rail’ freight
costs are $26 per thousand net tonne kilometres (ntk), or
2.6c/ntk, below that of road. This difference is significant
(over forty percent below average road costs across all
corridors) and, when applied to the estimated 14 billion ntk
of additional freight that can be carried by rail in ten years
time, will lead to annual savings that will steadily grow to
$370m per annum, with an overall net present value of
$5.2billion.
Additionally, the cost reductions and productivity
improvements possible will reduce the cost of the existing
rail task by around $8/’000 ntk on average, yielding benefits
to the economy of a further $130m per year, with a net
present value of $1.8 billion.
Rail reform therefore has the potential to deliver total direct
benefits to the economy of $7.0 billion, and the effect on
the wider economy will be more significant. The above
direct costs savings are estimated by Access Economics
to increase Australia’s Gross Domestic Product (GDP) by
$1.2 billion per annum by 2014 in 2004 prices. The net
present value of this annual GDP benefit is estimated to be
around $27 billion.
There is a need for significant public policy and
internal industry reforms to achieve these benefits
Rail can grow its volume and modal share with significant
policy and other supporting changes. Three important
changes are required (Exhibit 2). Specifically:
1. Policy changes are needed to ensure rail is on a level
playing fi eld with road. This requires changes in three
specifi c areas of public policy:
> First, Governments need to charge the heavier and
longer travelling trucks the true costs of the damage
they cause to roads. It is widely acknowledged that
smaller, shorter distance trucks cross subsidise
the heavier and longer travelling trucks, such as B-
doubles. What is also clear is that trucks as a whole are
signifi cantly cross-subsidised by cars in terms of the
user charges that they pay.
> Second, and closely building on the fi rst point,
Governments need also to remove the current
impediments to above rail investment by aligning the
road and rail regulatory principles for access pricing.
There currently exists the potential for track owners to
increase (approximately double) rail access fees over
time. This has the effect of putting at risk the benefi ts
from investments that above rail operators make to
expand capacity or improve effi ciency. Yet without this
investment in rolling stock and terminals, expansion of
this lower cost mode of transport cannot occur.
THE FUTURE FOR FREIGHT > 3
executive summary
> Third, important gains will come from additional
track investment above that currently planned by the
ARTC for investment and maintenance ‘catch-up’, yet
Governments continue to assess road funding more
favourably than rail funding. The fact is that, to the
extent the already mentioned bias in road user charging
and the misalignment of access regimes between road
and rail are not addressed, Governments must take on
a large role in funding this rail investment.
While road and rail are active day-to-day competitors,
Governments have set road access fees artifi cially low,
and so as not to require a heavy vehicle return on past
road investment. These factors allow road access fees
to effectively ‘cap’ and largely prohibit any return on
inter-capital track investment that would encourage
stand-alone private investment in track infrastructure1.
This is particularly relevant as the last remaining large
track investment on the east coast will, by necessity,
depend largely on private investment.
2. Rail needs to accelerate its internal industry reforms;
specifi cally:
> It is important that the ARTC reduce NSW rail
maintenance and operating costs as they have
foreshadowed. Such a reduction is appropriate, and is
fundamental to providing effi cient rail’s underlying cost
advantage on the North South corridors.
> Above rail operators must overcome their legacy of
poor customer service. Specifi cally, rail operators need
to better differentiate their price/service offerings to
target particular customer needs. In this area rail is
considerably behind other industries.
> Much closer co-ordination is needed between
track owners and rail operators. It is clear that
vertical separation has imposed large costs on rail.
Mechanisms and processes between above and below
rail operators need signifi cant further development to
alleviate these costs. Failing that, vertical separation
within rail may need to be reconsidered.
3. Governments and the rail industry must together pursue
these actions in a structured and co-ordinated process.
Given where the rail industry now stands, one cannot move
forward without the other.
Where to start?—practical suggestions on how to embark
on this process
In determining where to start, two points should be
uppermost in the minds of the rail industry. The rail
industry needs to:
> Deepen its knowledge in key areas and actively drive or
participate in the key debates. Specifi cally it should:
- Work to ensure a proper process is undertaken at
the national public policy level to align road and rail
access pricing and government funding principles
- Establish formal and robust coordination mechanisms
between above and below rail operations to remove
the obstacles to reducing operating costs and better
investment decision.
> Recognise that there currently exists a very favourable
environment for change. There is perhaps no better
indication of this than the call for reform than provided
in the AusLink green paper:
“Relying on the status quo to address these challenges
is clearly not in Australia’s interest. There is no ‘do-
nothing’ option. Incremental change is also inadequate.
Without major change to the planning framework, the
costs of providing an effective national land transport
network will be far higher. The economic and social
importance of the national land transport network
reinforces the need for Australia to undertake major
reform.”2
4 < AUSTRALASIAN RAILWAY ASSOCIATION
executive summary
1 In the intermodal freight market, rail is a price taker. Road transport effectively sets the prices that rail operators can charge, and thus the track access fees they can afford to pay to infrastructure owners.
2 Department of Transport and Regional Services, AusLink Green Paper, 2002 p.23
THE FUTURE FOR FREIGHT > 5
executive summary
EXHIBIT 3: TRENDS IN MODAL SHARE—ROAD VERSUS RAIL Percentage share of land freight by net tonne kilometres for inter capital corridors
Source: BTE Working Paper 40 'Competitive neutrality between road and rail', 1999
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
20201972 2010200019901980
Road
Rail
"With no change in relative input costs, and in theabsence of a solution to some of rail’s logistic difficulties relative to road, the long-term decline in rail’s share of the freight market is unlikely to change." BTRE, 2003 Actual Forecast
> Is inter-city rail freight in terminal decline, or can it make a significant contribution to the national economy?
> If it can make a significant contribution, what in broad terms is required to make this happen?
Key questions
Chapter 1—introduction |Ch.1>
1.1 Rail’s decline relative to road
In a 1999 paper the then Bureau of Transport Economics
(BTE) stated that...
“The long term evolution of transport modes is a
reflection of their patterns of growth and diffusion.” 3
The paper discussed a pattern of initial slow growth of a
particular technology as an idea or product takes hold;
followed by rapid growth as a ‘bandwagon’ effect then
occurs; then a levelling off as the technology matures; then,
after this period of maturity, there is finally a period of long
term decline and decay.
The BTE referred to the rise and fall of coastal shipping,
and then rail in this context. It noted that...“rail’s share has
been declining slowly but surely...” and then posed the
question... “can the march of history be altered?”
The rise of road at the expense of rail has been occurring
over at least the last 30 years. As shown in Exhibit 3, in
1999 the BTE forecast this trend to continue in inter-capital
freight based on various models of logistic substitution. The
BTE concluded that:
“With no change in relative input costs, and in
the absence of a solution to some of rail’s logistic
difficulties relative to road, the long term decline in
rail’s share of the freight market is unlikely to change.”
When forming views on rail’s future, however, it is crucial
to keep in mind some fairly recent history. It was only in
the early 1990s, for example, that National Rail was formed
out of the inter-capital city freight operations of each of the
State rail entities (National Rail has since been purchased
by Pacific National). Prior to National Rail’s formation rail
freight had suffered from poor co-ordination as there was
no single point of accountability for inter-capital freight
movement. It had also suffered from a lack of focus as
State rail entities saw urban passenger transport as their
first responsibility, particularly given the inevitable day-to-
day political pressures that come from being responsible for
vital daily commuter services. Throughout this period rail
freight, of course, was competing with a privately owned,
highly focussed, and increasingly competitive road sector.
1.2 Is inter-capital rail in terminal decline?
The starkness of the past and forecast future trends shown
in Exhibit 3, plus some understanding of recent history,
prompts a key question. Is inter-city rail freight in terminal
decline? Alternatively put, are we witnessing an inevitable
trend driven by shifting competitive technologies, or is
rail’s decline due to factors which have undermined rail’s
appropriate role in industrial transport?
THE FUTURE FOR FREIGHT > 9
chapter 1: introduction
The future of rail for inter-capital frieght transport has been long debated in the absence of a comprehensive fact base. This report provides a comparative fact base that underpins the conclusion that rail has an important future role in Australia's industrial transport system.
3 Bureau of Transport Economics, Working Paper 40, Competitive Neutrality between Road and Rail, 1999.
Given the vital role that transport plays in Australia’s
international competitiveness this is a vital question. It is
reinforced by the world class performance of the iron ore
railways in the Pilbara, and the much higher proportion of
freight carried on the East West corridor compared to the
low share of the freight task carried by rail on the North
South corridor.
Port Jackson Partners Limited (PJPL) has been
commissioned to address this question. The scope of this
study has been deliberately confined to inter-capital freight.
To address this question, and in the absence of a ready
repository of relevant and comparable cost and volume
data, the overall rail and road transport economics were
built “bottom up” using the best available data from both
the public and private sectors. Whenever possible, actual
operating data was used.
The objective of this work has been to provide an
analysis that is robust and defensible; there being no
merit in drawing conclusions based on a selective adoption
of measures that can be seen to favour one mode
over another.
1.3 Rail can regain significant share, and by doing so
will create significant value for the economy
The conclusion from this analysis is clear. Efficient rail can
significantly improve its share of inter-capital city transport
(Chapter 2) and, in so doing, make a major contribution
to the Australian economy (Chapter 4). Rail’s past decline
has been due to poor transport public policy, inappropriate
industry structures and a history of poor rail industry
performance which, together, have undermined rail’s ability
to compete with road transport (Chapter 3).
Without important policy and related changes, rail’s
situation and modal share will likely deteriorate further.
However, with these changes the next decade will see
a radical reshaping of inter-capital freight modal shares
in favour of rail. With a new approach emerging from
Governments, and now with a strong, private sector led,
commercial focus within the rail industry, major change is
both possible and can be extremely worthwhile.
On conservative estimates around $1.2 billion in direct
value (NPV) can be created through a program of reform,
and this program would increase Australia’s Gross Domestic
Product (GDP) by around $27 billion on a net present value
basis. Such reforms should see inter-capital rail freight as a
fast growing and significantly lower cost transport mode on
all inter-capital corridors.
While several changes are required (Chapter 5), the key
areas of change seem clear. With further analytical work
and wide stakeholder buy-in they are achievable.
The size of the rail reform benefits, and the need for co-
ordinated changes involving both the public and the private
sectors, makes it an important national policy agenda.
A country with vast distances between its major centres
cannot afford to carry the burden of inefficient transport in
an increasingly competitive global economy.
1.4 Outline of this report
The remainder of this report is arranged as follows:
Chapter 2 Establishes that ‘effi cient rail’ is a
considerably lower cost freight transport
mode than road on all inter-capital corridors.
Chapter 3 Identifi es the six key constraints preventing
rail from achieving its natural potential given
this cost advantage.
Chapter 4 Quantifi es the signifi cant benefi ts to transport
users, Governments, and the economy more
broadly, by rail gaining its natural share as
the cheaper transport mode (achieved by
implementing the recommendations made
in Chapter 5 to address the constraints
identifi ed in Chapter 3).
Chapter 5 Recommends key industry reforms needed
to overcome the identifi ed constraints, and
argues that it is appropriate to initiate such a
reform process now.
Chapter 6 Addresses the question ‘where to start?’, and
makes suggestions on how to practically and
quickly embark on this reform process.
10 < AUSTRALASIAN RAILWAY ASSOCIATION
chapter 1: introduction
THE FUTURE FOR FREIGHT > 11
chapter 1: introduction
EXHIBIT 4: ECONOMIC COST COMPARISON—ROAD VERSUS RAIL POST RAIL REFORM $ per '000 ntk
Source: PJPL analysis
Road
Rail
Operator costs (above) Infrastructure costs (below) Externalities Net
Operating Capital Operating Capital
11.4 + 4.1 + 3.0 + 1.2 + 6.0 = 25.7
31.0
19.6
8.54.4
9.66.6 4.8 3.6
7.3
1.3 25.7
Chapter 2—efficient rail is lower cost than road on all inter-capital corridors |Ch.2>
It has not previously been possible to draw a meaningful
conclusion as to the relative cost of road and rail because
of the absence of a comprehensive and consistent
comparative analysis. To be meaningful such a comparison
needs to be made on a corridor-by-corridor basis.
Indeed, to prepare a proper comparison there are a number
of relatively complex elements that need to be taken into
consideration in a consistent manner at each stage of the
transport network.
To prepare the necessary repository of relevant and
comparable data to underpin the analysis in this report, the
best available data sources from both the public and private
sectors have been used. The overall rail and road transport
economics have been built “bottom up”, where possible
using actual operating data. Where differing estimates exist
on some cost elements the “mid point estimate” has been
chosen with a view to favouring neither road nor rail in the
analysis. This data has then been combined to provide a
comprehensive assessment of overall road and rail transport
costs, with the assistance of leading public sector transport
economists to provide comment and guidance.
The analysis has been conducted on a forward looking
basis. That is, it is based on the costs of meeting future,
not existing, transport needs. For example, with capital
costs the focus has been on what is needed to meet future
demand, with no account being taken of past sunk costs in
either road or rail.
The overall conclusion is that efficient rail has a clear cost
advantage over road, which on average, across all corridors,
is $26/’000 ntk. Exhibit 4 summarises the relative costs of
road and rail at each stage of the cost structure, including:
(i) ‘Above’ road/rail operating costs; i.e., the cost of
running trucks and trains
(ii) ‘Above’ road/rail capital costs; i.e., the cost of
investing in trucks and trains to meet forecast demand
growth
(iii) ‘Below’ road/rail operating costs; i.e., the costs of
managing and maintaining inter-capital city roads and
tracks
(iv) ‘Below’ road/rail capital costs; i.e., the cost of
providing the necessary road and track infrastructure
required to meet forecast demand
(v) ‘Vertical co-ordination’; i.e., gains possible from
improved coordination between above and below rail
operators
(vi) ‘Externality effects’; i.e., the costs associated with
both road and rail transport where the impact is
borne by parties outside the direct road/rail transport
systems. These include factors such as the cost of
accidents, congestion (time), pollution (air, noise) and
greenhouse effects, all well recognised in conventional
economic literature.
THE FUTURE FOR FREIGHT > 15
chapter 2: efficient rail is lower cost than road on all inter-capital corridors
The analysis contained in this report concludes that efficient rail is the lowest cost mode of freight transport on all inter-capital corridors.
16 < AUSTRALASIAN RAILWAY ASSOCIATION
chapter 2: efficient rail is lower cost than road on all inter-capital corridors
Source: PJPL Analysis
* Assumes 50% reduction in RIC's costs and 100% growth in RIC's intermodal volume **Across 2014 volume shifted to Rail of 14b ntks
21
5
20
28
37
33
32
Syd - Bris
Melb - Syd
Melb - Bris
Melb - Adel
Adel - Perth
Melb - Perth
Syd - Perth
EXHIBIT 5: TOTAL COST COMPARISON$ per '000 ntk
Road RailCost benefit of Rail
64
60
58
66
63
61
57
43
55
38
38
26
28
25
Average**= 25.7
Below
Above
Variable Operating cost*Fixed operating cost*Capital recovery cost Variable operating costFixed operating costPick Up and Delivery (rail)Capital recovery cost Externalities
* Based on 38nt b-double
**Based on current PN operating performance
Source: Pacific National; PJPL analysis
38
40
39
43
45
40
39
29
32
23
30
20
21
15
9
8
16
13
24
19
25
Variable costsFixed costsCapital costsPUD
EXHIBIT 6: ABOVE ROAD/RAILCOMPARISON—OPERATING COST $ per '000 ntk
Road* Rail**Cost benefit of rail
Syd - Bris
Melb - Syd
Melb - Bris
Melb - Adel
Adel - Perth
Melb - Perth
Syd - Perth
represents a major Reducing PUD
opportunity for rail, especially on the shorter corridors
Exhibit 5 describes these same costs for each inter-capital
city corridor. It demonstrates that efficient rail has a cost
advantage on all corridors and that, not surprisingly,
rail generally enjoys a larger cost advantage on the
longer corridors.
2.1 Above rail’s operating cost advantage
The single largest source of rail’s overall $26/’000 ntk cost
advantage over road is its significant ‘above infrastructure’
cost advantage at the train versus truck running level. At its
most basic, a train can carry typical inter-capital freight over
large distances at a lower cost than a truck. Again, while
the magnitude of this cost differential varies by corridor,
it always holds true (see Exhibit 6).
In the case of road, these costs include variable costs
such as fuel and tyres, semi-fixed costs such as driver
costs, truck maintenance costs and fixed costs such as
registration, insurance and business overheads. This
analysis recognises that different truck types affect some of
these cost elements: for example, a B-double carries twice
the tonnage of a 6-axle semi but with the same one driver.
For the purpose of this exercise we have used the cost
structure of a 38 net tonne B-double which, while more
efficient than the average inter-capital city truck today,
reflects the more likely type of truck that will be used to
meet future inter-capital transport demand. Further details
on above road operating costs are provided in Appendix 1,
section 2.
In the case of rail, these costs include variable costs such
as fuel, semi-fixed costs such as crews, maintenance,
loading costs and local pick-up-and-delivery (PUD) costs;
and fixed costs such as general terminal and train control
overheads. The source of data for this was corridor specific
actual train operator costs and container traffic.
In calculating train costs some conservative assumptions
were used. For example, train lengths were assumed
to remain as they are today which, on the East Coast,
continues the restriction to 1,500m trains. This is
conservative because, for example, the same crew numbers
are needed independent of train length, and on the East
West corridor train lengths are now 1,800m. Likewise, no
allowance was made for changes to crew costs which are
currently in the process of shifting from two person crews
to ‘driver only’ operations.
Other important assumptions that affect unit cost outcomes,
such as freight densities, have been based on industry
experience. For example, the density of future additional
freight was assumed to vary by corridor (as it does to day).
Within the rail industry the lower the rail market share the
higher the density of the traffic being carried, as rail is most
competitive carrying heavier goods. Thus additional traffic
on the East West corridor was assumed to be between
250kg/M³ and 300kg/M³, 300kg/M³ on the Melbourne-
Brisbane leg and 350kg/M³ on the shorter Melbourne-
Sydney, Sydney-Brisbane and Melbourne-Adelaide legs.
Further details on above rail operating costs are also
provided in Appendix 1, section 3.
THE FUTURE FOR FREIGHT > 17
chapter 2: efficient rail is lower cost than road on all inter-capital corridors
18 < AUSTRALASIAN RAILWAY ASSOCIATION
chapter 2: efficient rail is lower cost than road on all inter-capital corridors
EXHIBIT 7: ABOVE ROAD/RAIL COMPARISON—CAPITAL COSTS $ per '000 ntk
Source: Pacific National; NECG; PJPL Analysis
7.6
8.5
8.5
10.0
11.0
8.6
7.0
Road Rail
Assumes:- 1 prime
mover and 2 trailer sets
- 5 year life- 20% salvage
value- 7% interest
rate
Assumes:- Wagon / loco
configurations varied by corridor
- 20 year life- No salvage
value- 7% interest
rate
3.7
4.4
3.6
6.7
7.7
4.5
5.0
Syd- Bris
Melb - Syd
Melb - Bris
Melb - Adel
Adel - Perth
Melb - Perth
Syd - Perth
Cost benefit of rail
Road weighted average= 8.5
3.9
4.1
4.8
3.3
3.3
4.1
2.0
EXHIBIT 8: BELOW ROAD/RAIL COMPARISON—OPERATING COSTS $ per '000 ntk
Variable costsFixed costs
Road Rail
10.1
10.2
11.6
8.6
7.3
7.6
9.6
5 axle semi
6 axle semi
> 6 axle semi
7 axle B -double
8 axle B -double
9 axle B double
Weighted avg.
- After reducing RIC costs by 50% and doubling volumes
Road weighted average= 9.6
In line with ARTC assumptions
11.4
9.8
10.6
6.6
6.6
6.6
8.5
8.0
Syd - Bris
Melb - Syd
Melb - Bris
Melb - Adel
Adel - Perth
Melb - Perth
Syd - Perth
Weighted avg.
2.2 Above rail’s capital advantage
The capital cost of moving freight by trains is also
significantly lower than by trucks, in fact about half the cost
(Exhibit 7). In both cases the ‘power unit’ (prime mover or
locomotive) is relatively expensive compared to the ‘carrying
unit’ (trailer or wagon). In the case of trains, however,
each locomotive can haul more carrying units on a dollar
equivalent basis compared to a prime mover.
In the case of trucks, this capital analysis was based on a
B-double to be consistent with assumptions made for the
above road operating cost analysis.
In the case of rail, capital costs were calculated using
corridor specific locomotive and wagon configurations.
These are dependent on train lengths and route
characteristics. Appendices 1, sections 2 and 3 provides
more detail on both above road and rail capital costs.
2.3 Rail’s advantage in infrastructure operating costs
Taking into account planned changes to below rail
operating costs on the North South corridor it can be
concluded that rail will have, on average, a small but
meaningful cost advantage over road in terms of the
annual operating and maintenance costs for the track
or road infrastructure (see Exhibit 8). This advantage is
significant on the East West corridor where the track is
generally in better overall condition (for example, due to
the concrete sleepering upgrade in the 1990s) and is an
inherently easier environment from a track maintenance
perspective (drier, more level terrain) compared to the
North South corridor.
In the case of road, infrastructure operating costs include
all items included in the national and state funding budgets
for road, excluding any capital expenditure. The allocation
of these costs between cars and trucks are dependent in
part on the axle load impact on the road and are discussed
in detail in Appendix 1, section 4.
There is some debate about the most appropriate
methodology for determining the impact of heavy vehicle
traffic on road maintenance expenditure. To summarise,
the National Road Transport Commission (NRTC) and
Bureau of Transport and Regional Economics (BTRE)
have drawn different conclusions on this issue. While the
NRTC methodology likely significantly underestimates the
impact of heavy vehicles on road expenditure, the empirical
research and emerging overseas evidence suggests the
BTRE’s estimates may also underestimate the level of
road costs attributable to the heavier and longer travelling
trucks. Nonetheless, for the purposes of Exhibit 8 we have
used the BTRE’s cost allocation methodology. This issue is
discussed further in 3.1 below.
In the case of rail costs, infrastructure operating and
maintenance costs include current actual expenditure
on the East West corridors and estimates of ‘efficient’
operating costs on the North South Corridors. The key
sensitivity in this analysis is the magnitude of the rail
infrastructure operating cost reduction expected on the
North South corridor.
The Australian Rail Track Corporation (ARTC) is seeking to
reduce the infrastructure costs on the North South corridor
to around half their current levels. In early September
2004 the ARTC took operational control of the NSW inter-
capital track. The ARTC had determined its view as to the
cost outcomes possible on the North South corridor as
a consequence of a restructured organisation under its
operating principles.
The ARTC’s cost reductions are based on operational
changes and some capital programs specified by the ARTC
within its overall merger plan.
THE FUTURE FOR FREIGHT > 19
chapter 2: efficient rail is lower cost than road on all inter-capital corridors
20 < AUSTRALASIAN RAILWAY ASSOCIATION
chapter 2: efficient rail is lower cost than road on all inter-capital corridors
EXHIBIT 9: BELOW RAIL OPERATING COST REDUCTION POTENTIAL
Source: ARTC 2002 Annual report; ‘Independent Review of RIC Metropolitan Maintenance Funding’ (Oct 2002); PJPL analysis
Total cost$ Millions
Unit cost$ per '000 ntk
Comments
Typical cost reduction of 40% in an inefficient organisation
- PJPL experience- Public sector privatisation - ARTC plans
ARTC estimate: $170m; NECG estimate $120m
n.a .
n.a .
32.0
3.0
16.0
4.7
11.6
5.7
Excludes passengers
Coal = 70; Grain = 179
Typical merger synergies
100% volume increase
5b ntk growing to 10bn ntk
10.5b ntk
409
70
160
64
14
82
34
116
66.7
179
Total RIC cost
Non -intermodal cost
Intermodal RIC cost
Inefficiencies (40%)
Merger synergy (15%)
Efficient RIC cost structure
Impact of volume increase
Achievable RIC cost
ARTC cost
In-line withARTC cost reduction projections of 58% over 5 years
13.0
Insurance
EXHIBIT 10: DERIVATION OF ‘BELOW RAIL’ OPERATING COSTS—RIC AND ARTC ACCOUNTS$ Millions
Source:
*Excludes depreciation and costs associated with non-intermodal traffic
Over ~5b ntk
~10b
32.9
14.4
8.7
97.7 2.9
35.7
6.012.6
5.0
6.4
4.9
5.7
39.1
49.8
6.0
4.3
13.2
16.8
5.0
RIC 02/03Budget*
Costreductions Merger
Synergies
RICnew coststructure
ARTC cost structure
Routinemaintenance
Major periodicmaintenance
External asset mtce Network services
External asset mgmtP/E, Materials
Overheads 160.0
66.7
Employees
Infrastructure Mtce
Operating Lease expensesProject/Development costs
Incident costsOther
64.0
81.6
Over ntk
ARTC 2002 Annual report; "Independent Review of RIC Metropolitan Maintenance Funding (Oct 2002); PJPL analysis
The cost reductions used in this analysis are judged to be
realisable, for three reasons.
> They are in line with the cost reductions that the
ARTC has assumed it will make on taking control of
the NSW inter-capital freight track. The ARTC had to
make a very practical assessment as its future business
viability depends on these savings being achieved.
The previous NSW Government-owned track operator
received large subsidies which ceased once the ARTC
took control of the track.
> These cost reductions accord with industry experience.
PJPL has observed major private sector performance
improvement programmes in many companies.
These have usually achieved cost reductions of
around 40% in organisations that have not had such
a programme for many years, especially in public
sector organisations. In addition, cost synergies (such
as overhead reduction with removal of one corporate
centre) of 15% are also commonly seen when two
companies have merged. Typically, these programmes
take up to three years for the full impact to reach the
bottom line of the business.
> Even after these cost reductions have been made, and
once North-South volumes have, say, doubled to reach
the current ARTC targets, NSW infrastructure operating
and maintenance unit costs (i.e., $/ntk) will still be
approximately double those currently being incurred by
the ARTC on the East West. There are obvious reasons
why NSW infrastructure costs should be higher,
as already discussed. These reasons may perhaps
justify double the maintenance expenditure, but they
would not appear to justify anywhere near the current
expenditure levels, as much rail maintenance is either
‘routine’ or based on standard life cycle considerations.
Exhibits 9 and 10 outline the North South Corridor rail
track operating and maintenance costs used in this
analysis, and illustrate the adjustments made. Further
details are in Appendix 1, section 5.
Overall, when these cost reductions are achieved ‘efficient
rail’ will have a material infrastructure operating and
maintenance cost advantage over road.
2.4 Rail requires less infrastructure capital to meet
forecast demand than road, at least in the short to
medium term
Our modelling has considered the infrastructure capital
required to meet the forecast growth in the transport
task over the next 10 years. The analysis recognises that
beyond this horizon, a fundamental change to the track
infrastructure on the East Coast would be required for
sustained growth. The modelling described in this section
considers the capital required to accommodate the target
freight volume in 2014, with no further growth for rail after
that point. On this basis, we conclude that the incremental
capital required for rail is less than that for road.
It is important to understand that, with small capital
investments to improve rail service levels, large
improvements in volume are achievable. The reason is
that the current infrastructure is capable of carrying many
more services—the limiting factor is the service level
(i.e., cut-off times and transit times) that operators can
provide to customers which can be improved with better
above and below rail co-ordination and with some modest
capital outlays.
Calculating capital costs to accommodate growth is
complex. Considerable effort has gone into ensuring that
the conclusions in this analysis have been properly drawn
(Exhibit 11).
In the case of road, capital investment forecasts detailed
in the BTRE’s Working Paper 35 “Roads 2020” have been
used and traffi c growth by vehicle type has been modelled
in terms of passenger car units (PCUs), which determine
the consumption of road capacity. The relationship between
heavy vehicle traffi c growth and expenditure is used to
quantify the capital required per unit of freight growth on
road, and hence the capital expenditure avoided through
modal shift to rail. This has been done for each major
highway. This is discussed in more detail in Appendix 1,
section 6.
THE FUTURE FOR FREIGHT > 21
chapter 2: efficient rail is lower cost than road on all inter-capital corridors
22 < AUSTRALASIAN RAILWAY ASSOCIATION
chapter 2: efficient rail is lower cost than road on all inter-capital corridors
EXHIBIT 11: CAPITAL REQUIRED FOR GROWTH—ROAD VERSUS RAIL*
Source: BTRE Working Paper 35: Roads 2020, 1997; PJPL analysis
*1.2m PCU÷3.5PCU per truck = 340,000 trips per year. 340,000 x 20t per truck x 1200km per trip = 8 billion net tonne kikometres
truck traffic
1.0 PCU/ veh
0.0m PCU
1.7 PCU/ veh
0.0m PCU
3.5 PCU/ veh
1.2m PCU
1.2m (4%)
Growth capitalper PCU$156/PCU
Allocated growth capital$0.2b
Growth in road freight8.0b ntk
Benefit of modal shift to rail0.1/'000ntk
Benefit of modal shift to rail$1.1/'000ntk
Road Capital Deficit$1.7/'000ntk
Rail Capital Deficit$1.6/'000ntk
Road Growth Capital$3.1/'000ntk
Rail Growth Capital$2.0/'000ntk
12.8m
5.9m
1.1m
after modal shift
Change in
PCUsTotal benefit of modal shift $1.2/'000ntk
Intercapital freight
Allocated capital deficit$0.1b
Capital deficit per PCU$113 /PCU
Local freight
Passenger vehicles12.8m
5.9m
1.4m
before modal shift
Growth in road freight8.0b ntk
EXHIBIT 12: POTENTIAL BENEFITS FROM IMPROVED VERTICAL COORDINATION
Source: Interviews with Pacific National operators
Possible initiative Detailed explanationPotential value
($'000 per annum) Key assumptions
On track refuelling Providing refuelling facilities on key loops would free up capacity/time in yards
500 On track refuelling saves 1 hour per journey on affected routes-saving crew time, allowing more effective use of yards and improving service reliability etc.
Improved sharing of train position data
Rail operators use GPS systems to track train movements. This data could be used by train controllers to improve network management
400 Key savings are crew time (e.g. overtime), headcount used in train monitoring activities, service penalties
Upgrading of remaining manual signals and points
Manual signals require driver to stop train
500 In deciding when to upgrade signals and points the track owner does not account for the fuel saved by not stopping and starting, injuries avoided, time saved, etc.
Wheel grinding Improved maintenance of wheels reduces their impact on track wear
tbd Cost of frequent wheel grinding is less than the cost of rail grinding/replacement. Whole of rail decisions are not being taken
EXAMPLES
In the case of rail, the required level of capital investment
to improve service levels and free up new capacity has
been comprehensively modelled. Initial increases in volume
can be achieved with limited capital investment by, for
example, better utilisation of both containers and slots.
However, as volumes increase, additional capacity becomes
increasingly expensive. Ultimately, additional capacity
can only be increased by adding or extending passing
loops to allow 1,800–2,000 metre trains, and/or by double
stacking, both of which are expensive from an infrastructure
investment perspective.
The analysis assumes that around $1b is spent on the North
South corridors. This is consistent with the $600m ARTC will
spend (as part of an overall $870m planned expenditure in
taking over the NSW track) and an additional $450m which
has now been made available for further improvements (as
part of the overall $900m AusLink funding). This funding
will overcome the existing maintenance defi cit and provide
capacity and service performance improvements consistent
with rail achieving the modal shares being forecast by the
key industry players.
Appendix 1, sections 6 and 7 provide more detail on below
road and rail capital costs.
2.5 Allowing for the benefits of improved
vertical co-ordination
This analysis has taken only minor account of the many
benefits available from improved co-ordination between
above and below rail operators. Only $1.00 of efficient
rail’s $26/000 ntk cost advantage over road is due to
our assumptions on the benefits of improved vertical
co-ordination. The benefits from improved vertical co-
ordination are embedded in the cost figures already
discussed. Exhibit 12 lists some examples and analysis
that have been used to derive our assumptions.
In essence, most actions of the track owner affect the train
operator in important ways, and vice versa, but each almost
always makes independent decisions due to the arms
length nature of the relationship on most of the inter-capital
track. Existing consultation processes are only a starting
point compared to the level of co-ordination required.
While work to date indicates that much larger gains are
possible, more effort would be required to substantiate
gains larger than the $1.00/000 ntk used in this analysis.
The importance of improved vertical co-ordination has
become apparent through discussions with industry
participants and is dealt with in more detail in Chapter 3.
2.6 Rail imposes significantly lower externality
(indirect) costs
In attempting to assess the relative transport costs of road
and rail it became apparent that there are real costs of
both modes that are not borne directly by either rail or road
operators or users. These costs, described by economists
as ‘externalities’, are no less real out-of-pocket costs. While
somewhat more difficult to assess than direct operational
costs, there are well established Australian and international
methodologies and data to estimate these costs. Australian
estimates of rail’s externality costs are significantly lower
than that of road's under any methodology.
While there are arguments for both raising and lowering
aspects of any one measure, rather than enter into a
detailed methodological debate, a mid-point estimate of the
upper and lower range of these estimates has been used in
this analysis (Exhibit 13).
Accident externalities are responsible for much of the
total difference between road and rail externalities. It
its well established that interstate heavy vehicles create
considerable costs for other road users because they cause
a relatively high number of accidents, which is not the
case with inter-capital freight trains4. Much of the public
opposition to the level of freight carried on the road seems
at least intuitively based on this observation. Appendix 1,
section 8 provides more details on externalities.
THE FUTURE FOR FREIGHT > 23
chapter 2: efficient rail is lower cost than road on all inter-capital corridors
4 Laird P., Land Freight External Costs in Queensland, 2002
24 < AUSTRALASIAN RAILWAY ASSOCIATION
chapter 2: efficient rail is lower cost than road on all inter-capital corridors
EXHIBIT 13: COST OF ‘EXTERNALITIES’—RURAL AREAS $ per '000 ntk
Source: BAH; NRTC; BTRE; Qld Transport; PJP analysis
* Qld Transport assume $25/t of CO2, Bus Industry Confederation assume $40/t of CO2
**Note that increased rail usage will incur road congestion costs around terminals
Low Case High Case
Greenhouse Gases*Accident costsNoise PollutionCongestion Costs**
4.6
0.81.4 0.6
3.2
0.2
Road Rail
10.0
1.6
1.7 1.1
7.0
0.3
0.50.8
Road Rail
Difference
3.8
6.0
8.4
0.8 0.6
3.0
6.7
0.30.8
Low Case
Assumed High Case
EXHIBIT 14: TOTAL COST COMPARISON—PRE RIC COST REDUCTION $ per '000 ntk Capital recovery cost
Variable Operating cost
Capital recovery cost
Fixed operating cost
Externalities
Variable operating cost
Pick Up and Delivery (rail)Fixed operating cost
43
55
38
38
26
28
25
Source: PJPL Analysis
Rail (RIC today)Cost benefit of Rail
Below
Above
Rail (RIC reduced)
(2)
(10)
(0)
28
37
33
22
Syd - Bris
Melb - Syd
Melb - Bris
Melb - Adel
Adel - Perth
Melb - Perth
Syd - Perth
66
69
59
38
26
28
35
Today = 3
Efficient operator = 17
Chapter 3—constraints preventing rail reachingits natural economic potential |Ch.3>
There are five critical structural barriers to rail achieving its
‘natural’ potential:
> Costs on the North South have not yet reached ‘effi cient
levels’
> Road usage (access) charges for heavy vehicles, which
compete most directly with rail, are being subsidised
by other vehicles; making rail appear relatively more
expensive to freight users
> Inconsistent access charging policies between road
and rail results in above rail operators having much less
certainty in the potential returns from new investments.
This makes capital investments for either expanded
capacity or improved productivity in above rail
unacceptably risky
> Inconsistent funding decision making criteria between
road and rail infrastructure results in rail receiving less
investment than it otherwise should
> Vertical separation between above and below rail
operators in the main inter-capital city networks drives
higher costs for all parties and make major investment
decision making more diffi cult.
Combined, these structural barriers have two fundamental
and debilitating impacts on the rail industry. Firstly, they
increase operator costs. Secondly, they mask the true
economic costs to transport users, and returns to investors;
resulting in both making choices that would be otherwise
made differently.
3.1 Inefficient North South below rail performance
As discussed earlier in Section 2.3, targeted below rail
operating costs for the NSW track are much lower than
those being incurred by the then NSW Rail Infrastructure
Corporation (now part of ARTC). As shown in Exhibit 14
(and in contrast to Exhibit 5 above), in the absence of
these NSW track maintenance cost reductions, rail has
no material cost advantage on the North South corridor.
THE FUTURE FOR FREIGHT > 29
chapter 3: constraints preventing rail reaching its natural economic potential
The economic advantages of efficient rail over road are clear and compelling. However this begs the obvious question as to why rail’s modal shares, particularly on the North South corridor, do not reflect such an advantage.
30 < AUSTRALASIAN RAILWAY ASSOCIATION
chapter 3: constraints preventing rail reaching its natural economic potential
EXHIBIT 15: MECHANISMS TO IMPROVE RELATIVE ROAD/RAIL ACCESS PRICING
Area Issue Policy change required
Heavy vehicle cost estimation methodology
- Current allocation template underestimates costs due to heavy vehicles
- Bring road access pricing practice into line with the emerging international evidence/theory
Mass-distance charging - Fuel based charges underrecover costs from the heaviest vehicles
- Begin a process of shifting road access pricing from fuel levies to mass-distance charges
Externalities - Rail imposes lower external costs than road (e.g. accidents, congestion, pollution), but this is not factored into modal choices
- Ensure that road and rail access pricing reflects externalities as well as direct costs
Description Examples* Possible impact
1. 'Equity' - Allocate ALL costs between users (the current Australian PayGo regime is an example)
- May or may not refer to marginal costs as a lower bound for allocations
- NRTC approach
- UK NERA approach
- US federal studies
- EU Commission study
- Outcome is heavily dependent on how 'non-separable' costs are allocated —by VKT or PCU. Current dominant methodology internationally; favours heavy vehicles if non-separable costs are allocated by VKTs
2. Engineering - Estimate the marginal cost of road usage, including impact on other road users due to road damage, based on engineering models
- 'Direct'— uses pavement management system models (e.g. HDM 4) to estimate the marginal cost of road use
- 'Indirect'— uses Newbery's theorem, linking ESALs to wear
- If Marginal Cost < Average Cost then will reduce costs and will fall short of PayGo
- If Marginal Cost = Average Cost, then because allocations are made based on ESAL's, it will result in increased heavy vehicle costs
3. Econometric - Use economic models on historical datasets to estimate the impact of traffic on costs
- Only works if there are strong datasets (few available now)
- VKT*, GVM*, ESAL* are correlated, making estimation of impact difficult
- The Link Study (2002)
- Li et al. (2001)
- Martin (1994)
- If successful, likely to result in increased allocation to heavy vehicles BUT currently does not take account of the affect of road damage on other vehicles (road damage externalities)
EXHIBIT 16: METHODOLOGIES FOR CALCULATING ROAD USAGE COSTS
*VKT = Vehicle Kilometres Travelled; PCU = Passenger Car Unit; GVM = Gross Vehicle Mass; ESAL = Equivalent Standard Axle Load
Source: Data cited in "Measuring the Marginal Cost of Road Use—an International Survey", Nils Bruzelius, 2003
3.2 Undercharging of heavy vehicles
It is widely accepted that the heaviest trucks do not pay
their appropriate share within the current pricing principles.
The Commonwealth Government in its 2002 Transport
Green Paper stated that:
“While charges for heavy vehicles are calculated by
the National Road Transport Commission (NRTC) to
recoup the costs of road wear... those trucks that carry
greater than average loads and travel greater than
average distances bear less than the costs allocated
to them by the NRTC”. 5
Trucks have a natural advantage over rail over shorter
distances, however, the current charging regimes blurs
these economics. There are significant inherent cross
subsidies from trucks with lighter loads moving over shorter
distances to those travelling the longer inter-capital city
corridors with heavier loads. In effect, short distance truck
operators are charged higher access fees than are justified
and interstate truck operators are charged less.
Furthermore, passenger vehicles subsidise all types of
trucks, again biasing charges in favour of the heaviest
vehicles. As a result rail, which is actually cheaper, safer
and less polluting, is chosen less often by transport users
than it otherwise should be.
The bias in user charging that favours the heaviest vehicles
occurs in more ways than is commonly realised. Exhibit 15
illustrates three mechanisms through which relative access
pricing between road and rail is flawed.
3.2(i) The currently employed vehicle costing
methodology is flawed
The currently employed methodology is out of line with the
emerging international evidence and theory. This work,
which has accumulated over at least the last 15 years,
allows road access charging methodologies to be divided
into at least three categories (Exhibit 16).
Australia, like many other countries, uses what may be
described as an ‘equity’ allocation approach. It seeks to
allocate all costs between users in an ‘agreed’ fashion
based on prescribed principles that, among other things,
emphasise simplicity, efficiency and equity. More recent
work suggests that this approach is inferior to both
‘engineering’ (based on engineering models of vehicle-
pavement interactions) and ‘econometric’ (based on
empirical studies of road use and pavement cost data)
methodologies. These have a common link in that they
seek to base charges on the actual marginal costs of road
usage, rather than sometimes discretionary cost allocations.
Furthermore engineering models can be used to quantify
what are termed ‘damage externalities’—the costs imposed
on other vehicles imposed by an individual user’s road
damage, which can be significant particularly on older and
less 'robust' roads. The ‘engineering’ approach seeks to
use pavement management systems that determine when
various maintenance tasks are necessary. The ‘econometric’
approach uses historical data to estimate the actual impact
of traffic on costs. It is empirical, not theoretical.
The current approach results in low allocations to interstate
freight for a number of reasons. The first relates to the
application of the ‘equity’ approach.
The NRTC divides all road expenditure into ‘allocated’ and
‘non-allocated’ categories. The latter includes expenditure
on vehicle registration and heavy vehicle registration costs.
It is unclear why this approximately 14% of expenditure is
excluded for purposes of calculating road user charges.
Most important, the NRTC then divides the allocated
costs into separable and non-separable costs. The latter
represents 70% of allocated costs, as shown in
Exhibit 17. This expenditure is supposed to correspond
to the cost of building a minimum standard road, as well
as some minor operational expenditure such as mowing
roadside verges.
THE FUTURE FOR FREIGHT > 31
chapter 3: constraints preventing rail reaching its natural economic potential
5 Department of Transport and Regional Services, AusLink Green Paper, 2002
32 < AUSTRALASIAN RAILWAY ASSOCIATION
chapter 3: constraints preventing rail reaching its natural economic potential
Source: NRTC Technical Paper, September 1998; BTRE working paper 40 "Competitive neutrality between road and rail",1999; Team analysis
ESAL = Equivalent Standard Axles LoadsGVM = Gross Vehicle Mass-passenger vehicles are given a value of 0.0 PCU = Passenger Car Units-a measure of the space taken up by a vehicle on the road VKT = Vehicle Kilometres Travelled
Total road expenditure
- Opex (e.g. servicing
- Capex (e.g. pavement construction)
Allocates all costs to trucks with ~55% going to the heaviest vehiclesAllocates all costs to trucks with ~50% going to the heaviest vehiclesAverage heavy truck modeled as being equivalent to 3.5 cars based on 'footprint'
Cars and trucks treated as having equal impact on road costs
NRTC does not differentiate between cars and trucks for 70% of road costs
Allocated$9.1
allocated$0.3
All costsNon-separable
Separable30%$7.8
70%$1.3
EXHIBIT 17: PAY AS YOU GO (PAYGO) ROAD ALLOCATION
% of costsESAL
GVM
PCU
PCU
VKT
GVM
Parameter
26%
31%
0%
0%
100%
44%
VKT 0%
$9.4
$ per '000 ntk
NRTC
6%
8%
6%
89%
5%
74%
12%
45%$11.8
55%$3.3
$15.1
$0.3
$15.4
BTREImpact from moving from:
142
EXHIBIT 18: INTERNATIONAL COMPARISON OF ROAD MARGINAL COSTS
Marginal cost
Aust. cents/km
Swissstudy
Austrianstudy
NERA/ITS
NRTC*
EQUI
TYEC
ONOM
ETRI
C**
***Imputed externality based on costs derived from Swedish study, which explicitly includes road damage externality costs
Source: "Measuring the marginal cost of road use—an international survey", Bruzelius 2003; "Updating Heavy Vehicle Charges", NRTC 1998
ENGI
NEER
ING Swedish
Direct
6.0 PCU
Imputed road damage TrucksCars
'externality' cost**
SwedishIndirect
Martin(1994)Rosalin/Martin(1999)
Truck:Car cost ratio
Truck c/km as multiple of car c/km
CarsTruckCarsTruckCarsTruckCarsTruckCarsTrucksCarsTruckCarsTruckCarsTrucksCarsTrucks
Two opportunities:
1) Change parameters used in current equity system
2) Change to engineering based allocation
Relative allocation
31
31
15
142
53
88
15
10
31
0.436.37
0.1420.10
0.2312.27
0.4813.76
0.182.73
0.65
0.4813.76
0.4813.76
0.4813.76
11.51
% share of total costs
47
9
20
32
47
43
32
32
32
53
91
80
68
53
57
68
68
68
NRTC 'Marginal cost' calculated from maintenance costs only to be consistent with international studies
The NRTC allocates 100% of these non-separable costs
by vehicle kilometres, and so treats a car and a B-double
in the same way. At a minimum this expenditure should
instead be allocated using Passenger Car Units (PCUs),
which is a capacity measure of the space taken up by a
vehicle on the road. Using PCUs, a B-double is equal to
four cars, and this is more closely representative of the
impact of different vehicle types on the need to incur
non-separable costs. This change alone would signifi cantly
re-weight road user charges in a more equitable fashion.
The second reason for a low allocation to heavy vehicles
flows from the findings of the ‘engineering’ or ‘econometric’
approaches. On the one hand, using this research would
likely see non-separable expenditure at less than 70%
of allocated costs. That is, more expenditure would be
allocated by vehicle mass or axle loads than is currently
assumed. More work is needed to determine the
appropriate extent of non-separable expenditure.
On the other hand, research suggests that loaded axles
per vehicle (referred to as Equivalent Standard Axle Loads,
or ESALs), rather than vehicle mass per se, is a better
parameter for the allocation of separable expenditure.
This is the consensus of the emerging evidence from the
econometric and engineering models, which show that axle
configurations make an important difference to the level of
damage caused by vehicles, even when the gross vehicle
mass is the same, and so are a better predictor of the costs
an individual user will impose on the road system.
The effect of these assumptions is that the NRTC’s
application of the equity approach results in lower
allocations of costs to heavy vehicles than other countries
which use a comparable technique (e.g. the UK, which
uses similar cost categories, but makes greater use
of mass and axle-mass parameters to apportion costs
between users) and those that use different approaches
(i.e., engineering and econometric approaches). Exhibit 18
shows a comparison of the NRTC’s methodology with the
outcome of several studies using alternative methodologies.
Without trying to resolve this issue conclusively, some
adjustments to the NRTC methodology were made in
estimating heavy vehicle costs. Exhibit 17 shows how the
BTRE would alter the expenditure allocation parameters6.
Exhibit 19 shows how various changes in road cost
allocation would affect road user charges. It is more than
likely that with further research even the BTRE assumptions
may prove to be too conservative.
Given that much of the research in these areas has been
initiated in Australia, refinement of these models could be
done relatively quickly. Appendix 2 brings together some of
this research to form an important starting point.
3.2(ii) Absence of mass distance charging
is a second methodological flaw
There are further problems with the current charging
methodology for heavy vehicles that also must be
addressed. The BTRE has recently described this
problem clearly:
“Specifically, the current fuel-based heavy vehicle
charges increase linearly with distance but at a
declining rate with respect to vehicle load... for more
heavily laden vehicles the costs of road wear per net
tonne-kilometre increases with mass whereas the fuel-
based charge per net tonne-kilometre decreases with
mass.” 7
A related issue, according to the BTRE, is that:
“Registration charges are set based on fleet average
utilisation. The effect is that vehicles that carry less
mass or travel below average distances pay a higher
per unit road use charge than vehicles carrying more
mass or travelling above average distances”.
Both these provide a further inherent cross subsidy for the
heavy long-haul trucks which compete with rail.
THE FUTURE FOR FREIGHT > 33
chapter 3: constraints preventing rail reaching its natural economic potential
6 Bureau of Transport and Regional Economics, Working Paper 40, Competitive Neutrality Between Road and Rail, 19997 Bureau of Transport and Regional Economics, Working Paper 57, Land Transport Infrastructure Pricing, 2003
34 < AUSTRALASIAN RAILWAY ASSOCIATION
chapter 3: constraints preventing rail reaching its natural economic potential
EXHIBIT 19: IMPACT OF CHANGES TO THE CURRENT COST ALLOCATION METHODOLOGY$ per '000 ntk
*Net charge for the difference in Road/Rail externality costs based on the mid-range of a number of studies on externality costs for rural areas
Source: NRTC Technical Paper: Updating Heavy Vehicle Charges, September 1998; BTRE working paper 40,1999; Queensland Government Rail Studies
NRTCtoday
NRTC(with externalities)
NTRC(externalities plus PCU)
BTRE
Var OpexFixed Opex
MPM
CapexExternalities*
Higher proportion of costs treated as separable and allocated by mass-distance measures
As current NRTC allocation but non-separable expenditure allocated by PCUsrather than VKTs
Implied >100% increase
9.1
15.117.4
21.1
2.3 2.3 2.5 3.80.7 0.7 1.312.2 2.2 2.5
3.83.9 3.95.1
6.566
6
EXHIBIT 20: ACCESS REGIME COMPARISON—ROAD VERSUS RAIL
Road Rail
"Instead of separately costing past efforts to construct roads and future maintenance requirements, it is assumed that current expenditure provides a reasonable proxy for annualised costs of providing and maintaining roads for the current vehicle fleet.
This approach is known as the PAYGO, or pay-
targets"
NRTC, "Updating Heavy Vehicle Charges",
September 1998
"The Ceiling Limit means the Charges which, if applied to all operators of a Segment or a group of Segments would generate revenue for ARTC sufficient to cover the Economic Cost of that Segment or group of Segments"
Economic Costs include:
- Segment specific costs and an allocation of non-segment specific costs
- Depreciation of segment specific and non-specific assets
- A return on segment specific and non-specific assets based on DORC (revalued every five years)
ARTC Access Undertaking, May 2002
as-you-go, approach to setting cost-allocation
THE FUTURE FOR FREIGHT > 35
chapter 3: constraints preventing rail reaching its natural economic potential
Percent of total
Variable Op. Cost
Fixed Op. Cost
Depreciation / sustaining capital
Return on sunk capital
EXHIBIT 21: COMPARISON OF ACCESS PRICING POLICY REGIMES*
Typical Industries Rail Road
Coal Intermodal
Regulated price cap
12 17 17 19
23 19 19 13
2713 13 17
3851 51 x
Electricity,Gas,Telecommunications
*Estimate for ‘Typical Industries’ taken from the IPART determination for the Regulation of NSW Electricity Distribution Networks (pg 77); Coal assumed to have same distribution by cost component as Intermodal; Road sunk capital assumed to be same proportion of total access fee as for Intermodal
Road not required to pay
Can be much larger if significant 'growth' capital expenditure is also being undertaken
The solution to these problems is to use mass distance
charges instead of either fuel-based or registration charges.
This would further rebalance the bias against rail as the
latter competes against the heaviest trucks travelling
the longest distances. Currently, a number of European
countries are in the process of introducing mass distance
charging. Switzerland introduced a mass distance charge in
2001, and Germany and the UK are both planning truck-
km charges that will take account of vehicle environmental
and road damage characteristics.8 In considering the need
for mass distance charging it is important to recognise
that fuel-based charging is not the same as mass distance
charging because fuel consumption does not vary linearly
with increasing mass.
3.2(iii) No accounting for ‘externalities’ is a third
methodological flaw
Externalities, discussed in Chapter 2 (Exhibit 13) are not
included in either rail or road access costing policies. This
provides a further systematic bias in favour of trucks as
road transport typically has larger safety, environmental,
and congestion impacts compared to rail.
3.2(iv) The magnitude of heavy vehicle undercharging
is significant
While the existence of undercharging for heavy vehicle
access is becoming broadly accepted, the exact magnitude
is still subject to debate and beyond the scope of this
report. However, as discussed above, even on the more
conservative revision, the size of the current pricing
distortion arising from current public policy is large.
As shown earlier in Exhibit 19, simply moving from NRTC’s
current costing methodologies to BTRE’s approach, and
by including a mid-point estimate from the range of
externalities costs, heavy vehicle access charges should be
twice their current levels. This is before taking into account
some of the further increases that could be expected by
application of the emerging engineering-based costing and
more appropriate charging mechanisms such as mass
distance charging (which is in effect what is used for rail).
3.3 Inconsistent access charging policies makes
above rail investment unacceptably risky
The principles underpinning regulated access pricing
ceilings between road and rail are fundamentally different,
to the detriment of rail (Exhibit 20).
In essence, the access charging principles for road do
not seek to achieve a return on past investment, rather
they seek only to cover current road capital and operating
expenditures. In the case of rail, by contrast, access
prices can reach a ceiling that covers a return on all past
investments that are revalued every five years to achieve
a Depreciated Optimised Replacement Cost, or DORC
valuation. The effect is that rail access charges can be set
to recover more than road access charges, (although much
depends on how much new investment is being undertaken
relative to past ‘sunk’ levels). While Exhibit 21 illustrates
these differences, the obvious question is why would two
competing industries be subjected to different access
charging principles?
Inter-capital rail freight prices access at levels allowed
by competition from road and so, particularly given the
cross subsidies received by inter-capital road freight as
just discussed in 3.2 above, on inter-capital routes rail
access fees never reach ceiling levels. This creates a
major problem. Under the current regulatory conditions
the below rail access providers could significantly increase
access fees over time. Their only limit on doing so is the
current low profitability of rail operators. Access pricing
has been held down by the ARTC to assist the rail industry
to gain market share, even though the ACCC indicate that
these revenue levels may not sustain the infrastructure. In
theory at least, present rail access prices can, in ARTC’s
jurisdiction, be more than doubled within the floor/ceiling
limits established by the DORC method in the ACCC
Access Undertaking.
Exhibit 22 shows that access fees could double versus
current levels and still be within the regulatory ceiling.
Such a price movement would create even more distortion
between road and rail pricing.
36 < AUSTRALASIAN RAILWAY ASSOCIATION
chapter 3: constraints preventing rail reaching its natural economic potential
8 Perkins S., Recent developments in road pricing policies in Western Europe, 2002
This issue can inhibit investment by commercial rail
operators. The current rail access regime can, in its effect,
extract all additional profit from new rail investment and
leave rail operators with profits only at stay-in-business
levels. Whatever the stance or intentions of track owners,
the train operators will not be able to risk large investments,
and take business risks in freight markets, while access
fees can rise to take the incremental profits they create
from improved service or increased investment.
3.4 Absence of a consistent approach to the assessment
of road and rail capital funding
Currently there are three problems with comparative
road and rail funding, which bias heavily to road at the
expense of rail. This bias further tilts the ‘playing field’ to
road, and damages efficient resource allocation within the
Australian economy.
> The most fundamental problem is, as just discussed,
that artifi cially low road user charges in inter-capital
road freight ‘cap’ the access fees that rail can charge.
With rail access fees below cost recovery, new rail
investment, and even major periodic maintenance,
are unprofi table.
THE FUTURE FOR FREIGHT > 37
chapter 3: constraints preventing rail reaching its natural economic potential
EXHIBIT 22: RAIL’S REGULATORY REGIME ALLOWS FOR LARGE ACCESS PRICE INCREASES
100%
Energy Australia
AGL TelstraLocalCalls
50 50
Coal ARTC RIC
Typical regulatory regimes for separated infrastructure
Elect-ricity Gas
Tele-phony Rail
Current access price $5–7/'000 ntk
Allowableaccess price $10-12/'000 ntk
Permissibleincrease in price
~$100m pa*
Implications for rail operatorsRegulatedceiling price
Intermodal
2x
Percent of total defined costs
*Calculated as $6/'000ntk increase across current intermodal rail task of ~16bn ntk
> Secondly, markedly different assessment criteria are
used in making road and rail investment decisions.
This difference was summarised neatly by the
Commonwealth Government in its 2002 Transport
Green Paper:
“This issue is compounded by different assessment
criteria for road and rail infrastructure investment.
Rail infrastructure projects are commonly appraised
on financial rather than economic cost-benefit
criteria. Financial analysis presents higher hurdles
than economic analysis by excluding benefits for
organisations or groups and only considering those
for the investor. Financial analysis also has to take
account of corporate taxation and does not include
consumers’ surplus gains, which can make an
important difference for large lumpy investments.” 9
The Government has said in the 2004 AusLink White
Paper that it intends to address this issue:
“A new project assessment methodology will be
progressively introduced to ensure neutrality between
transport modes, proponents and construction and
non-construction solutions, in assessing the broad
range of potential projects.” 10
All else equal, then, there will be less investment in rail
than in road when funding proposals with equal merit
are considered.
The States and the Commonwealth have focussed
heavily on road investment, but not rail investment.
This is illustrated by programmes for National
Highways, Roads of National Importance, Roads to
Recovery and Black Spots. In contrast, for example, in
the Commonwealth’s own words:
“… Commonwealth engagement [on rail funding] has
been ad hoc and intermittent compared to its focus on
the road system”. 11
Further, the Government has also recently stated that:
“Rail infrastructure investment has been largely ad
hoc. The arrangements for the planning and funding
of rail network infrastructure reflect, in large part, the
origin of the rail network in separate State-based rail
systems. These have been independently run and
managed with funding decisions historically driven by
local needs. The overall amount of funding available for
rail infrastructure has also been severely limited.”12
> Thirdly, added to these problems has been rail’s history
of poor performance, particularly as inter-capital
rail was run by separate State-based and focussed
entities. When added to inadequate user charges for
heavy road vehicles, rail infrastructure in the past has
seemed a poor investment.
The effect of this record of poor investment is
profound. Current track quality, particularly on the
North South, is very poor resulting in slow travel times,
trains not available when they are needed and not
reliably meeting timetables, and track and terminal
congestion which also limits overall capacity.
3.5 Negative impact of structural separation has not been
properly overcome through alternate vertical co-ordination
mechanisms
As has been recognised by numerous studies, rail differs
from other network industries (electricity, gas and water) in
a number of important ways. For example:
> The nature of the close physical interaction between
the network (rails) and the network users (trains). For
example, rolling stock design, maintenance and day-to-
day operation affect track maintenance and operation,
and vice-versa. Decisions by half of the rail equation
can have a large impact on the performance and costs
of the other, resulting in overall higher costs.
38 < AUSTRALASIAN RAILWAY ASSOCIATION
chapter 3: constraints preventing rail reaching its natural economic potential
9 Department of Transport and Regional Services, AusLink Green Paper, 2002, p.2710 Department of Transport and Regional Services, AusLink White Paper, 2004, p.2511 Department of Transport and Regional Services, AusLink Green Paper, 2002, p.712 Department of Transport and Regional Services, AusLink White Paper, 2004, p.13
> The need for access to particular paths through the
network at particular times. Rail traffi c must travel
between origin-destination pairs to meet specifi c
delivery windows. This is in contrast to, say, electricity,
gas and water where the product is homogenous and
so the path of particular electrons, molecules, etc
through the network is irrelevant to producers and
customers alike.
Managing these two factors requires levels of coordination
between the track and train owners that are not necessary
in most other network industries. The physical interaction
of train and track requires that, for optimal performance
of the system as a whole, both ‘above rail’ and ‘below rail’
choose investments, maintenance activities and operational
practices that may be considered suboptimal from the point
of view of either enterprise. Differing demand for specific
paths between nodes in the network places additional
constraints on capacity, which require coordinated
investments in track, terminals and rolling stock, and
optimisation of train scheduling if the infrastructure is to be
used efficiently.
Industry participants recognise there is a need for much
improvement in the way above and below rail operators
interact with each other to optimise the industries
performance. Four types of ‘vertical market failure’ have
been identified. These are summarised in Exhibit 23, and
some examples include:
> Operational links. Signifi cant cost trade-offs and
burdens can be placed on either the track owner or the
train operator by decisions made in a number of areas.
For example:
- Rail grinding of track versus wheel profi les. The profi le
determined for wheel profi les can have a signifi cant
impact upon the maintenance cost of the train
operator. Conversely the rail head grind profi le of the
track has a signifi cant impact upon the above rail
wheel profi le life. If wheel profi les are not in alignment
with the track grind profi le signifi cant maintenance
cost is applied to the track owner and rail operator.
- Train speeds versus axle load. The effect upon train
operators of standard track speeds for capacity
utilisation may have a signifi cant impact upon fuel
usage, staff rosters and terminal down time. The
choices made have signifi cant impacts upon above
rail costs and below rail costs.
- Train control. In a vertically separated rail industry,
functions are necessarily duplicated in both above
rail and below rail organisations. An example is train
control (tracking), where both the track manager and
train owner will need to monitor train movements
through the system. There is considerable cost
associated with duplicating such functions and their
supporting infrastructure. Additionally, if information
is not shared between the two systems in a timely or
accurate fashion, then the performance of the rail
system as a whole will be impaired.
> Critical investment decisions. For effi cient rail,
synchronised and complementary investments in track,
terminals and rolling stock must be made (for example,
investments in longer trains require parallel investments
in longer passing loops). With large amounts of capital
at stake, any uncertainty regarding the likelihood of
the complementary investment taking place can cause
both sides to delay their plans, sometimes indefi nitely.
> Risk management. Accidents are a major issue for both
above and below rail and both sides sustain damage
and loss as a result of the same incident. To address
these issues through processes other than litigation
and with greater simplicity would assist both. To
objectively determine major risk mitigation collectively
would also be benefi cial to each. Occupational Health
and Safety are also issues in which the action or
inaction of the above and below rail operator can have
risk consequences for the other. This is an area in
which closer collaboration and benefi t and disbenefi t
assessment for resolution would assist each other.
THE FUTURE FOR FREIGHT > 39
chapter 3: constraints preventing rail reaching its natural economic potential
40 < AUSTRALASIAN RAILWAY ASSOCIATION
chapter 3: constraints preventing rail reaching its natural economic potential
EXHIBIT 23: FOUR TYPES OF 'VERTICAL MARKET FAILURE' NEED TO BE ADDRESSED
Description Example
Operational - Operational activities or decision making, including operational negotiation between above and below rail operators. Also extends to safety
- Optimising wheel maintenance and rail track grinding to minimise ‘through-chain’ costs
- Optimising train speeds and track utilisation to maximise through-chainprofits and safety
- Minimising time consuming/difficult negotiations over timetabling and access arrangements
Capital Investment
- Investment delayed waiting both above and below rail sign-off as co-ordinated investment is required
- Investment in passing loops
- Customer specific sidings
Risk Management
- Difficulty of installing track monitoring equipment for benefit of above rail
- Equipment to monitor condition of bearings would help train operators, and reduce derailments
Marketing - Customer acquisition and retention initiatives requiring coordination between above and below rail
- Better control over customer offering and customer service
- New customer introductory rates to attract them away from road (modal shift requires significant customer investment)
> Offering to customers. Winning new customers to
rail requires service (cut-off and transit times) and
reliability undertakings that often require effort from
both the above and below rail operator to be fulfi lled.
For example, co-ordinated introductory pricing offers
may be necessary to entice current road users to make
the investments necessary to convert their freight
operations to rail (e.g. container purchases, new
warehouse facilities, and so on). Unless both above and
below rail operators co-ordinate to provide these types
of offers, there will be continued under investment in
‘customer acquisition’. This is particularly true when the
customers are entrenched road users.
A further example of this is the prioritisation of trains
through congested parts of the network. If scheduling
confl icts are resolved by reference to simple contract
rules rather than taking into consideration the nature
of the freight involved, low value services may be given
priority over higher value services (e.g. containerised
freight versus steel) causing delays to the freight where
speed matters most to the end users.
Chapter 4—significant benefits will flow from lower cost rail growing modal share |Ch.4>
However, before embarking on industry reform it is
necessary to know whether the benefit is worth the effort
(discussed in Chapter 5) necessarily involved.
To assess the benefits of rail industry reform we have
modelled the effects of some specific changes that reflect
the direction that needs to be taken. These changes will
not only ‘unmask’ the true economic signals, as discussed
in Chapter 3, but will allow rail to increase its modal
share significantly.
The model calculates the ‘direct benefits’ from rail reform
as the cost advantage rail achieves multiplied by the extra
tonnage the reforms will encourage to be carried by rail.
These direct benefits have then been used by Access
Economics to determine the broader economic benefits
of change.
This chapter addresses in turn:
> The specifi c changes that have been modelled
> The expected increase in rail’s modal share that should
be anticipated
> The resultant economic gains that can be expected.
4.1 Changes that have been modelled
Four specifi c changes have been modelled, refl ecting the
required policy changes described in detail in Chapter 5.
Together these changes provide the necessary ‘shock’ to
the current industry dynamics that drive a fundamentally
different outcome.
4.1(i) Efficient costs on the North South corridor
The modelling assumes that the ARTC makes the savings it
has foreshadowed and that were discussed in section 2.4.
This is fundamental, for two reasons:
> It is vital for the ARTC’s future viability, as the ARTC
will not receive the level of subsidies that the NSW
Government previously provided to offset the high cost
of NSW infrastructure maintenance. If the ARTC misses
its planned savings targets, and as a publicly owned
organisation they could be diffi cult to achieve, then it
may be forced to raise access fees to compensate. Any
such increase will offset the benefi ts of rail reform and
frustrate the outcomes captured in this modelling.
> Without these cost savings rail has no material cost
advantage over road on the North South corridors,
as was discussed in section 3.1. That is, there is no
compelling case for rail over road on these corridors
without these available cost reductions being achieved.
The ARTC has, therefore, a critically important role to play
in rail reform.
THE FUTURE FOR FREIGHT > 45
chapter 4: significant benefits will flow from lower cost rail growing modal share
The previous chapter identified the constraints holding back rail from achieving its natural equilibrium position within the broader freight transport system. While not wanting to trivialise their importance, it is clear that these constraints can be relatively easily dealt with through industry reform.
46 < AUSTRALASIAN RAILWAY ASSOCIATION
chapter 4: significant benefits will flow from lower cost rail growing modal share
95%
95%
95%
95%
95%
95%
75%
75%
75%
80%
80%
80%
99%
99%
99%
99%
99%
99%
60%
75%
85%
75%
85%
95%
15
11
9
43
55
17
10.5
29
12
57
69
Transit time
EXHIBIT 24: IMPACT OF ARTC TRACK INVESTMENT ON SERVICE CHARACTERISTICS
15
11
33
9
43
55
21
13.5
36
13
58
72
Syd-Bne
Mel-Syd
Mel-Bne
Mel-Adl
Mel-Per
Syd-Per
* Percent of services arriving within 15mins of scheduled time**Extent to which mode offers services at times the market demands
Source: ARTC Track Audit 2001, undertaken by Booz Allen; Memorandum on ARTC lease of NSW track, June 2004
95%
95%
95%
95%
95%
95%
50%
55%
45%
74%
66%
70%
99%
99%
99%
99%
99%
99%
25%
50%
60%
70%
80%
83%
Hrs
Reliability*
%
Availability**
%
Transit time
Hrs
Reliability*
%
Availability**
%
Now After ARTC investment in track improvements
RoadRailRail improvement
4.1(ii) Improved service performance, particularly on the
North South
Rail service performance levels on the North South are a
major constraint to rail achieving greater modal shares. Put
simply, despite its discounted price to road, rail’s service
performance was insufficient to continue to attract more
customers. On each of the three service measures (transit
time, reliability and availability) rail was found to be a
clear underperformer.
The reason for this underperformance lay largely with
the under maintained NSW section of the North South
track and inherent problems such as passage through
the Sydney metropolitan commuter bottleneck. Analysis
undertaken in the course of this work and drawing on the
ARTC track audit of 2001 and discussion with rail operators
pointed to a conclusion that in the order of $1 billion of
investment was needed on the North South network. This
was about double the investment planned by ARTC under
their emerging lease arrangements with NSW Government
under negotiations at the time.
The modelling assumed a $1 billion investment in track
targeting performance improvements on individual corridors
(Exhibit 24). This performance improvement, coupled with
other changes below, would form the basis for modal share
growth assumptions. After this analysis was complete the
AusLink funding announcements were made, providing
the remaining investment funding to meet the $1 billion
program we had identified.
4.1(iii) Removal of access pricing uncertainty
to provide investment certainty
As discussed in section 3.3, and related to the issue
discussed in 4.1.(ii) above, the differences between
current road and rail access pricing policies means that
the proposed track investments could be undermined by
increased access charges. There are, of course, two ways
to practically remove this uncertainty:
> Road and rail access regimes could be aligned so that
the same defi nition of full cost recovery is used for
both (in addition to correcting the current road cost
allocation and recovery methodologies). With road
pricing also set to include a return on sunk capital,
rail access fees can then increase to their maximum
(ceiling levels) with no scope for further increases as
above rail operators increase their profi ts.
> Alternatively, access fees can be capped via long term
(say, 15 year) access agreements so that above rail
investors can invest with certainty.
For the reasons listed in Section 4.1(i) above, we have
assumed the latter in our modelling. It is the simplest way
to address what is a serious and immediate problem for the
rail industry.
4.1(iv) Achieving improved vertical co-ordination
Rail, particularly on the North South corridors, cannot afford
to carry any cost inefficiency. This is a major issue given
the need for co-ordinated above and below rail investment,
and the operational effectiveness and reliability required to
restore rail's modal share.
The modelling has assumed, therefore, that this issue is
in part addressed. The magnitude of cost improvement
assumed by improved vertical coordination (discussed
in section 2.5) was small relative to its overall potential;
specifically applying a 10 percent efficiency to both above
and below rail operating costs. For our purposes the
modelling did not need to assume how this was achieved
and where specifically it would impact on the cost structure.
THE FUTURE FOR FREIGHT > 47
chapter 4: significant benefits will flow from lower cost rail growing modal share
4.1(v) Improved customer service through above
rail innovations
Rail operators are realising the scope they have to tailor
particular service and price offerings to their customer’s
individual circumstances. For example, they can provide
improved container tracking to minimise lost or delayed
containers; they can offer higher priced and faster services
for some, and lower priced and slower services for others,
rather than the current ‘vanilla’ offering; or they can provide
customised rolling stock for specific customer segments,
such as automobile manufacturers. Such initiatives,
of course, require significant investments to be made,
including in rolling stock.
Traditionally, pricing and service offering has not properly
segmented the markets compared to other industries such
as airlines.
The modelling assumes that rail operators do significantly
improve their customer service relative to past performance.
Indeed, rail freight operators are increasingly doing so.
Much will, of course, depend on all the other changes
being made that have been discussed above. This does
not impact operating costs directly but rather recognises
that service offerings will need to be more sophisticated to
achieve the modal shares anticipated.
4.2 Expected modal shares
It is necessary to estimate the likelihood of additional rail
volumes on a corridor-by-corridor basis. This is particularly
important because of the significant differences in the road
versus efficient rail cost differentials along specific corridors.
It is estimated that, by the end of the next decade, some 16
billion net tonne kilometres of additional freight task should
be carried by the inter-capital city rail network as a result of
these changes compared to a “business-as-usual” scenario.
Underpinning these estimates are three key issues and
projections:
> The overall growth in the total inter-capital city freight
task (i.e., underlying market growth)
> The likely rail volumes without industry reform
(i.e., ‘business as usual’)
> The share of the overall inter-capital rail freight growth
that should be carried by rail with these changes
(i.e., modal share after rail reform).
4.2 (i) Expected growth in the total inter-capital city
freight task
In broad terms this analysis has adopted the corridor-by-
corridor market growth estimates of the BTRE which are
in line with those of other forecasting bodies. Overall, it is
projected that the total inter-capital city freight market will
grow on average by 4.5% p.a. According to the BTRE:
“Inter-capital non-bulk freight flows have grown faster
than national income. Over the periods 1972 to 1980
and 1980 to 1990, the freight growth rate was on
average 1.3 times the growth rate for the economy as
a whole. Between 1990 and 2000 inter-capital freight
grew at 1.5 times the growth rate of the economy.” 13
The 4.5% overall freight market growth can either come
from assuming Australia’s economic growth will be 3%,
which is lower than recent trends, and assuming that the
freight task will grow at 1.5 times; or assuming Australia’s
growth will be 3.4%, which is more reflective of recent
trends, and assuming that the freight task grows at 1.3
times that of the economy.
4.2(ii) ‘Business as usual’ modal share
To determine the likely rail volumes on a business-as-usual
basis we used the BTRE estimates for future rail volumes as
a starting point. These are shown in Exhibit 25.
An examination of these estimates, however, revealed at
least two concerns:
> Firstly, these trends did not accord with recently
slowing of growth of rail volumes on, particularly, the
East West and Melbourne Brisbane corridors. The
formation of National Rail in the early 1990s appears
to have driven volume growth in these corridors, but
based on the most recent privately available data, the
impetus appears to have slowed signifi cantly.
48 < AUSTRALASIAN RAILWAY ASSOCIATION
chapter 4: significant benefits will flow from lower cost rail growing modal share
13 BTRE Information Sheet 22: Freight between Australian Capital Cities
> Secondly, rail operators are expressing a range of
growing concerns that only policy changes can
address. These have been discussed in Chapter 3.
Despite the above questions about the BTRE’s forecasts,
these forecasts have been adopted as base case
volumes, which will occur under business-as-usual.
Furthermore, where BTRE had declining volumes on the
short Melbourne-Sydney, Sydney–Brisbane and Adelaide-
Melbourne corridors, these volumes have been held
constant in the base case. The adopted base case volume
assumptions can therefore be regarded as being highly
conservative as they in-effect over estimate the volumes
we judge can reasonably be expected to be carried
by rail, thereby reducing the measured benefits of the
changes recommended in this report. This approach is
consistent with the conservative approach underpinning the
conclusions of this report.
Given the importance of this starting point volume
assumption, sensitivities are conducted in section 4.3 below
using an alternate base case (halving BTRE’s assumed
volume growth on all corridors) starting point.
THE FUTURE FOR FREIGHT > 49
chapter 4: significant benefits will flow from lower cost rail growing modal share
EXHIBIT 25: BTRE FORECAST RAIL VOLUMES BY CORRIDOR
Source: BTRE Information sheet 22: Freight between Australian capital cities
1972 1980 1990 2000 2010 2014 1972 1980 1990 2000 2010 2014
North-South Corridors
Mel-BneSyd-BneMel-Syd
Mel-Adl
Eastern capitals-Perth
East-West Corridors
Kilotonnes Kilotonnes
Formation of National Rail
Formation of National Rail
Forecast Forecast
0
500
1000
1500
2000
2500
3000
3500
4000
4500
0
500
1000
1500
2000
2500
3000
3500
4000
4.2(iii) Modal share possible with reform
The changes assumed in this modelling are dramatic
compared to business-as-usual, and are forecast to result
in rail volumes doubling over 10 years. This is justified by a
combination of improved service and price.
To maintain its current share rail already offers significant
price discounts over road. On the East West and Melbourne
Brisbane corridors rail offers price discounts around 30%,
while on the shorter haul particularly North South routes the
price discounts are around 20%.
These discounts, in effect, compensate for rail’s less flexible
service levels. On the East West corridors service levels are
closer to those provided by road, and high modal shares
are achieved (discounts are also large). On the North
South corridors service levels, and so modal shares, are
disappointingly low.
The ‘shock’ to the status quo, therefore, is the combination
of service level improvements from the $1 billion investment
in rail, service innovations by operators and lower below rail
operating costs (discussed in section 4.1); which in turn
drives unit cost reductions as modal shift drives increasing
economies of scale; working together in a ‘virtuous circle’.
As seen Exhibit 24 above, service levels on all corridors
increase, with the largest improvements being seen on the
East Coast. Our modal share projections therefore reflect
the fact that larger shifts in service levels can be expected
to drive larger changes in modal share, as well as the
relationship between rail’s discount to road prices and
modal share. It can be seen in Exhibit 26 that the projected
modal share increases and cost reductions maintain this
relationship. It is important to note that the North South
modal shares assumed in this analysis are consistent with
the operating plans of key industry operators such as ARTC
and Pacific National.
4.2(iv) Overall Volume
Overall it is estimated that the combination of market
growth and modal share gains will drive an additional 14
billion ntk being carried on rail compared to a business-as-
usual outcome (Exhibit 27). Exhibit 28 shows significant
share gains on all corridors, particularly the Melbourne
Brisbane corridor.
Comparing the business-as-usual volume base used in this
report with the projections made by the BTRE it can be
seen that our assumptions lead to a slightly higher base
growth assumption than is contained in the BTRE analysis.
This is shown in Exhibit 29 across all corridors. Relative to
today’s rail freight volume, it can be seen that 75% of the
additional ntk calculated in this report are additional to the
business-as-usual projections used.
Finally, some international comparisons can be useful. The
USA has previously undergone major rail reform, and saw
significant modal share gains as a result (see Exhibit 30).
Prior to these reforms, rail tariffs were set by the Interstate
Commerce Commission, unprofitable lines could not be
closed, companies could not secure long term contracts
and invest with confidence, and price discrimination
between customers was prohibited. The 1980 Staggers Act
removed these constraints and allowed rail to reorganise
itself to compete with road, focussing on longer corridors
and pricing to get volume. Consolidation saw 63 companies
become 7, with the survivors able to exploit scale
economies on the remaining network.
The modal share shifts determined in this analysis
represent a major change from past trends. They will clearly
not occur on a ‘business-as-usual’ basis.
4.3 Shifting to a lower cost transport mode
drives significant economic benefits
When rail reaches efficient costs on the North South
corridor, and grows its volumes and modal share,
significant economic benefits are available. This can be
illustrated in three ways.
> The direct benefi ts from the projections and changes
discussed in this report
> The wider ‘2nd order’ benefi ts that fl ow to the economy
> The conservative nature of the proposed changes
discussed in this report.
50 < AUSTRALASIAN RAILWAY ASSOCIATION
chapter 4: significant benefits will flow from lower cost rail growing modal share
THE FUTURE FOR FREIGHT > 51
chapter 4: significant benefits will flow from lower cost rail growing modal share
EXHIBIT 26: RELATIONSHIP BETWEEN PRICE DISCOUNTS AND MARKET SHARE
Source: BAH/ARTC Track audit; PJPL analysis
Rail's modal share(Percent)
CurrentAfter 10 years
*(road price-rail price)/ road price
Current price differential market share relationship
North
Impact of cost reductions resulting from rail reform
Rail's modal share(Percent)
Discount to road* Discount to road*
SB SM
MB
MA
AP
MP
SP
0%
10%
20%
30%
40%
50%
60%
70%70%
0%
10%
20%
30%
40%
50%
60%
0 10 20 30 40 50 60 70 80 90 1000 10 20 30 40 50 60 70 80 90 100
North-South
East-West
Impact of cost reductions resulting from rail reform
Discount to road* Discount to road*
SB SM
MB MA
AP
MP
SP
Source: BAH; PJPL Analysis
EXHIBIT 27: THE VOLUME SHIFT TO RAIL—RAIL REFORM VERSUS BUSINESS AS USUAL
Additional volume carried by rail= 14.0b
By corridor
2.4
*Business as usual assumptions: BTRE forecast growth rates on the long corridors, constant volume on the short corridors
Additional volume carried by rail= 14.0b ntk by 2014
2.4
7.6
36.3
2.9 2.7 2.6
9.8
4.0
6.7
MB SB MS MA MP AP SP
1.81.0 1.4 1.0
5.5
2.23.5
0.9
2.4
1.0
1.5
4.8
1.9 1.3 1.6
1.9
0.8
1.7
16.5
5.8
14.0
Total
Currentdemand
Business as usual*
Impact ofRail
Reform
52 < AUSTRALASIAN RAILWAY ASSOCIATION
chapter 4: significant benefits will flow from lower cost rail growing modal share
EXHIBIT 28: GROWTH IN FREIGHT TASKS—ROAD AND RAIL
*4.5% per annum until 2014Source: BAH; PJPL Analysis
CurrentGrowth
Rail's modal share under Rail Reform
Percent
Current
50 Based on:
Current taskBillion ntk
Future task (in 10 years)
Billion ntk
Underlying freight growth*
Billion ntk
80
M-B S-B
M-S
M-A
M-P A-P
S-P
M-B S-B
M-S
M-A
M-P A-P
S-P
M-B S-B
M-S
M-A
M-P A-P
S-P
M-B S-B
M-S
M-A
M-P A-P
S-P
0.9
2.4
1.01.5
1.81.0
1.41.0
5.5
2.2
3.5
7.6
2.9 2.7 2.6
9.8
4.0
6.7
53
2717
40
80 80 80
20 14 14 23
70 70 65
Current
0-5yrs
Share gain
5-10yrs
- BAH analysis for ARTC- Total cost comparison
by corridor- Modal share modelling - Company projections
EXHIBIT 29: COMPARISON OF PJPL AND BTRE FORECAST RAIL VOLUMES ACROSS ALL CORRIDORS
%
Source: BTRE Information sheet 22: Freight between Australian capital cities
ntk
Rail reform volume
BTRE volume
PJPL(business-as-usual)
0
5
10
15
20
25
30
35
40
1972 1980 1990 2000 2010 2014
4.4%
2.7%
7.4%
2.8%
% = CAGR
THE FUTURE FOR FREIGHT > 53
chapter 4: significant benefits will flow from lower cost rail growing modal share
EXHIBIT 30: TRENDS IN US RAIL PRODUCTIVITY: 1964–PRESENTRail KPIs change relative to 1981 base (100)
Source: Railroad Facts, ARR, Cambridge Systematics
0
50
100
150
200
250
300
1964 1970 1980 1990 2000
Price
Revenue
Volume
Productivity
2010: 4 Class I Rail companies
1976: 63 Class I Rail companies
1997: 7 Class I Rail companies
’60s and ’70s: ROI for Rail co ’s falls to 2.5%
1980: Staggers Act replaces Interstate Commerce Act which prevented:- pricing freedom- horizontal mergers- divestment of assets- confidential contracts
By 1976: 1/3 of Rail co ’sbankrupt; maintenance deferred on low density lines; network deteriorates
EXHIBIT 31: DIRECT SAVINGS FROM IMPROVING RAIL'S COSTS AND MODAL SHAREToday
32
1826
EW NS Average
*After RIC cost reductions, volume increases and cost reductions from improved vertical coordination and productivity improvements
Source: BTRE; ARTC; Pacific National; Port Jackson Partners analysis
28
7
18
EW NS Average
Rail's cost advantage over Road
$ per '000 ntkToday Estimated*
72%
32%
50%
EW NS Average
59%
16%
35%
EW NS Average
Rail's share of freight task
PercentToday Estimated*
Outcome for the economy
Benefit from modal shift:
Total cost savings = $26/'000 ntk
Volume shifted to rail = 14.0b ntk
Average annual benefit = $370m
Value created = $5.2b NPV
Benefit on existing volumes:
Incremental cost savings = $8/'000 ntk
Existing task = 16.5b ntk
Average annual benefit = $127m
Value created = $1.8b NPV
Total value created = $7.0b NPV
GDP benefit pa $1,213m
GDP benefit NPV $27b NPV
Effect of changes discussed in this report
4.3(i) The gains to users from the projections
and changes discussed in this report
Chapter 2 showed that when the inter-capital freight
services operate at normally expected levels of efficiency,
then rail is a significantly lower cost transport mode on all
rail corridors. Rail becomes 30% cheaper on the North
South corridor, and 50% cheaper on the East West corridor.
This conclusion was based on a ‘bottom up’, corridor-by-
corridor examination of above and below rail and road
operating and capital costs.
The previous section (4.2) described the 14 billion ntk of
rail freight volume increases possible with industry reform.
Moving large volumes of freight from a more expensive
(road) to a cheaper (rail) transport mode therefore results
in direct savings as much as $370m pa.
Exhibit 31 summarises the gains and the savings to be
made. These gains will in part be to the benefit of:
> Transport users; through lower prices (as it is assumed
that productivity gains are passed on to existing users),
as well as new users benefi ting from the shift to the
lower cost mode
> Government; through reduction in effective subsidies
in different forms such as inadequate user charges and
externalities, and
> The rail industry; through the benefi ts of greater
volume (at constant unit margins).
On a net present value basis these direct savings to users
and to Government amount to $7.0 billion. The NPV model
used to calculate the total value created took current
performance (including the current RIC cost structure)
as its Year 0 starting point, then phased in cost savings
assumed for RIC cost reductions and improvements in
vertical coordination over the first 5 years of the forecast.
Unit costs for each year were calculated by applying cost
reductions to the previous year’s figures and then the
impact of volume growth in that year on the fi xed cost base.
This gave a through chain road/rail unit cost differential
for each year of the forecast. Savings were then calculated
by multiplying costs saved by the cumulative growth in rail
volume, relative to the base, for that year.
The NPV calculation assumed a real discount rate of 7%
which likely lies between a public policy and a commercial
discount rate (and is again a conservative assumption in
the context of this analysis).
4.3(ii) The benefits that flow to the wider economy
PJPL commissioned Access Economics to prepare an
independent assessment of the national economic benefits
likely to flow from these cost savings to inter-capital freight.
Their focus was on the net economy wide effects—on
output, consumption and employment—that could be
expected to emerge over the longer term. The report by
Access Economics is attached as Appendix 3.14
Rather than model the year-by-year gains, Access
Economics have modelled the benefits available in the year
of maximum gain; that is in 2014. Their results are shown
in Exhibit 32, which provides the summary results for their
preferred case.
The economy wide benefits from these assumed rail
reforms are very significant. Real GDP is increased by
around $1.2 billion per annum. The benefit to the economy
is around $27 billion, again assuming a conservative
7% real discount rate.
By way of comparison, the recent Council of Australian
Government’s Energy Market Review of December 2002
estimated that the impact of the major policy changes
required in Australia’s electricity and gas markets would
increase real GDP by just under $2 billion per annum by
2010. Many have seen energy as Australia’s main area of
microeconomic reform focus. If energy reform can assume
the mantle of necessary reform, then so should rail reform.
54 < AUSTRALASIAN RAILWAY ASSOCIATION
chapter 4: significant benefits will flow from lower cost rail growing modal share
14 In Appendix 3, section 2, Access Economics refers to rail’s ‘headline’ unit cost advantage over road as being $28/’000 ntk providing an annual benefit of $393m. This is an aggregate number and is consistent with the ‘headline’ numbers presented in the Executive Summary and Chapters 2 and 4 of $26/’000 ntk and $370m pa respectively for additional volume, and $8/’000 ntk and $132m pa respectively for existing volume. In addition, Access Economics nets of the implicit financial subsidy in under-recovery of costs when calculating broader indirect economic benefits.
THE FUTURE FOR FREIGHT > 55
chapter 4: significant benefits will flow from lower cost rail growing modal share
EXHIBIT 32: BROADER IMPACT ON THE ECONOMY (2014 PREFERRED CASE*)$2004 (real dollars)
*Assumes flexible exchange rate, or a fixed trade balance; and a small labour supply response to higher wages (a positive labour supply elasticity with respect to real wages of 0.2)
Source: Access Economics
AnnualBenefit
- GDP increases by $27billion in net present value terms- Consumption increases by $18billion in net present value terms
Increase in GDP, without externality gains $1,070m
Increae in GDP, with externality gains $1,213m
Increase in consumption $802m
Increase in employment 2,480 FTE
EXHIBIT 33: POSSIBLE CONSERVATIVE ASSUMPTIONS
Source: PJPL analysis
Area Treatment NPV impact
Lower ‘Business as Usual’ growth
Evidence to suggest that growth in rail volume on the longer corridors may be slowing-could model with half the BTRE’s forecast growth rate
$1.4b
Larger vertical coordination synergies
Opportunities available from improved industry coordination have only been investigated at the operations level within one operator's activities. Benefits generated by better alignment of strategic goals, investment priorities, etc across the industry are likely to far exceed the current estimates of direct process improvements
~$0.7b
Above road operating costs Have used 38 net tonne B-Double as the base truck-more efficient than the average fleet vehicle so has lower per ntk costs than the average
$0.3b
Above rail operating costs 1,500m trains assumed North South—likely willactually be longer in near future. Also, no benefit assumed for driver only operations
$0.1b
Pick Up and Delivery costs Have not modelled any reduction in PUD costs, although this is likely to occur as commercial operators develop better integrated solutions
$0.5b(20% saving over 5 years)
4.3(iii) The conservative nature of the proposed changes
discussed in this report
The economic measures described in this Report are
necessarily based on assumptions. The purpose of any
modelling is more to indicate the likely broad size of the
gains, rather than to seek a precise estimate.
The cost and volume analysis described above has been
based on a conservative approach taken to some of the
underlying measures and assumptions from the point
of view of rail. These are summarised in Exhibit 33. In
particular, attention has already been drawn to the use of
a 38 tonne B-double as the assumed vehicle for all future
road freight growth, yet the exclusion from the cost gains of
any benefit from longer trains on the North South corridor
or the introduction of driver only train operation. Attention
has also been drawn to the limited analysis that has so far
been undertaken in relation to the potentially large benefits
from improved co-ordination between above and below
rail operators.
Conservative assumptions have also been applied in
defining the ‘business as usual’ volume scenario against
which the incremental benefits from rail reform are
calculated. In forecasting what will happen to rail freight in
the absence of reform, we have firstly adopted the BTRE’s
latest view of rail growth trends (which we judge are likely
to be overly optimistic) and then, where BTRE has forecast
falling absolute volume on the short corridors, we have
held volumes constant. By assuming that rail will achieve
more growth in the absence of reform than is actually likely
to be the case, we have implicitly understated the volume
of rail freight growth that can be attributed to the reforms
described in this report.
Naturally, some conservatism in approach is required
because there will always be areas where the assumptions
made could be seen to have overstated the estimated
$7.0 billion of value.
The key area where this could be the case is the
assumption that there are no further productivity gains
assumed in road freight, while some further rail gains are
factored in (albeit only on the below rail infrastructure
costs). The view taken was that the productivity gains for
road were reasonably exhausted, to the point where now
there are concerns about trucks operating in ways that
already test the bounds of regulation in terms of speed
limits and rest breaks. In any event, the assumption made
in this analysis that B-doubles will be used for all future
road freight, compared to their current approximately
20% share of heavy truck tonne-kilometres, in itself brings
major future productivity gains over the currently used
fleet. Furthermore, it is judged that there remain further
productivity gains in areas such as train and terminal
operations which have been excluded from this analysis.
For example, the benefits of single driver train operations
(currently being implemented but not in the 2003 costs
adopted in this report) have not been included in ‘efficient
rail’ costs.
Overall, different combinations of reasonable assumptions
would still yield an estimate of at least $7.0 billion from
rail reform. Exhibit 34 illustrates that, even if there are
areas where it is judged that some assumptions are more
favourable for rail than they should be, there is considerable
conservatism in the overall approach supporting the
conclusion that the estimated benefits of $7.0 billion are
robust and realistic.
It should be recognised that this analysis relates only to
the inter-capital corridors. The Brisbane Cairns route, for
example, has not been included in the estimated benefits of
rail reform. This would add considerable further benefits to
those estimated.
56 < AUSTRALASIAN RAILWAY ASSOCIATION
chapter 4: significant benefits will flow from lower cost rail growing modal share
THE FUTURE FOR FREIGHT > 57
chapter 4: significant benefits will flow from lower cost rail growing modal share
EXHIBIT 34: SENSITIVITY OF VALUE FORECAST TO MODEL CHARACTERISTICS
1.4
0.7
0.5
0.3
0.1
Halve 'Business as Usual' growth'
Larger vertical coordination synergies
Pick Up and Delivery savings
Above road operating costsbased on most efficient truck
East coast rail operating costsbased on shorter trains
Item
Estimate based on PJPL experience of likely gains from optimising linkages within an industry
Model truck costs based on actual average fleet costs
Changes to model and NPV impacts$ Billions
Model impact of 20% PUD productivity improvement over 5 years
Model impact of 1800m trains on the East Coast and Driver Only Operations
Use only half the BTRE’s projected growth for rail for the next 10 years as the base line
EXHIBIT 35: THREE KEY AREAS OF REFORM
1. Rail needs a level playing field with road transport to ensure efficient choices are made between transport modes and to enable investments to be made with certainty. This requires consistent:
- Access usage charging methodologies
- Capital recovery policy
- Investment decision making criteria
2. Rail industry needs to accelerate internal reforms
- Reduce NSW track costs to efficient levels
- Innovate customer service offering
- Improve vertical coordination
3. A framework for Governments and the rail industry together to pursue a structured and co-ordinated process to achieve the above is required.
Chapter 5—overcoming the constraints holding back inter-capital rail transport |Ch.5>
Chapter 3 discussed a number of unnecessary constraints
being imposed on the rail industry. Chapter 4 established
that there are very significant benefits to be gained in
overcoming these constraints. Rail reform is therefore a
source of major benefit to the Australian economy. This is
not surprising when considering the importance of efficient
transport in a nation with the geographic spread of Australia.
To realise this benefit, the Australian rail industry needs
an integrated reform plan that is comprehensive, and
that closely involves both industry and Governments.
The changes are required to achieve these benefits
(summarised in Exhibit 35) fall under three headings and
are addressed in turn in the remainder of this chapter.
1. Rail needs a level playing fi eld with road transport to
ensure effi cient choices are made between transport modes
and to enable investments to be made with certainty.
2. Rail needs to accelerate its internal industry reforms;
specifi cally:
> ARTC must ensure it quickly captures the expected
operational cost savings by bringing NSW track under
its management
> Above rail operators must overcome their legacy of poor
customer service
> Track owners and train operators must quickly achieve
improved vertical coordination;
3. Governments and the rail industry must together pursue
these actions in a structured and co-ordinated process.
5.1 Providing rail with a level playing field to facilitate
efficient choice and appropriate investment
There is a consensus within industry and Governments
that rail does not compete with road on a level playing
field. Views differ, of course, on the degree of ‘tilt’ but,
as discussed in sections 3.2, 3.3 and 3.4, the cumulative
impact is large. What is needed is policy that ensures
that road/rail usage choices reflect relative economic
attractiveness, and not distortions.
We would recommend a properly constituted national
inquiry, involving both Federal and State policy makers,
be held with the objective of making recommendations
to achieve:
(i) Appropriate costing methodology to remove the current
cross subsidy to heavy vehicle inter-capital movement
to the detriment of rail
(ii) Mechanisms to allow consistent road and rail funding
decisions to be made, recognising the substitutable
nature of road and rail freight
(iii) Common access pricing policy that factors in the need
for consistent expectations of the capital returns on
invested capital.
These issues are linked in that quick moves in (i) reduces
the burden on Government in (iii).
THE FUTURE FOR FREIGHT > 61
chapter 5: overcoming the constraints holding back inter-capital rail transport
While private operators have an important role to play, by their very nature, most of the constraints identified arise largely from a legacy of poor transport public policy.
5.2 Rail needs to accelerate its internal industry reforms
5.2(i) ARTC must ensure it quickly captures the
expected operational cost savings by bringing NSW track
under its management
As discussed in sections 2.4 and 4.1(i), the delivery by
ARTC of its planned efficiencies on the NSW track are
critically important to the success of the rail industry.
While these improvements are judged to be feasible,
achieving them quickly is not trivial, particularly for a
public sector organisation.
While on the surface this can be seen as an ‘internal’
issue to ARTC, reducing operating costs without reducing
maintenance activities (i.e., achieving true productivity
gains), is an issue that all rail operators are absolutely
reliant on. In rail, as in many infrastructure industries,
operating costs (largely maintenance) can be quickly
reduced by reducing maintenance activities below
sustaining levels. Clearly this is not an acceptable option
as its effectively passes costs on to above rail operators
and customers through deteriorating service performance.
Therefore, every effort must be made to ensure nothing
undermines or distracts ARTC from this important element
of a wider industry reform program.
5.2(ii) Above rail operators must overcome legacy
of uneven customer service
Customer interviews reveal that rail freight has a legacy
of poor customer service experiences that needs to be
overcome. Many customers would prefer to use the
cheaper rail transport but feel prevented by service levels
that they perceive do not meet their needs. This situation
has arisen in part because of:
> The previous separate State-based rail freight
operations, with no single point of accountability
to customers
> Public ownership that can be seen to lack the
intensity of commercial focus that can come with
private ownership
> Other factors described in this chapter that have held
back rail.
In general terms rail must improve its service offering in
ways that appeal to specific customers so that its lower
cost structure can be translated into higher volumes. This
could involve making investments in sidings equipment or
in containers, and offering transit time and price/service
arrangements that are better suited to customer needs.
While this area of change will be fundamental, it will
largely depend on all the other changes being made. For
example, the necessary investment will not be made given
the current access fee uncertainty (see section 3.1(ii)
above); and given the current state and operation of the
North-South track as, in particular, offering transit time
undertakings could be counter productive.
5.2(iii) Mechanisms must be found for track owners and
train operators to achieve improved vertical coordination
The first important step in this process is to undertake
a rigorous assessment of the cost of all identifiable
coordination failures. Having done this the causes of
the most important can be identified and a process of
determining the best mechanism to tackle each of them
can then begin. Formation of more ‘general working parties’
is not the right starting point.
5.3 Governments and the rail industry must
together pursue these actions in a structured and
co-ordinated process
To achieve the benefits of rail reform there needs to be
a well structured and co-ordinated process of change
involving both the Government and the rail industry. This is
required for two reasons.
First, piecemeal change will not work. The required
changes in one area often depend on changes in other
areas. This can be seen most easily in areas of above
rail investment, where major decisions must await policy
change to remove access fee uncertainty. It can also be
seen in the link between the need for track investment and
the need for the heaviest road vehicles to pay their true
share of road costs, including externalities. Externalities,
for example, can be factored into access pricing or into
investment decisions.
62 < AUSTRALASIAN RAILWAY ASSOCIATION
chapter 5: overcoming the constraints holding back inter-capital rail transport
Second, Governments need to acknowledge their role
and be fully involved in the change process. The role of
Governments in the policy decisions described above is
obvious. It is also clear, however, that Governments have
assumed for themselves a large role in rail reform as
industry owners, particularly of track through the ARTC,
and also as rail operators in the case of Queensland Rail.
THE FUTURE FOR FREIGHT > 63
chapter 5: overcoming the constraints holding back inter-capital rail transport
EXHIBIT 36: FURTHER REFORM IS TIMELY
Old world Major change New world
> Rail world- state based- all publicly owned
1990s – First direct Federalinvolvement in rail (NR, ARTC)
> Direct responsibility for capital and operations inefficiency for first time
> Poor but rapidly improving understanding of the opportunities
> Road world- Federal
involvement in roads provision
- privately owned operation
- strong and well organisedadvocacy
2000/ – Private involvement2001 (e.g ARG, PN)
> Operators outside of Government
-commercial focus-ability to lobby where required
2002/3 – Federal policy bodies beginning to shift from primarily roads to road/rail focus
> AusLink Green Paper> NRTC, NTC> BTRE
2003 – ARTC/RIC merger announced
> More coordinated approach possible across state boundaries
2004 ––
Auslink released
ARTC takes controlof NSW track
> Rail reform momentum gaining ground
Chapter 6—where to start—immediate actions required to begin the reform process |Ch.6>
In undertaking reform it can sometimes be difficult to know
how to proceed, and particularly how to ensure success
when past efforts to tackle many of the issues raised above
have not been successful. In this regard it seems three
points should be uppermost in the minds of the rail industry.
6.1 The rail industry needs to deepen its knowledge in
key areas so it can actively drive or participate in the key
public policy debates
There is considerable continuing work to be done to
advance all of the issues raised in Chapters 3 and 5.
Specifically the industry should:
> Extend and deepen its detailed knowledge of key
issues. Attempting to address well known problems, no
matter how rational the argument, with an absence of
the relevant facts often leads to inertia
> Work to ensure a proper process is undertaken at the
national policy level to align road and rail access pricing
and government funding principles
> Establish formal and robust coordination mechanisms
between above and below rail operations to remove
the obstacles to reducing operating costs and better
investment decision.
This report is an attempt to make a significant first step
down this path.
6.2 Recognise that there is currently a favourable policy
environment for change
Considerable change has occurred over the last decade or
so that has fundamentally altered the landscape for rail.
Exhibit 36 summarises this change.
In essence, rail has moved from being operated by state-
based, fully publicly owned entities that had their major
focus away from inter-capital freight, to an industry with a
mix of private and public ownership, but nevertheless with
a national freight focus. Private sector involvement obviously
brings a clearer commercial focus and a voice to issues that
may otherwise be left to a different time frame.
The Commonwealth’s involvement in rail is also significant.
It is the level of Government most responsible for, and able
to deal with, inter-capital freight issues, but previously it
had not engaged significantly with rail reform issues. The
recent AusLink papers may have changed this. There is
perhaps no better indication of the urgent need for reform
than provided in the AusLink Green and White Papers:
“Relying on the status quo to address these challenges
is clearly not in Australia’s interest. There is no
‘do-nothing’ option. Incremental change is also
inadequate. Without major change to the planning
framework, the costs of providing an effective national
land transport network will be far higher. The economic
and social importance of the national land transport
network reinforces the need for Australia to undertake
major reform.”15
THE FUTURE FOR FREIGHT > 67
chapter 6: where to start—immediate actions required to begin the reform process
The size of the gains demonstrated in this report suggests that removing the constraints to rail achieving its appropriate place in inter-capital land freight transport deserves the same level of attention and commitment as that directed to similar industries such as energy, another important national infrastructure reform agenda.
15 Department of Transport and Regional Services, AusLink Green Paper, 2002, p.23
“Australia cannot afford poor and uncoordinated
infrastructure decisions that impose high costs on the
community, the economy and the environment... The
existing planning and decision-making framework is
short-term, ad hoc and fragmented across transport
modes and jurisdictional boundaries. The development
and implementation of a national vision for critical land
transport links is vital.”16
6.3 While public policy changes are needed, there
is much the industry can and should do today
While this report has highlighted several important changes
that must occur at the public policy area, it has also
highlighted several that are within the industry’s control.
Of these the most important are:
> ARTC achieving the targeted productivity gains on the
NSW leased track
> Service and ‘product’ innovation by above rail operators
to as rapidly as possible attract volume to rail for line-
haul freight movements
> Establishing appropriate mechanisms to improve
coordination between above and below rail operators on
both operational activities that otherwise impose higher
overall costs, and on investment decision making that
otherwise results in delayed or inappropriate outcomes.
68 < AUSTRALASIAN RAILWAY ASSOCIATION
chapter 6: where to start—immediate actions required to begin the reform process
16 Department of Transport and Regional Services, AusLink White Paper, 2004, p.viii
Appendices—1. Derivation of Road and Rail freight costs;2. Comparing international road costing methodologies and charging regimes;
3. National economic benefits of cost savings on inter-capital rail freight;4. Bibliography; 5. Glossary
|App.>
THE FUTURE FOR FREIGHT > 73
appendix 1: derivation of road and rail freight costs
This appendix discusses in detail the derivation of relative road and rail costs that lead to the conclusion that rail was a lower cost freight transport mode on all inter-capital corridors.
1. Introduction
In order to determine the cost relativity of rail to road on
each corridor, a total cost comparison was performed.
Total cost includes all the direct cost components of above
road and below rail operations, together with the cost
of externalities.
Access fees for both road and rail were not included as part
of the above road/rail operating costs, as these represent a
transfer price within the value chain of each industry, rather
than an actual cost. In effect we have calculated true access
costs and used them rather than current charges.
The total cost comparison has been calculated under
two scenarios:
> Assuming NSW track costs can be reduced to effi cient
levels, together with the benefi ts of improved vertical
coordination and volume growth (see Exhibit A1.1)
> Assuming NSW track costs remain at their current
levels, but the other benefi ts of improved vertical
coordination and volume growth are realised (see
Exhibit A1.2).
These fi gures show that effi cient rail has a cost advantage
over road on all corridors (i.e., when NSW track costs can
be brought to effi cient levels). However, if NSW track costs
cannot be reduced then rail has no material advantage over
road on the North South corridors.
This appendix describes in detail the sources, approaches
and methodologies used to derive through-chain economic
costs for inter-capital road and rail freight. These drivers,
together with volume forecasts, form the inputs to the Net
Present Value model. It addresses in turn:
> ‘Above road’ operating and capital costs
> ‘Above rail’ operating and capital costs
> ‘Below road’ operating costs
> ‘Below rail’ operating costs
> ‘Below road’ capital costs
> ‘Below rail’ capital costs
> Externalities associated with road and rail transport.
74 < AUSTRALASIAN RAILWAY ASSOCIATION
appendix 1: derivation of road and rail freight costs
$ per '000 ntkEXHIBIT A1.1: TOTAL COST COMPARISON—POST RIC COST REDUCTION
Source: PJPL Analysis
Syd - Bris
Melb - Syd
Melb - Bris
Melb - Adel
Adel - Perth
Melb - Perth
Syd - Perth
Road RailCost benefit of Rail
64
26
28
25
21
5
20
28
37
33
32
60
58
66
63
61
57
26
28
25
43
55
38
38
26
28
25
Average**= 25.7
Below
Above
Variable Operating cost*Fixed operating cost*Capital recovery cost Variable operating costFixed operating costPick Up and Delivery (rail)Capital recovery cost Externalities
*Assumes 50% reduction in RIC’s costs and 100% growth in RIC’s intermodal volume
**Across 2014 volume shifted to Rail of 14b ntks
Source: PJPL Analysis
Today = -3
Efficient operator = 17
EXHIBIT A1.2: TOTAL COST COMPARISON—PRE RIC COST REDUCTION $ per '000 ntk
Rail (RIC today)Cost benefit of Rail
Below
Above
Rail (RIC reduced)
Syd - Bris
Melb - Syd
Melb - Bris
Melb - Adel
Adel - Perth
Melb - Perth
Syd - Perth
Variable Operating cost*Fixed operating cost*Capital recovery cost Variable operating costFixed operating costPick Up and Delivery (rail)Capital recovery cost Externalities
43
55
38
38
26
28
25
(2)
(10)
(0)
28
37
33
22
66
69
59
38
26
28
35
*Assumes 50% reduction in RIC’s costs and 100% growth in RIC’s intermodal volume
2. Derivation of ‘Above Road’ operating and capital costs
The ‘above road’ operating costs were based on a 38 net
tonne B-double truck. All unit cost inputs were sourced
directly from a leading road transport company. Utilisation
levels were varied from 10% to 100% (with 100% being
the legal load limit of 38 net tonnes) for each corridor, and
the economics of each individual corridor were modelled at
each utilisation level.
The operating costs can be split into three groups:
> Costs which vary directly with the number of tonnes
transported (e.g., fuel)
> Costs which are independent of tonnes transported
but vary with other factors such as kilometres travelled
(e.g., labour costs, maintenance costs)
> Fixed costs (e.g., insurance, management overheads)
The detailed parameters considered are shown in
Exhibit A1.3.
‘Above road’ capital costs were derived to determine a
charge for both the return of, and return on, assets. Capital
costs comprised depreciation and interest charges, with
straight-line depreciation over a 5 year life (20% residual
value) and 7% interest assumed.
To allow road and rail costs to be evaluated for an
equivalent task, the utilisation levels were related to product
density (since truck and container volumes are not the
same). In the above road model the assumption was that
volume of a 38 net tonne (maximum) B-double is 150m3.
Fully loaded (i.e., both mass and volume limits reached)
implies a freight density of about 250 kg/m3. Utilisation
varies with density: for example, assuming a product has a
density of 100 kg/m3 and the volume of the truck is filled
(150m3), the weight of the freight would be 15 tonnes. This
correlates to a utilisation level of ~40%.
In the case of the B-doubles modelled, operating costs
do not decrease for products with a density greater than
250 kg/m3. This is because no more than 38 tonnes of
freight can be carried (i.e. a full 150 m3 truck load at 250
kg/m3) so no further benefits of increased utilisation can be
achieved once this mass limit is reached.
The reference densities chosen for comparing road and
railcosts was 250 kg/m3—i.e. at the 100% utilisation level
for the B-doubles modelled.
THE FUTURE FOR FREIGHT > 75
appendix 1: derivation of road and rail freight costs
76 < AUSTRALASIAN RAILWAY ASSOCIATION
appendix 1: derivation of road and rail freight costs
EXHIBIT A1.3: DERIVATION OF ABOVE ROAD OPERATING COSTS—B-DOUBLES$ per '000 ntk
RegistrationInsuranceManagement overheadsKms/annumNet tonnes transported
Fuel consumptionFuel costNet tonnes transported
Cost/kmLong distance multiplierNet tonnes transported
Cost/kmNet tonnes transported
Cost/tyreTyres/truckKms/tyreNet tonnes transported
No. hours
On-cost multiplierCost/hour
Pick-ups per tripKm per tripNet tonnes transported
Vehicle Maintenance
Labour
Tyres
PUD
Above road operating cost($/ '000 ntk)
Costs variable with tonnes transported($/ '000 ntk)
Cost variable with kilometres travelled($/ '000 ntk)
Fixed costs($/ '000 ntk)
EXHIBIT A1.4: DERIVATION OF ABOVE RAIL OPERATING COSTS$ per '000 ntk
Above rail operating cost
Cost variable with train kilometres and container numbers
Cost variable with tonnes transported
Fixed costs
Fuel consumption(loco / wagon/ freight)
Fuel cost
Crew
Maintenance
Lifting costs
PUD
Terminals
Operation centre costs
Cost / hr No. hours per tripNo. crew
Cost / km (loco; wagon)
Kilometres per tripCost per lift
No. containers per wagonNo. wagons per train
2 lifts per container
Cost per containerNo. containers per wagonNo. wagons per train
KEY ASSUMPTIONS- 1,500m trains N/S- 1,800m trains E/W- 65t per wagon (max)- 2–4 locos per train (depending
on weight transported)
3. Derivation of ‘Above Rail’ operating and capital costs
The input parameters to the ‘above rail’ operating model
were sourced from Pacific National’s operations, based
on actual 2003 operating conditions and costs in an open
book manner.
The key operating condition assumptions were:
> Train length: 1,500m on North—South Corridor;
1,800m on East—West corridor
> Maximum freight weight per wagon: 65 tonnes,
based on 1 x 20 ft container and 1 x 40 ft container
(equivalent to 100% utilisation)
> Number of containers per wagon: 2 (giving ‘wagon’
volume of 115m3)
> Number of wagons per train: dependent on train length
and number of locomotives required to pull the load
> Number of locomotives: dependent on trailing mass of
the train.
The key cost components and their drivers are shown
in the schematic in Exhibit A1.4. Above rail capital costs
were provided by the NECG based on specific wagon and
locomotive configurations, on a corridor-by-corridor basis.
Capital cost drivers were loco cost (~$4m) and wagon cost
(~$2m). A 20 year life with no residual value was assumed
(due to the absence of an active market for used rolling
stock). The calculations allowed for both a return of, and
return on, the assets employed using a discount rate of 7%.
As with the road modelling, operating costs per net tonne
kilometre for various product densities were calculated.
A volume of 115m3 per wagon (again based on 1 x 20 ft
container and 1 x 40 ft container) was assumed. Wagon
tonnage capacity was assumed to be 65 tonnes, meaning
that rail can accommodate full containers of higher density
products (over 500kg/m3) and therefore continues to
realise the benefits of reducing costs as densities increase
above the truck limit of 250kg/m3.
4. Derivation of ‘Below Road’ operating costs
The methodology used to derive the ‘below road’ operating
costs followed the general methodology of the NRTC,
as outlined in the Updating Heavy Vehicles Charges:
Technical Report (September 1998)17. Exhibit A1.5 shows
a schematic of the process used in this analysis to derive
the operating costs.
Total road expenditure was based on the 1997/8 Arterial
and Local Road expenditure, which was taken from the
NRTC’s Regulatory Impact Statement in February 200018.
Individual vehicle parameters were taken from Appendix
B.2: Road Use Data by Vehicle Type (1997) in the
September 1998 paper.
Total costs are classified by the NRTC as ‘Allocable’ and
‘Non-Allocable’.
Non-Allocable costs are excluded from the NRTC’s charging
regime and include items such as:
> Driver licensing administration costs
> Vehicle registration administration costs
> Loan interest costs
> Heavy vehicle enforcement costs
> Access costs for local roads.
All other costs are ‘allocable’ and are included by the NRTC
in determining vehicle charges.
Within the group of ‘allocable’ costs, some were excluded
from the operating cost analysis in order to prevent double
counting with the road capital cost analysis. Those costs
that were excluded were:
> Pavement Constructions
> Land Acquisition
> Earthworks
> Other Extension / Improvement expenditure.
THE FUTURE FOR FREIGHT > 77
appendix 1: derivation of road and rail freight costs
17 National Road Transport Commission, Updating Heavy Vehicles Charges: Technical Report, September 199818 Note that the charges on page 17 of the September 1998 report were incorrect for some cost categories. The Regulatory Impact Statement contains up-to-
date and correct data for all cost categories.
The remaining allocable costs were split into separable and
non-separable components. Separable costs are those that
differ depending on the level of road use and the type of
vehicle. Non-separable costs have little relation to road use,
such as mowing roadside verges and the costs of building
a minimum possible standard of road. The split between
separable and non-separable components for each cost
category was based on the NRTC’s assumptions, except for
routine maintenance and reseals. The NSW and Victorian
benchmarks (rather than the Australia wide benchmarks)
were used for these two categories.
Various measures of road use (cost drivers) were used to
allocate the separable component of each cost category
to specific types (e.g., a 6-axle semi). ESAL-kilometres
(a measure of the relative road wear responsibilities of
different loads on different axles) were used to allocate
costs for routine maintenance, road rehabilitation and
periodic maintenance of sealed roads.
Gross vehicle mass kilometres (GVM-km) were used to
allocate bridge maintenance and construction costs, while
passenger car unit kilometres (PCU-kms) were used to
allocate all other cost categories. All non-separable costs
were allocated to each vehicle type based on PCU-kms,
except for bridge construction costs, which were allocated
using GVM-kms.
Values for each cost driver were arrived at by multiplying
annual vehicle kilometres travelled for each vehicle type by
parameters representing individual vehicle characteristics.
For example, to get GVM-kilometre figures for trucks,
annual kilometres travelled were multiplied by average gross
vehicles masses for the different truck types. Differences in
vehicle characteristics therefore have a significant impact on
the cost driver values for each vehicle type and the share of
costs attributed to each vehicle—e.g., cars and other light
vehicles are assumed for the purpose of cost allocation by
the NRTC to have an effective average gross mass of zero
(relative to trucks) and so receive no cost allocation where
the allocation parameter chosen is GVM-km.
78 < AUSTRALASIAN RAILWAY ASSOCIATION
appendix 1: derivation of road and rail freight costs
EXHIBIT A1.5: DERIVATION OF BELOW ROAD OPERATING COSTS
costs
Local Road
etc
Capital improvements
Non-capitalcosts
Allocatedcosts
Non-allocated
Arterial/
Expenditure (97/98)
Non-Separable
Separable
Construction—pavements
Construction—land
Construction—earthwork
Excluded from operating cost analysis
Heavy Vehicles
Other Vehicles
Heavy Vehicles
Other Vehicles
Heavy Vehicles
Other Vehicles
Construction—other
7 axle B-double
8 axle B-doubleetc.
7 axle B-double
Allocated based on PCU-kms excludes local roads access costs
Items included in operating cost analysis
5 axle Semi
6 axle Semi
5 axle Semi
6 axle Semi
8 axle B-double
etc
To calculate costs per ntk for each heavy vehicle type, costs
in each category are split into their separable and non-
separable amounts. The separable amount is then divided
by the total value of the chosen cost driver to get a dollar
per unit cost figure (e.g. dollars per GVM-km for separable
bridge costs). Unit costs are then multiplied by the cost
driver values for each vehicle class to give an allocation
to each vehicle type. Allocations in each cost category are
summed to give a total allocation to each vehicle type. In
the case of freight vehicles, these allocations can then be
divided by total ntk travelled to give a final figure in dollars
per net tonne kilometre. Since GVM-km data was already
available, net tonne kilometres were derived from GVM-km
using a ratio of 0.57 net tonnes per gross tonne.
The cost drivers chosen for the separable components were
largely in-line with the NRTC’s allocation procedures used
in the 2nd Heavy Vehicle Pricing Determination19. The only
area of difference is the use of ESAL-kms for allocating
the costs of routine maintenance, road rehabilitation and
periodic maintenance of sealed roads. The NRTC currently
uses GVM-kms, however in meetings with the NRTC it
became clear that their methodologies were under review
for the 3rd determination. The NRTC indicated that it
was possible that ESAL-kms would be used in future
determinations, as this measure better reflected road wear
by vehicle type than GVM-kms.
Exhibit A1.6 summarises the separable and non-separable
allocations assumed by the NRTC and the BTRE, as well as
the cost allocation drivers used.
In order to obtain a split between fixed and variable costs
(as distinct from ‘separable’/’non-separable’ which only
refers to the ability to allocate costs to a particular vehicle
type), variable costs were considered to be those due to
‘wear and tear’ on the roads. This encompassed three of
the cost categories as defined by the NRTC:
> Routine Maintenance
> Road Rehabilitation
> Periodic Maintenance of Sealed Roads
All other costs were considered to be ‘fi xed’. The weighted
average operating cost for B-doubles and semi trailers with
more than 5 axles was determined to be $9.5 per ’000 ntk.
Variable costs were $5.7 per ’000 ntk, with the remaining
$3.8 per ’000 ntk being attributable to fi xed costs. The
weighted average cost was used in all analyses and was
assumed to be independent of the freight corridor.
In order to determine a total operating cost for the road
system, non-allocable costs were included as part of this
analysis (excluding local road access costs). These costs
were attributed to each vehicle type based on its number
of PCU-kms. These non-allocable costs are generally
recovered through registration fees charged by each State
in addition to the national heavy vehicle charge. For the
purpose of this analysis, it was assumed that there was full
recovery of these costs.
Exhibit A1.7 compares the operating cost results by truck
type from four separate sources. The methodologies used
by PJPL, NECG and BTRE are generally similar, with
differences in the final results being attributable to the
treatment of capital costs and the cost allocation drivers
selected. All are higher than operating costs calculated
using the current NRTC allocations—the key difference
being in the share of costs treated as separable (i.e. related
to individual vehicle characteristics).
THE FUTURE FOR FREIGHT > 79
appendix 1: derivation of road and rail freight costs
19 National Road Transport Commission, Updating Heavy Vehicles Charges: Technical Report, September 1998
80 < AUSTRALASIAN RAILWAY ASSOCIATION
appendix 1: derivation of road and rail freight costs
Source: BTRE Working Paper 40; NRTC Updating Heavy Vehicle Charges
Cost Category
NRTC Percent Non-Sep
BTREPercentNon-Sep Cost allocation driver used
Separable Non-Separable
Percent non-separable based on NSW and VIC benchmarks
Excluded from operating cost analysis—included in capital analysis
EXHIBIT A1.6: COMPARISON OF BELOW ROAD COST ALLOCATION METHODOLOGIES
Different parameters used
Routine Maintenance 50 20 ESAL-km PCU-kmReseals 50 20 ESAL-km PCU-kmRoad Rehabilitation 55 20 ESAL-km PCU-kmServicing 100 100 PCU-km PCU-kmBridge repair 67 33 GVM-km PCU-kmLow cost improvements 0 0 PCU-km PCU-kmConstruction—Bridges 85 55 GVM-km GVM-kmPavement constructions 55 55 n.a. n.a.Land 90 90 n.a. n.a.Earthworks 90 90 n.a. n.a.Construction—other 90 90 n.a. n.a.Miscellaneous works 100 100 PCU-km PCU-kmCorporate Services 100 100 PCU-km PCU-km
5 axle semi
6 axle semi
> 6 axle semi
7 axle B -double
8 axle B -double
9 axle B -double
PJPL assumption NECG estimate BTE (WP40) NRTC estimate
-
Variable costs
Fixed costs
costs allocated by costs allocated by
Key notes:
(only calculated for 6 axle semi)10.2
9.4
10.4
7.6
7.2
7.6
6.1
5.7
6.9
3.2
3.4
4.1
4.1
3.8
3.5
4.4
3.7
3.5
12.1
10.8
12.3
8.1
7.5
8.1
8.9
8.1
10.1
4.4
4.9
5.9
3.2
2.7
2.2
3.7
2.6
2.2
11.27.3 3.9
6.5
6.1
6.4
5.1
4.9
5.1
5.1
4.9
5.4
3.8
3.9
4.2
Avoidable costs allocated by esal-kms
- Non-separablecosts allocated by PCU-kms
- Non-allocated costs included and excludes capital costs.
- Net tonnes: gross tonnes = 0.57
- Avoidable costs allocated by esal-kms
- Non-separablecosts allocated by PCU-kms
- Non-allocated costs excluded and includes capital costs
- Net tonnes: gross tonnes = 0.57
- Avoidable costs allocated based on esal-kms
- Non-separable
PCU-kms- Non-allocated
costs excluded. - Includes capital
costs- Net tonnes: gross
tonnes = 0.50
- Avoidable costs allocated based on GVM-kms
- Non-separable
VKT- Non-allocated
costs included and excludes capital costs
- Net tonnes: gross tonnes = 0.57
EXHIBIT A1.7: BELOW ROAD—COMPARISON OF COST ESTIMATES
THE FUTURE FOR FREIGHT > 81
appendix 1: derivation of road and rail freight costs
EXHIBIT A1.8: BELOW RAIL OPERATING COST COMPARISON$ per '000 ntk
PJPL assumptions* NECG estimates BTE estimate (WP40)
*Assumes that RIC's costs will reduce by 50% post-merger under ARTC management and volumes increased by 100%
Based on analysis of RIC and ARTC's cost structure from annual reports, government reviews
Based on annual reports; assumes 53% increase in ARTC's cost structure; minimal reduction in RIC cost structure
Assumed that rail freight operators currently pay the full cost of their infrastructure use
Syd -Bris
Melb-Syd
Melb-Bris
Melb-Adel
Adel-Perth
Melb-Perth
Syd -Perth
Syd -Bris
Melb-Syd
Melb-Bris
Melb-Adel
Adel-Perth
Melb-Perth
Syd -Perth
11.4
9.8
10.6
6.6
6.6
6.6
8.5
6.7
5.2
5.9
2.0
2.0
2.0
3.9
4.7
4.6
4.7
4.6
4.6
4.6
4.6
15.2
12.5
13.9
7.0
7.0
7.0
10.5
3.3
3.3
3.3
3.3
3.3
3.3
3.3
11.9
9.2
10.6
3.7
3.7
3.7
7.2
8.7
8.7
8.7
8.7
8.7
8.7
8.7
Variable costsFixed costsFixed and variable
5. Derivation of ‘Below Rail’ operating costs
Operating costs were calculated for both ARTC and RIC
track, based on publicly available data.
ARTC costs were taken from the 2001/02 Annual Report
and assumed a current freight task of 10.5bn ntk. It was
assumed that 90% of the ARTC cost base was related to
the Intermodal task. Employee costs, maintenance costs,
operating lease expenses and insurance and project
development costs were assumed to be ~60% variable,
while incident costs and other expenses were taken as
100% fixed. The resulting costs per ’000 ntk for ARTC
are $4.6 fixed and $2.0 variable.
RIC’s cost structure was based on the 2002/03 budget
for the Access Division, taken from the “Independent
Review of RIC Metropolitan Maintenance Funding (October
2002)”. It was assumed that 40% of the cost base was
attributable to Intermodal freight, with the remaining 60%
being split between Coal ($70m) and Grain ($180m). The
Intermodal costs were based on a freight task of ~5bn ntks
per annum. Employee costs, external asset maintenance
costs and materials were assumed to be 20% variable.
Costs associated with major periodic maintenance were
assumed to be 50% variable, and all other costs were taken
as 100% fixed. Under these assumptions, RIC’s variable
costs are ~$13.5 per ’000 ntk, with fixed costs of ~$18.5
per ’000 ntk.
However, it is expected that RIC’s cost structure will be
significantly reduced once it comes under the control of
ARTC. Therefore, the analysis has used estimates of an
‘efficient RIC cost structure’ in order to compare the total
cost of road against rail. To arrive at an efficient cost level,
it was assumed that all of RIC’s costs could be reduced
by 40% (which is indicative of cost reductions achieved in
public sector organisations after privatisation) and that a
subsequent 15% cost reduction could be achieved through
merger synergies, giving a total cost reduction of 50%.
These cost reduction assumptions are in line with those
put forward by the ARTC in their 2002 annual report. It is
also anticipated that in reducing these costs, the volume
transported by Intermodal freight on RIC’s corridors will
double in the next five years. The combined effect of these
cost reductions and volume increases is an operating cost
per ’000 ntk of $6.7 for variable costs and $4.7 for fixed
costs (a reduction of $20.6 per ’000 ntk). Further volume
growth obviously reduces the fixed costs per unit on both
RIC and ARTC track.
Exhibit A1.8 compares the derived ‘below rail’ operating
costs with those from other sources. The NECG estimates
differ to the PJPL estimate due to more conservative cost
reduction assumptions on the RIC cost base by NECG,
and assumptions regarding the potential for cost increases
in ARTC’s existing cost structure due to current low
maintenance requirements.
6. Derivation of ‘Below Road’ capital costs
Below road capital requirements for freight users were
derived principally from the BTRE’s forecast of highway
spending requirements laid out in Working Paper
35—“Roads 2020”20. The report considered highway
expansions required both for traffic growth and to ease
current bottlenecks. However, for the purposes of this
analysis, both the ‘backlog’ and ‘growth’ investments were
considered to be necessary for growth (i.e., the backlog
capacity investment would not be undertaken if no further
traffic growth was to happen). Traffic volume data for each
corridor (in average vehicles per day) was provided by
the BTRE, split into passenger vehicles and commercial
vehicles. The inter-capital freight task in tonnes per year
was converted into inter-capital freight vehicles per day
using an assumption of 20t per vehicle (consistent with
BTRE methodology). The BTRE’s commercial vehicle
volume data was thus split into inter-capital and non inter-
capital freight.
There are a number of alternative methods by which
investments in road capacity could be allocated to users.
These include Vehicle Kilometres Travelled (VKT)—a
measure of straight traffic volume; Passenger Car Units
(PCUs)—a measure of how much a vehicle contributes
to congestion both through its ‘footprint’ and its speed/
acceleration characteristics; and Equivalent Standard Axle
Loads (ESAL kilometres)—a measure of how much load
the vehicle puts on the road surface. PCUs were adopted
as the allocation parameter as investments in new capacity
are driven primarily by considerations of congestion easing
(although a more complex model of economic and financial
considerations underpins the BTRE’s forecasting model,
including vehicle costs, accident avoidance, etc). PCU
parameters for cars, non-inter capital freight, and inter
capital freight were set at 1, 1.7 and 3.5 respectively;
based on parameters for cars, light trucks and a heavy
truck mix of semis and B-doubles used by the BTRE and
NTRC. The parameters used were for travel on flat straight
roads—gradients and curvatures significantly amplify the
contribution of heavy vehicles to congestion.
82 < AUSTRALASIAN RAILWAY ASSOCIATION
appendix 1: derivation of road and rail freight costs
20 Bureau of Transport and Regional Economics, Working Paper 35, Roads 2020, 1997
THE FUTURE FOR FREIGHT > 83
appendix 1: derivation of road and rail freight costs
EXHIBIT A1.9: DERIVATION OF BELOW ROAD CAPITAL COSTS
12.8m veh/year
* Figures relate to projected growth in traffic and capital required to support this growth**Weighted average based on corridor lengths and freight tasks
1.0 PCU/veh
5.9m veh/year
1.7 PCU/veh
3.5 PCU/veh
1.4 veh/year
Growth Capital $6.8bn
Road Capital Deficit$1.7/'000ntk
Road Growth Capital*$3.1/'000ntk
Total below road capital$4.8/'000ntk
4.9m PCUIntercapital freight
Capital deficit $3.1bn
10.2m PCULocal freight
12.8m PCUPassenger vehicles
PCU growth27.9m
Growth Capital$243/PCU
Capital deficit$113 /PCU
3.5pcu/veh20t/veh1200km/trip**
3.5pcu/veh20t/veh1200km/trip**
Thousand vehicles per day Thousand passenger car units (PCU) equivalentsCars only AADT'Local' freight AADTIntercapital freight AADT
15.1 15.1
18.5
5.6
10.3
2.2
3.9 4.1
13.6
9.8 10.5
2.7
5.9
1.2 2.2 2.9
4.54.2
1.4
2.7
0.81.4 1.1
1.50.8
3.7
1.5
1.7
14.0
12.7
14.1
4.0
7.9
1.8
3.13.6
13.6
9.8 10.5
2.7
5.9
1.22.2 2.9
2.62.5
0.8
1.6
0.5
0.80.6
0.4
0.2
1.1
0.4
0.5
Sydto
Bris(PacificHwy)
Sydto
Bris(inlandroute)
Melbto
Syd
Melbto
Bris
Melbto
Adel
Adelto
Perth
Melbto
Perth
Sydneyto
Perth
Sydto
Bris(Pacific Hwy)
Sydto
Bris(inland route)
Melbto
Syd
Melbto
Bris
Melbto
Adel
Adelto
Perth
Perthto
Melb
Sydto
Perth
EXHIBIT A1.10: ROAD TRAFFIC FLOWS BY MAJOR CORRIDOR
In this respect the allocation methodology chosen erred on
the side of conservatism. Traffic flows for the three classes
of vehicle were converted from vehicles per day into PCUs
per day for each corridor. Thus the relative use of the road
capacity by each class of user could be calculated.
Since the investment data used was a forecast based on
future traffic growth, it was the growth in each type of road
use, in PCUs, that was used as the basis for calculating $/
PCU. This was then converted to $/’000ntk figure for freight
using the PCU/vehicle and tonne per vehicle assumptions.
Exhibit A1.9 shows the structure of the calculation and
Exhibit A1.10 shows the impact of converting vehicles to
PCUs on the shares allocated to different vehicle classes.
7. Derivation of ‘Below Rail’ capital costs
No direct equivalent to the BTRE’s Roads 2020 forecast
exists for the rail network, although the ARTC Interstate Rail
Network Audit21, undertaken by Booz-Allen & Hamilton
Consulting, did consider investments requirement to
improve service levels on the rail network (along with the
capacity enhancements required to absorb the additional
freight attracted to rail by improved service). Therefore a
model to derive below rail capital required for growth was
developed from a combination of publicly available data and
consultation with Pacifi c National personnel in infrastructure
planning and operations areas; providing a unique and
comprehensive view of capacity and requirements from both
an above and below rail perspectives.
In considering capacity investments required to support rail
freight growth on the existing track, capacity increases were
considered as coming from two sources:
> Improvements in utilisation of existing rolling stock and
track infrastructure, and
> Increases in the capacity of the track infrastructure
itself. Additionally, considerations of service
enhancements required to attract additional freight
volume to rail were also taken into account.
Indications from customer interviews undertaken in the
middle of 2003 were used to gauge how much additional
freight could be gained through capacity provision alone
and how much would require service levels to be improved.
This led to a ‘threshold’ level of capacity increase, below
which no investment in service levels would be required.
Opportunities to increase capacity were considered
as discrete units in a logical, chronological sequence
(e.g. increase tonnes per container, increase containers per
wagon, increase wagon per train, etc). They were quantified
using operational data (containers per wagon, train
lengths, etc) and simple models of track capacity (e.g. the
relationship of crossing loops to path availability) and below
rail investments necessary to support each additional
increment were then costed. Service level investments
were allocated to capacity increments once the cumulative
increases had crossed the ‘threshold’ level mentioned
above. Projected rail freight growth to 2014 was used to
calculate the investment required on a dollar per ntk basis
Exhibit A1.11 lays out the methodology for costing the
basic. capacity increases and the drivers used.
8. Externalities
In the context of transport, externalities refer to the costs
occasioned by users that are not internalised (e.g. by
insurance). In determining the externalities associated with
both road and rail, the analysis has taken into account the
following measures:
> Noise pollution
> Air pollution
> Congestion costs
> Greenhouse gas emissions
> Accident costs
84 < AUSTRALASIAN RAILWAY ASSOCIATION
appendix 1: derivation of road and rail freight costs
21 Booz Allen & Hamilton/ARTC, Interstate Rail Network Audit: Final Report, 2001
A variety of sources have been collated on these measures,
as shown in Exhibits A1.12 and A1.13. Where available,
a low, medium and high value was estimated for each
component of the externality cost, based on the sources
quoted. The only exception to this was for the derivation of
accident and greenhouse externalities for road, where more
detailed analysis was performed (discussed below). In order
to be conservative, the medium estimate has been used in
all further analyses of total costs. The rural option has been
chosen where available, as the majority of the intermodal
transport occurs in rural areas. Exhibit A1.14 compares
the values adopted for road and rail intercapital freight
transport, and shows that the assumed difference between
road and rail externalities for the purpose of this analysis
was $6/’000 ntk.
The area of work surrounding the cost of externalities is
constantly evolving and it is expected that the accuracy of
these sources will improve over time.
Particular attention was paid to the road greenhouse and
accident externalities—as these are the most significant
externality cost items and contribute to the majority of
the $6/’000ntk difference. Estimates in the literature
vary widely, due to differences in the data sources used
(e.g. accident cost estimates have been revised sharply
upwards in later exercises as more costs are taken into
account, and assumed cost per fatality has increased22),
and assumptions over key drivers (e.g. kg of CO2 per litre
of diesel, for greenhouse estimates). We have made the
following assumptions.
THE FUTURE FOR FREIGHT > 85
appendix 1: derivation of road and rail freight costs
Growth in rail taskBelow rail capital($/ '000 ntk)
Capital for capacity expansion($/'000ntk)
Capital for capacity expansion($/'000ntk)
Market growth (4.5%)
Share growth(targets by corridor)
Capacity opportunities
ARTC investment projections
Mass utilisation Slot utilisation Length utilisation Path utilisation 1900m pathsNew pathsDouble stacking
New track ($455k/km)New turnouts ($350k each)
Growth possible within current service levels
Cost of improving service(BAH data for MS, SB, SP and MP)
Growth potential with/without service level improvements gauged from interview with a sample of rail customers
Investments split into 'loop enhancements' and 'other' and allocated to capacity increments
New signalling ($900k each)
EXHIBIT A1.11: CALCULATING BELOW RAIL CAPITAL REQUIREMENTS
22 For example, the total crash cost estimate for 1993 was ~$6.1b (BTRE Information Sheet 14), but 1996 estimates using higher fatality and injury costs evaluated crash costs at ~$15b (BTRE Report 102: Road crash costs in Australia)
86 < AUSTRALASIAN RAILWAY ASSOCIATION
appendix 1: derivation of road and rail freight costs
$ per '000 ntk EXHIBIT A1.12: EXTERNALITY ASSUMPTIONS—ROAD
Source: Laird P., Land freight external costs in Queensland, 2002; Bureau of Transport and Regional Economics, Working Paper 40: Competitive neutrality between road and rail, 1999
* Booz Allen & Hamilton — figures from the Interstate Rail Network Audit, 2001 **Qld Transport assume $25/t of CO2, Bus Industry Confederation assume $40/t of CO2
Rural 0.03 0.03 0.03 0.25Metro 0.06 1.32 0.06 2.10 0.06 1.00
0.34
Air Pollution Rural 0.00 0.00 0.00 0.00Metro 1.10 1.20 1.10 2.10 1.10 1.15 1.2
0.10
Greenhouse Gases** 1.60 not calculated 1.70 3.20 1.40 1.55 1.70
Congestion/Enforcement cost
Rural 0.00 not calculated 0.00 0.40 0.80Metro 0.90 0.90 not calculated 0.80 0.85 0.90
0.80Accident costs Rural 3.20 7.00 3.20 5.10 7.00
Metro 3.20 7.00 3.20 5.10 7.003.20
Range used
Externality Measure BAH* NRTC
Bureau Transp . Econ
(1999)Qld
TransportBus Industry
Confederation Low Medium HighNoise Pollution 0.50
1.32
Totals Rural 4.8 8.7 3.2 4.6 7.3 10.0Metro 6.9 10.8 7.4 6.8 9.3 12.1Total 12.8 5.8
0.10 0.200.10 0.20
0.00 0.000.15 0.30
1.10
0.000.00
0.300.30
1.601.90
0.000.04
0.000.30
0.60 0.90
0.00 0.000.00 0.00
0.24 0.270.24 0.27
0.84 1.270.90 1.40
0.04 0.700.18
0.30 0.700.04
not calculated 0.64 1.10
not calculatednot calculated
0.240.24
0.300.88 1.101.22 2.50
1.90
Rural 0.00Metro 0.40
Rural 0.00Metro 0.30
1.10
Rural 0.00Metro 0.00
Rural 0.30Metro 0.30
Rural 0.40Metro 0.74Total
Range used
Externality Measure BAH* NRTC
Bureau Transp . Econ
(1999)Qld
TransportBus Industry
Confederation Low Medium HighNoise Pollution
Air Pollution
Greenhouse Gases**
Congestion/Enforcement cost
Accident costs
Totals
0.52
$ per '000 ntk EXHIBIT A1.13: EXTERNALITY ASSUMPTIONS—RAIL
Source: Laird P., Land freight external costs in Queensland, 2002; Bureau of Transport and Regional Economics, Working Paper 40: Competitive neutrality between road and rail, 1999
* Booz Allen & Hamilton — figures from the Interstate Rail Network Audit, 2001 **Qld Transport assume $25/t of CO2, Bus Industry Confederation assume $40/t of CO2
THE FUTURE FOR FREIGHT > 87
appendix 1: derivation of road and rail freight costs
Greenhouse GasesAccident costsNoise Pollution
EXHIBIT A1.14: COST OF ROAD AND RAIL EXTERNALITIES—RURAL AREAS$ per '000 ntk
*See previous exhibitSource: Laird P., Land freight external costs in Queensland, 2002; PJP analysis
Low Case High Case Difference
Congestion Costs*
4.6
0.81.4
0.6
3.2
Road Rail
10.0
1.6
1.7 1.1
7.0
0.3
0.50.8
Road Rail
3.8
6.0
8.4
0.8 0.6
3.0
6.7
0.30.8
Low Case
Assumed High Case
Road accident externality:
We have used the methodology used by the BTRE in
Working Paper 40 (Competitive Neutrality Between Road
and Rail, 1999), but with more up to date 1996 accident
cost data rather than the 1993 data in that study. Thus key
assumptions are:
> Total accident costs: $15b in 1996 dollars (BTRE
Report 102)
> Accident costs in 2004 dollars: $18b
> Articulated truck share: 5% (BTRE Working Paper 40)
> 6-axle semi share by PCU-kms: 55% (BTRE WP 40)
> % of costs internalised through insurance: 50% (BTRE)
> 6-axle semi annual ntkms: 57.9b ntk (BTRE Working
Paper 40)
> (19,000 x 0.05 x 0.55 x 0.5) ÷ 57.9 = $4.3/’000ntk
Since ATSB data shows that articulated truck crashes occur
disproportionately on roads with speed limits > 100kph
(ATSB website), we assume that crash costs per truck
kilometre on intercapital highways (e.g. Pacific Highway) are
20% higher than national average—i.e. $5.1/’000ntk
Greenhouse externality:
We reviewed the methodology and estimates of greenhouse
externalities cited in Philip Laird’s study of externality values
for Queensland Transport, as well as receiving specific
comments from the BTRE on their own work on greenhouse
costs (e.g., the CSIRO/BTRE/ABARE study ‘Appropriateness
of a 350ML biofuels target’).
There are three key assumptions behind most estimates
of greenhouse costs per tonne kilometre:
> Cost per tonne of Carbon Dioxide ($/t)
> Emissions per litre of diesel (kg/l)
> Fuel consumption (litres per ntk).
88 < AUSTRALASIAN RAILWAY ASSOCIATION
appendix 1: derivation of road and rail freight costs
EXHIBIT A2.1: METHODOLOGIES FOR CALCULATING ROAD USE COSTS
Description Examples* Possible impact
1. 'Equity' - Allocate ALL costs between users (the current Australian PayGo regime is an example)
- May or may not refer to marginal costs as a lower bound for allocations
- NRTC approach
- UK NERA approach
- US federal studies
- EU Commission study
- Outcome is heavily dependent on how 'non-separable' costs are allocated —by VKT or PCU. Current dominant methodology internationally; favours heavy vehicles if non-separable costs are allocated by VKTs
2. Engineering - Estimate the marginal cost of road usage, including impact on other road users due to road damage, based on engineering models
- 'Direct'— uses pavement management system models (e.g. HDM 4) to estimate the marginal cost of road use
- 'Indirect'— uses Newbery's theorem, linking ESALs to wear
- If Marginal Cost < Average Cost then will reduce costs and will fall short of PayGo
- If Marginal Cost = Average Cost, then because allocations are made based on ESAL's, it will result in increased heavy vehicle costs
3. Econometric - Use economic models on historical datasets to estimate the impact of traffic on costs
- Only works if there are strong datasets (few available now)
- VKT*, GVM*, ESAL* are correlated, making estimation of impact difficult
- The Link Study (2002)
- Li et al. (2001)
- Martin (1994)
- If successful, likely to result in increased allocation to heavy vehicles BUT currently does not take account of the affect of road damage on other vehicles (road damage externalities)
*VKT = Vehicle Kilometres Travelled; PCU = Passenger Car Unit; GVM = Gross Vehicle Mass; ESAL = Equivalent Standard Axle Load
Source: Data cited in "Measuring the Marginal Cost of Road Use—an International Survey", Nils Bruzelius, 2003
We have adopted the fuel consumption assumption from
BTRE Working Paper 40 (0.0265 l/ntk), which is the same
as that being used in Philip Laird’s calculations.
For the emissions per litre assumption, we adopted the
2.69kg/l figure from Philip Laird’s paper (noting that the
same emissions assumption appears to have been used
in the recent Auslink guidelines on externality values).
The final assumption to be made was the cost of a tonne
of Carbon Dioxide. Estimates of future greenhouse gas
compliance costs vary widely. In their 1999 Discussion
paper, National Emissions Trading: Issuing the permits, the
AGO suggest a range from $10-$50 per tonne based on the
different national and international abatement schemes,
with a midpoint of $30. A figure of $10/t is felt by the
BTRE to be more appropriate for stage 1 emissions trading
schemes under the current Kyoto agreement (due to expire
in 2012). In an attempt to take a more forward-looking
view, we have chosen a value of $20/t ($22/t in 2004
dollars). This gives a road greenhouse externality value of
$1.55/’000ntk.
1. Introduction
This appendix reviews current heavy vehicle user
charges, taking into account two emerging international
developments.
The developments are:
> Changing views relating to road cost allocation
methodologies and outcomes between different users
and user classes
> The development of technologies facilitating mass-
distance charging and the early implementation of
those technologies in a number of European countries.
In summary, we conclude that heavy vehicles receive
unduly favourable treatment using the current cost
allocation methodology and charging regime. The
remainder of this appendix addresses in turn:
> Why current allocation methods result in undercharging
of heavy vehicles
> The need to consider the shift to mass-distance
charging.
2. Why current cost allocation methods result in
undercharging heavy vehicles
2.1 Overview of the Three Different Cost Allocation
Methodologies
Internationally, there are three types of cost allocation
methodologies. Bruzelius (2003) classifies these as ‘Equity’,
‘Engineering’ and ‘Econometric’ approaches (Exhibit A2.1):
2.1(i) Equity methods
Equity methods seek to allocate costs amongst users on the
basis of fairness. They treat road users as if they belong to
a club that must collectively cover all road costs. Allocation
of costs within the club aims to ensure that users pay a fair
or ‘equitable’ contribution on the basis of the nature and
amount of their road use. The majority of cost allocation
methodologies used by roads authorities around the world
make use of this approach. Its appeal lies in the fact that
all costs can be recovered from the road user ‘club’ as
a whole—with non-separable costs split up on the basis
of some ‘fair’ parameter. However, the allocation of costs
within the ‘club’ is sensitive to the parameters chosen to
divide costs up amongst members of the club.
THE FUTURE FOR FREIGHT > 89
appendix 2: comparing international road costing methodologies and charging regimes
This appendix provides further elaboration on heavy vehicle access charging discussed in Chapter 3.
The main parameters used in Equity method cost
allocations are:
> Vehicle Kilometres Travelled (VKT)—all vehicles treated
as being equal.
> Passenger Car Unit kilometres (PCU)—vehicles are
weighted by their size relative to a passenger car. Some
approaches also take account of characteristics such
as acceleration and braking.
> (Average) Gross Vehicle Mass kilometres (AGM/GVM)—
vehicles are weighted by their total mass.
> Equivalent Standard Axle Load kilometres (ESAL)—the
ESAL value is calculated for each axle of a heavy
vehicle as [Actual Axle load/Reference load]4. These
are summed to give a total ESAL value for the vehicle.
Because the guiding principle is the notion of ‘fairness’,
parameter choices involve an element of discretion on
the part of the road authority, rather than being purely
scientific. This is important as the choice and value of the
parameter used to allocate a particular cost item can have
a significant impact on the proportion of that cost attributed
to different classes of road user.
Exhibit A2.2 illustrates the generic equity allocation process
and Exhibit A2.3 describes the general characteristics of
and issues with the approach.
2.1(ii) Engineering methods
Engineering cost allocation methodologies seek to allocate
costs on the basis of engineering models of road damage.
Two approaches exist:
> Direct Approach. A Pavement Management System
(PMS) is used to forecast road management costs
resulting from incremental traffi c fl ows. The PMS
contains a set of cost drivers for road damage (both
direct repair costs and costs to road users) together
with models relating road use to road damage,
congestion, etc. Costs of additional units of different
types of road use can be compared with a base case
to calculate their marginal costs. Importantly, because
decisions in PMS systems take account of both the
costs of road repairs and the costs incurred by users,
the Direct Approach also captures the “damage
externality”—the costs that one road user’s road
damage imposes on subsequent users.
> Indirect Approach. Makes use of a theoretical
relationship between road use and road wear
developed by Professor David Newbery (Exhibit A2.4).
Newbery’s Theorem calculates marginal cost as the
average cost of road repairs per ESAL-km, by dividing
the cost of periodic overlays by the accumulated ESAL-
kms of traffi c load carried between overlays. Unlike
the Direct Approach, road wear costs are the costs of
road repairs only, and do not take account of externality
costs imposed on other road users. In terms of cost
recovery, costs allocated to traffi c are scaled by a factor
(<=1) that takes account of non-load factors such as
weather. Consequently, marginal costs so calculated
may be less than average costs unless the proportion of
wear due to non-load factors is set to zero.
90 < AUSTRALASIAN RAILWAY ASSOCIATION
appendix 2: comparing international road costing methodologies and charging regimes
THE FUTURE FOR FREIGHT > 91
appendix 2: comparing international road costing methodologies and charging regimes
EXHIBIT A2.2: ‘EQUITY’ OR ‘CLUB’ APPROACH—BASIC FORMULA
-
-
-
-
-
--
-
Allocated ‘arbitrarily’— generally based on VKTs
Many different bases upon which costs are allocatedNo damage externalities taken into account in allocating costs
French use PCUs* BTRE** uses PCUs
Separable
Non-separable
Allocated
Non-allocated
All costs
How Allocated
Allocated based on wear and tear impact usually ESALs* or GVM*
Not allocated
Comments
Not allocated
*VKT = Vehicle Kilometre Travelled ; PCU = Passenger Car Unit; GVM = Gross Vehicle Mass; ESAL = Equivalent Standard Axle Load **Bureau of Transport and Regional Economics
Source: National Road Transport Commission, Updating Heavy Vehicle Charges: Technical Paper, 1998;Bureau of Transport and Regional Economics, Working Paper 40: Competitive neutrality between road and rail, 1999Perkins S., Recent developments in road pricing in Western Europe, 2002
*VKT = Vehicle Kilometre Travelled; PCU = Passenger Car Unit; GVM = Gross Vehicle Mass; ESAL = Equivalent Standard Axle Load
Source: Bruzeliu N., Measuring the Marginal Cost of Road Use—An International Survey, 2003
Generally splits costs into fixed (independent of use) and variable (with road use)—also applied to capital through a base incremental split
Variable costs allocated by usage data
Fixed costs allocated by neutral parameter (usually kilometres travelled)
Actual costs from road authority accounts—do not take account of road damage externality
No fixed rules for the allocation of costs between fixed and variable—significant variations by country
Sensitive to parameter chosen
Choice of VKT* vs. PCU* is important as the latter assigns more cost to large vehicles
Costing based on a snapshot rather than an average over time
Externalities not taken into account
EXHIBIT A2.3: OVERVIEW OF THE EQUITY OR ‘CLUB’ APPROACH
CHARACTERISTICS
-
-
-
-
-
-
-
-
-
ISSUES
92 < AUSTRALASIAN RAILWAY ASSOCIATION
appendix 2: comparing international road costing methodologies and charging regimes
EXHIBIT A2.4: COSTS ALLOCATED BY THE ‘INDIRECT’ APPROACH—THE NEWBERY THEOREM
Source: Bruzeliu N., Measuring the Marginal Cost of Road Use—An International Survey, 2003
Share of road wear due to traffic (%)
Cost/Km ($)
Years between overlays
ESAL-kms/year
Total allocated cost / overlay
Total traffic load in ESAL-kms/overlay
Marginal cost per ESAL-km
> RUC30 model (Newbery theorem)—effectively a simplified version of the HDM4
> Swedish study found 1:15 MC relativity between cars and heavy vehicles (based on allocation of variable costs)
> Lindberg (2002) estimated marginal costs of ECU 0.3/km—ECU 1.9/km for goods vehicles
THE NEWBERY FORMULA STUDIES RELATINGTO THE INDIRECT APPROACH
÷
x
x
EXHIBIT A2.5: COMPARISON OF MARGINAL COSTS DERIVED FROM INTERNATIONAL STUDIES
from international
17
31
31
31
15
142
53
88
15
47
9
20
32
47
43
32
32
32
53
91
80
68
53
57
68
68
68
0.436.37
0.1420.10
0.2312.27
0.4813.76
0.182.73
0.6511.51
0.4813.76
0.4813.76
0.4813.76
* NRTC ‘Marginal cost’ calculated from maintenance costs only to be consistent with international studies **Imputed externality based on costs derived from Swedish study, which explicitly includes externality costs
from international
CarsTrucksImputedexternality cost**
MARGINAL COSTAus. cents/km
TRUCK: CAR COST RATIOTruck cents/km as multiple of car cents/km
RELATIVE ALLOCATIONPercent share of total costs
ENG.
Swiss study
Austrian study
EQUI
TYEC
ONOM
ETRI
C**
Swedish Direct
Swedish Indirect
Martin (1994)
Rosalin /Martin(1999)
6.0 PCU
NERA/ITS
NRTC*
Source: Bruzeliu N., Measuring the Marginal Cost of Road Use An International Survey, 2003;National Road Transport Commission, Updating Heavy Vehicle Charges: Technical Paper, 1998
‘Average’ split
2.1(iii) Econometric methods
Econometric methods use regression modelling to establish
“best fit” relationships between observed road costs
and a set of explanatory variables (of which road use is
one). Differentiation of these relationships then gives the
marginal costs. Because these methods include non-load
factors (weather, etc) in the explanatory variables used,
the marginal cost of road use calculated will be less than
average cost. Additionally, as with the indirect method
only road maintenance costs are taken into account and
damage externalities are ignored.
2.2 Comparing Outcomes from the International
Studies—Current Equity Methodology Results in
Undercharging of Heavy Vehicles
We have compared heavy vehicle cost allocations from
international studies with the current NRTC methodology
and found that, in most cases, moving from the current
NRTC cost allocation methodology to one of the alternatives
would increase the share of costs allocated to heavy
vehicles. To do this, we took the heavy vehicle results from
each study and compared them with the current NRTC
approach in three different ways:
> Marginal costs—The marginal cost impact of both
cars and trucks across the different methodologies
was compared. This demonstrated that while the
NRTC attributes a marginal cost of $6.37/km for
trucks, the numbers are much higher with alternative
methodologies, typically between $10 and $20/ km.
> Truck/car cost ratio—While marginal cost estimates are
useful, they are sensitive to exchange rates as well as
differences in conditions across different countries. To
allow for this, the ratio of marginal costs between trucks
and cars was considered. Whereas the NRTC truck/car
cost ratio is about 15 to 1, other studies suggest higher
ratios in most cases, in some cases over 100 to 1.
The two exceptions are the Swedish attempt to apply
the indirect methodology, and the Swiss econometric
study23.
> Relative cost allocation—Each of the studies results
in an implied allocation of total costs between trucks
and cars which, with no exceptions, results in a higher
allocation to trucks than the NRTC approach. This is
partly due to the different ratio of trucks to cars in each
of the countries where the studies were conducted,
however for the most part it is due to fundamental
differences in methodology.
Exhibit A2.5 compares the outcomes from the application
of various studies in the Australian context. The results
show that the Australian NRTC approach appears to
underestimate the impact of heavy vehicle costs as
compared with other counties applying the same “equity”
framework (e.g., the UK), and that moving to other more
objective methodologies (Econometric and Engineering) will
also result in a greater allocation of costs to heavy vehicles.
The remainder of this section describes the results of these
various studies.
2.2(i) Equity Approaches
Even when applying a more traditional equity approach,
the NRTC allocations are at the lower end of the spectrum.
The 2000 NERA report on Lorry Track Costs for the
UK Department of Transport gives details of the British
equity approach. The differences between the British and
Australian allocation regimes are outlined in Exhibit A2.6.
The key difference lies in the share of costs assigned by
each parameter. A greater share of costs is allocated by
parameters that give high weighting to trucks, resulting
in greater allocations to heavy vehicles. The effect is to
increase truck/car unit cost relativities by a factor of 10 and
to shift the overall heavy vehicle share of variable costs from
~50% to >80% after usage volumes are taken into account.
THE FUTURE FOR FREIGHT > 93
appendix 2: comparing international road costing methodologies and charging regimes
23 The Swedish study appears to have relied significantly on experience rather than a formal analysis to arrive at its cost allocations, with the Newbery formula only being applied at the final stage of analysis. The Swiss study was based on very limited data.
94 < AUSTRALASIAN RAILWAY ASSOCIATION
appendix 2: comparing international road costing methodologies and charging regimes
Source: NERA Report on Lorry Truck and Environment Costs, UK Department of Transport, 2000;National Road Transport Commission, Updating heavy vehicle charges: Technical Paper, 1998;Bureau of Transport and Regional Economics, Working Paper 40: Competitive neutrality between road and rail, 1999
EXHIBIT A2.6: UK COST ALLOCATION METHODOLOGY VERSUS THE AUSTRALIAN NRTC METHODOLOGY
*VKT = Vehicle Kilometre Travelled; PCU = Passenger Car Unit; AGM = Average Gross Mass; ESAL = Equivalent Standard Axle Load
40000
1007015151515150
50500000000000
00
450000
4500000
40100100
000000000
000000
85858585850
0000
332015
1010100
200000
3000000
100
505055
10067808555909090
100
Australia UK Australia UK Australia UK Australia UKCOST CATEGORY
VKT*PCU*
KilometresESAL*
KilometresAGM*
Kilometres
Routine MaintenanceResealsRoad RehabilitationServicingBridge repairLow cost improvementsBridgesPavement constructionsLandEarthworksConstruction—otherMiscellaneous worksCorporate Services 100 100 0 0 0 0 0
EXHIBIT A2.7: COMPARISON OF NRTC AND BTRE ALLOCATIONS WITH MARTIN'S ECONOMETRIC FINDINGS**
Source: Bruzeliu N., Measuring the Marginal Cost of Road Use—An International Survey, 2003;Bureau of Transport and Regional Economics, Working Paper 40: Competitive neutrality between road and rail, 1999
*VKT = Vehicle Kilometre Travelled; PCU = Passenger Car Unit; AGM = Average Gross Mass; ESAL = Equivalent Standard Axle Load **Based on Australian data
-separable
44% 74%
26% 4% 100%
30% 8% 0%
100% 0%
0% 82%
100%
0%
100%
0%
0%
100%
0%
0%
30% 45% 53%
70% 55% 47%
ESAL*
GVM*
PCU*
VKT*
PCU*
GVM* 0% 18% 0% 0%
50%
50%
NRTC BTRE
NRTC BTREMartin(1994)
Rosalin& Martin(1999)
Martin(1994)
Rosalin& Martin (1999)
Non
Separable
Allocated VKT* 0% 12% 0% 0%
Perkins (2002) gives another example of an alternative
equity based approach. Reference is made in his paper
to the French study that concludes that the passenger car
equivalent ratio for a “truck” should be 6:1. This figure
was arrived at after taking into account the acceleration
and braking characteristics of vehicles, as well as their
‘footprint’. The NRTC currently use values of between
1.7:1 and 4 :1 with an average of 3:1. For comparison,
the French value was applied within the NRTC framework,
resulting in an increase in the repair and rehabilitation cost
allocation to trucks from ~50% to ~65%.
2.2(ii) Engineering Approaches
Two numeric examples used the Direct and Indirect
Engineering Approaches with Swedish data.
The Direct Approach calculated both the road damage
costs and the damage externality per VKT for a truck and
a car and found a relativity of ~140:1 when the externality
was included. Application of this result to Australian road
usage data allocated ~70% of costs to trucks, versus the
~50% allocated under the current NRTC system.
The Indirect Approach produced lower absolute figures
and a lower truck:car cost relativity (15:1), implying that
the impact of including the damage externality is much
greater for trucks than for cars (indirect calculations deal
with repair costs only, rather than total economic costs of
road damage). Total cost allocation was in the same ratio as
NRTC cost allocations. The literature gives other examples
of attempts to apply Newbery’s theorem in different
countries and to different types of road cost, but none
provide enough information to allow a comparison with
Australian data to be made.
2.2(iii) Econometric Approaches
A number of attempts have been made internationally to
apply econometric techniques to the calculation of marginal
road user costs. However, the shortage of sufficiently
detailed or deep data sets means that econometric analyses
need further development before any hard conclusions
can be drawn from the numerical results. Whilst their
usefulness is therefore currently limited, two inferences can
be drawn from the studies.
The first is that in constructing best fit equations to explain
road wear costs, the majority of studies have used ESAL or
GVM kilometres to explain the traffic dependant portion of
the wear costs.
Secondly, the majority of studies, including two carried
out on Australian data (Martin in 1994 and Rosalin and
Martin in 1999), find the share of costs attributable to traffic
volumes (i.e., separable rather than non-separable) to be
at least 50%. This has implications for the current NRTC
system, which on average allocates only 30% of costs
as separable. Exhibit A2.7 gives details of the allocations
for the NRTC, a suggested allocation from the BTRE’s
paper on Competitive Neutrality, and the two Australian
econometric studies.
THE FUTURE FOR FREIGHT > 95
appendix 2: comparing international road costing methodologies and charging regimes
96 < AUSTRALASIAN RAILWAY ASSOCIATION
appendix 2: comparing international road costing methodologies and charging regimes
3. Need to plan for a shift to mass distance charging
Not only do current cost allocation methodologies result
in undercharging of heavy vehicles, current excise
and registration based charging regimes are especially
favourable to the long heavily laden inter-capital truck
journeys that compete with rail. The BTRE and NTRC,
as well as transport economists in other countries, are
increasingly recognising the distorting effect this has on
cost recovery (Exhibit A2.8). As a result the logic for, and
experience with, ‘mass-distance charging’ (i.e., charging
truck per tonne kilometre travelled) is increasing rapidly.
3.1 Comparing different heavy vehicle charging regimes
Traditional charging regimes in Australia and around the
world have focused on standard vehicle type registration
fees as well as excises on fuel. At the same time, the road
damage driven by a vehicle is a function of (amongst other
things) the distance it travels and the weight of its load.
Therefore, the current charging regime only approximates
the true damage incurred as a result of the activities of an
individual vehicle.
The impact of vehicle weight on the road damage incurred
is illustrated in Exhibit A2.9. Depending on the cost
allocation methodology employed, once a 6 axle articulated
truck reaches approximately 20 tonnes in weight, they are
undercharged, and the size of the ‘subsidy’ increases with
increased weight. At 30 tonnes the undercharging could
be as high as $6/‘000 ntk, even using an ‘equity’ cost
allocation methodology.
A similar relationship is observed when cost recovery
across different vehicle classes is considered. Exhibit
A2.10, based on the BTRE’s analysis of NRTC costs,
illustrates the variation of fuel charges with vehicle size and
mass (measured in terms of ESALs). The Exhibit shows
the average fuel revenue in cents per ESAL-km for each
vehicle type, and the average avoidable cost per ESAL-km
to be recovered from all vehicles. It can be seen that the
NRTC’s current charging regime over-recovers costs from
smaller vehicles and under-recovers costs resulting from
heavier vehicles.
3.2 Emerging international experience with mass
distance charging
Mass distancing charging is already moving from theory
to practice. Systems for charging heavy vehicles based
on vehicle mass (and other characteristics) and distance
travelled are in place in Switzerland and Austria and are
planned for implementation in Germany and the UK within
the next 4 years. The systems make use of a combination
of GPS technology, roadside transponders and onboard
units to accurately record vehicle movements through the
road network. In some cases, the charges are being set
by both vehicle mass and emissions class and so have
the potential to recover both road wear and environmental
externality costs.
Exhibit A2.11 summarises the schemes and technologies;
it is interesting to note that the Swiss system explicitly
hypothecates revenue from the system to rail infrastructure
Investment.
4. Implications
The conclusion of this work is that, while heavy vehicle
charges are clearly too low, more work is required to
provide an exact calculation of the right charges. To achieve
this, it will be necessary to establish the correct framework
for how charges should be set, and then apply it in the
Australian context.
THE FUTURE FOR FREIGHT > 97
appendix 2: comparing international road costing methodologies and charging regimes
EXHIBIT A2.8: MASS-DISTANCE CHARGING—VIEWS FROM THE TRANSPORT ECONOMISTS
"For road transport there is a fixed annual registration charge and a variable fuel charge... this charging structure
does not closely match the amounts paid to the individual vehicle's marginal cost of road use. Highly utilised
vehicles and those with good fuel consumption rates pay too little’
NRTC 3rd Heavy Vehicle Pricing Determination Issues Paper, 2003
"BTE results indicate that heavily laden vehicles are currently undercharged, lightly laden vehicles are overcharged
and the current imputed fuel excise credit does not recover the road wear costs caused by heavy vehicles. Some form
of mass distance charge would be more efficient."
BTRE Working Paper 40 "Competitive Neutrality between Road and Rail", 1999
"Passenger vehicles are expected to overpay federal user fees by about 10%, while single-unit and combination
trucks will underpay by about 10 percent (in 2000)... In virtually all truck classes, the lightest vehicles pay more than
their share of highway costs, and the heaviest vehicles pay considerably less than their share of costs. Modifications
to the HVAT* rate schedule or new taxes such as a WDT or axle-WDT could result in larger gains in equity."
US Federal Highway Cost Allocation Study, 1998
*Heavy Vehicle Utilisation Tax
Source: Literature search
$ per '000 ntk
Source: Bureau of Transport and Regional Economics, Working Paper 40: Competitive neutrality between road and rail, 1999
NRTC avoidable cost
NRTC fuel charge
BTRE avoidable cost
EXHIBIT A2.9: ‘AVOIDABLE’ * ROAD WEAR COSTS AND CHARGES—6 AXLE ARTICULATED TRUCK
*Avoidable costs are those resulting directly from traffic interactions with the highway—predominately wear and tear on the road surface
0
2
4
6
8
10
12
14
16
18
20
5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30Tonnes of freight per vehicle
98 < AUSTRALASIAN RAILWAY ASSOCIATION
appendix 2: comparing international road costing methodologies and charging regimes
EXHIBIT A2.10: NRTC AVERAGE HYPOTHECATED FUEL CHARGE AND AVOIDABLE ROAD WEAR COSTS*Cents per ESAL -km
Source: Bureau of Transport and Regional Economics, Working Paper 40: Competitive neutrality between road and rail, 1999
ESALs per vehicle
Estimated avoidable road wear cost
NRTC average fuel charge
*All heavy vehicles
0
5
10
15
20
25
30
35
0 0.5 1 1.5 2 2.5 3 3.5 4 4.5
EXHIBIT A2.11: SUMMARY OF EUROPEAN MASS-DISTANCE CHARGING INITIATIVES
Source: Forthcoming BTRE paper —based on public information
Country Date Technology Coverage Charged by Use of revenue
Distance driven, registered weight and emission class
Distance driven, no. of axles
Distance driven, no. of axles and emission class
Switzerland
Austria
Germany
United Kingdom
Jan-01
Jan-04
Jan-05
2007/8
GPS and roadside electronic system with manual option
Fully electronic (Roadside shortwave radio)
GPS based electronic system with manual option
Likely to be GPS based
Trucks and buses with GVM >3.5t (all roads)
Trucks and buses with GVM >3.5t (motorways and certain expressways)
Trucks with GVM 12t+ (motorways)
Trucks with GVM >3.5t
Transport infrastructure, including public transport (Rail)
To maintain and develop the road infrastructure
Transport infrastructure
Revenue neutral
THE FUTURE FOR FREIGHT > 99
appendix 3: national economic benefits of cost savings on inter-capital rail freight
Executive Summary 100
Conclusions 100
Basis For Conclusions 101
1. Focus Of Report 101
2. Rail Freight Cost Advantages: Overview 102
2.1 The Magnitude Of Estimated Rail Cost Advantages 102
2.2 The Sources Of Estimated Rail Cost Savings 102
2.3 The Distribution Of Estimated Rail Cost Savings 103
3. Specifying The Modelling Economc ‘Shock’ 104
3.1 The Need To Quantify An Economic Shock 104
3.2 Road To Rail: Modal Shift Assumptions 104
3.3 Estimated Operator Cost Savings (‘Above Rail’) 104
3.4 Estimated Infrastructure Cost Savings (‘Below Rail’) 105
3.5 Impact Of Subsidy Arrangements 105
3.6 Industry Allocation Of ‘Non-externality’ Benefits 105
3.7 Estimated ‘Externality’ Cost Savings 106
4. Modelling Analysis 107
4.1 The General Modelling Approach 107
4.2 Market And Other Closure Assumptions 107
4.3 Closure Assumptions Used In This Report 107
4.4 Modelling Results 108
4.5 Net Present Value Estimates Of Economic Benefits 109
5. Conclusions 109
5.1 National Benefits From Rail Cost Savings 109
Attachments 110
Attachment A—Detailed Modelling Of Inter-capital Rail Freight Changes 110
Attachment B—The AE-CGE Model 113
Attachment C—Measurement Of Economic Welfare 114
A report prepared by Access Economics Pty Limited for Port Jackson Partners Limited December 2004
100 < AUSTRALASIAN RAILWAY ASSOCIATION
Executive Summary
This report has been commissioned by Port Jackson
Partners Limited (PJPL) and prepared by Access
Economics.
PJPL has requested Access Economics to prepare an
independent analysis of the national economic benefits
likely to flow from direct cost savings to inter-capital freight
estimated to be obtainable from a substantial modal switch
from road to rail, as the combined result of:
> the achievement of efficient costs on the North South
corridor, through the realisation of available cost
reductions by the Australian Rail Track Corporation
(ARTC);
> removal of access pricing uncertainty to provide
investment certainty; and
> improved rail industry performance more generally,
including better: customer service, vertical
coordination, innovation, reliability and transit times.
The focus of this report is on the net economy-wide
economic effects—concentrating on real output, real
consumption (as a proxy for economic welfare effects),
and employment—that could be expected to emerge in
the longer term from this modal switch as a result of these
inter-capital rail freight reforms.
In addition, Access Economics has been asked to estimate
the net present value to the economy of these benefits
flowing from sustained cost savings of the magnitude
estimated in the PJPL report.1
Conclusions
Annual Real National Benefits, 2014 (In 2004 Dollars)
The national economic benefits from cost savings derived
from a substantial modal shift for inter-capital freight from
road to rail are large:
> The PJPL report estimates cost savings in 2014 of
about $393 million per annum, (including about
$308 million per annum in cost savings excluding
‘externality’ benefits).
> On the basis of Access Economics’ preferred modelling
assumptions, these annual savings in 2014 would be
reflected in increased real GDP of around $1.2 billion,
(or around $1.1 billion excluding ‘externality’ benefits).
> There would be corresponding increases in real
consumption (used as a proxy for increased
community welfare or living standards) of about $800
million, (or around $650 million excluding ‘externality’
benefits).
> There would also be modest employment gains of
about 2,500 jobs, (or 2,200 jobs excluding ‘externality’
benefits).
Net Present Value (NPV) Of Real National Benefits, 2014
(In 2004 Dollars)
Assuming that rail freight maintains its market share on
the East-West corridor after 2014, that there is no further
growth on the North-South corridor after 2014, adopting
the PJPL assumption of a 7% real discount rate, and using
Access Economics’ preferred modelling assumptions:
> The 2014 NPV for real GDP is around $27 billion,
(or $24 billion excluding ‘externality’ cost savings).
> The 2014 NPV for real consumption spending is
$18 billion, (or $15 billion excluding externality
cost savings).
100 < AUSTRALASIAN RAILWAY ASSOCIATION
appendix 3: national economic benefits of cost savings on inter-capital rail freight
1 The Future for Freight—economic analysis of the cost of moving freight on the inter-capital city corridors, Report prepared for the Australasian Railway Association by Port Jackson Partners Limited, December 2004 (hereafter referred to as the PJPL report).
THE FUTURE FOR FREIGHT > 101
Basis for Conclusions
Access Economics’ conclusions are based on:
> Estimates of cost savings to direct users of inter-capital
city rail freight services arising from a substantial future
modal switch away from road, as presented in the
PJPL report to which reference is made above.
> Estimates of cost savings from reduced ‘externality’
damage arising from use of road freight services (in
the form of improved health, reduced accidents,
congestion and noise, and reduced greenhouse gas
emissions) as presented in the PJPL report to which
reference is made above. These do not appear as
direct savings to inter-capital city freight customers,
but as community-wide savings via lower expenditure
on health, time savings, etc.
> Application of these estimated cost savings to Access
Economics’ AE-CGE computable general equilibrium
model to provide real 2014 ‘snapshots’ of the long run
real national economic benefits (and corresponding
NPV values in that year).
1. Focus Of Report
This report has been commissioned by Port Jackson
Partners Pty Ltd (PJPL) and prepared by Access
Economics.
PJPL has requested Access Economics to prepare an
independent analysis of the national economic benefits
likely to flow from direct cost savings to the national freight
task, as a result of a significant modal shift of inter-capital
freight from road to rail. The cost savings are estimated to
be obtainable as a result of the combined result of reforms
including:
> The achievement of efficient costs on the North South
corridor, through the realisation of available cost
reductions by the Australian Rail Track Corporation
(ARTC)
> Removal of access pricing uncertainty to provide
investment certainty
> improved rail industry performance more generally,
including better: customer service, vertical
coordination, innovation, reliability and transit times.2
The focus of this report is on the net economy-wide
economic effects—on real output, real consumption (as a
proxy for economic welfare effects), and employment—that
could be expected to emerge in the longer term as a
result these inter-capital rail freight reforms. In addition,
Access Economics has been asked to estimate the net
present value to the economy of these benefits flowing from
sustained cost savings of the magnitude estimated in the
PJPL report.
The rest of this report is organised as follows:
> Section 2 summarises the sources of the inter-capital
rail freight cost savings as estimated in the PJPL
report.
> Section 3 sets out how Access Economics has
specified the economic ‘shock’ arising from these cost
savings in a form appropriate for application in its
AE-CGE general equilibrium model of the Australian
economy.
> Section 4 outlines Access Economics’ modelling
approach, and market and other closure assumptions.
It presents Access Economics’ detailed modelling
results and our NPV estimates of the economy-wide
benefits of sustained cost savings of the magnitude
estimated in the PJPL report.
> Section 5 summarises Access Economics’ conclusions,
given the cost savings estimated in the PJPL report.
> More detailed material is presented in Attachments to
the full report.
THE FUTURE FOR FREIGHT > 101
appendix 3: national economic benefits of cost savings on inter-capital rail freight
2 See The Future for Freight—economic analysis of the cost of moving freight on the inter-capital city corridors, Report prepared for the Australasian Railway Association by Port Jackson Partners Limited, December 2004 (hereafter referred to as the PJPL report).
102 < AUSTRALASIAN RAILWAY ASSOCIATION
2. Rail Freight Cost Advantages: Overview
The PJPL report identifies a variety of cost advantages
and additional cost savings that currently exist or could
be realised for inter-capital rail freight services relative to
road services.
This section of this report summarises the magnitude, the
broad sources, and the distribution of these cost savings
assumed for purposes of our modelling of economy-wide
net effects.3
2.1 The Magnitude Of Estimated Rail Cost advantages
As stated in the PJPL report, with rail reform:
> On average across all corridors, and including
‘externalities’, the average cost of inter-capital
city rail freight in 2014 should be around $28 per
thousand net tonne kilometres (ntk) below that of
road (expressed in 2004 dollars)
> This equates to average rail freight costs around 43%
below average road costs across all corridors in 2014
> When applied to the estimated 14 billion ntk of
additional freight that could be carried by rail in ten
years’ time, this would result in a saving of $393
million per annum in 2014 (expressed in 2004
dollars).
2.2 The Sources Of Estimated Rail Cost Savings
The broad sources of the estimated cost advantage of
inter-capital rail freight, relative to road freight in 2014,
are roughly as follows:
> Operator costs (above rail)—current: $222 million
> Operator costs (above rail)—capital: $75 million
> Infrastructure costs (below rail)—current: $124 million
> Infrastructure costs (below rail)—capital: -$8 million
> Change in financial subsidy: -$105 million
> ‘Externality’ benefits: estimated at about $85 million
in 2014. (This is not a direct saving to rail freight
customers. By definition, as an externality, it accrues
as broader cost savings for the economy as a whole.)
> Total benefit: $393 million per annum in 2014
(of which about $308 million is non-externality
cost savings).
In particular, the cost advantages for inter-capital rail freight
relative to road freight were derived by PJPL by considering:
> Above road/rail operating costs (ie, the costs of
operating trucks versus trains)
> Above road/rail capital costs (ie, the cost of investing
in trucks and trains to meet forecast freight demand
growth)
> Below road/rail operating costs (ie, the operating costs
of road and track to meet forecast freight demand
growth)
> Below road/rail capital costs (ie, the cost of providing
the necessary road/track infrastructure to meet forecast
freight demand growth. The negative figure of -$8
million reflects the fact that rail is at a slight cost
disadvantage in this category
> Financial subsidy arrangements. There is a decrease in
the overall level of government subsidy to inter-capital
freight of $105 million. In particular, the subsidy to rail
is reduced per ntk as increased revenues accrue to
rail operators with reform, due to cost reductions and
volume growth
> Estimated ‘externality’ effects, including reduced costs
of accidents, reduced congestion (time savings),
improved health (reduced air pollution and noise), and
reduced greenhouse gas emissions.
The estimated cost savings are largest in respect of above
rail operating costs. Geographically, the savings are larger
on the longer East-West corridors than the North-South
corridors (but it is also the case that a much higher
proportion of these are already captured on the East-West
corridor than on the North-South corridor).
102 < AUSTRALASIAN RAILWAY ASSOCIATION
appendix 3: national economic benefits of cost savings on inter-capital rail freight
3 See PJPL report, Chapter 2 and related appendices for full details.
The cost savings also reflect:
> Assumed growth over the next ten years in the overall
inter-capital city freight task (reflecting a BTRE
judgement that average annual growth in demand
will be 4.5% per annum, a growth rate that has
been observed in the past, but which is about 1.5
percentage points per annum faster than growth
projections for the economy as a whole).
> A substantial modal shift in favour of rail and away
from road, especially on the North-South corridor,
driven by the inland rail investment in relation to that
corridor, and better service quality, exploiting the
underlying cost advantage for rail. Additional cost
savings described above also drive an economies
of scale ‘virtuous circle’.
> These demand growth and modal switch assumptions
result in an additional 14 billion ntk of inter-capital city
freight carried on rail by 2014.
2.3 The Distribution Of Estimated Rail Cost savings
For the purposes of the general equilibrium modelling
described in the remainder of this report, Access
Economics makes the standard long term assumption that,
like increases or reductions in taxation, the cost advantages
estimated in the PJPL report are fully forward shifted to the
immediate users of inter-capital rail freight services.
In turn, the immediate users of inter-capital rail freight
services pass these cost savings on to their customers, and
so on down the value chain. Ultimately, both directly and
directly, these cost savings flow on in the form of lower
freight service prices to all components of final demand,
including Australian household consumption, investment,
imports, and government demand; as well as production,
(which includes all of the above, plus exports less imports).
This assumption is based on the proposition that the market
for inter-capital freight services, which is defined to include
all feasible transport modes, is both competitive and
contestable. It also assumes that, on average, downstream
markets are similarly competitive and contestable.
It is also assumed for modelling purposes that inter-capital
road freight costs are not amenable to further significant
reductions (inter alia because the PJPL cost scenarios for
road freight already assume 100% use of B-double trucks).
In addition, the PJPL base case assumes that the effective
cross-subsidy of road freight by other road users (arising
because, while road funding is ‘pay-as-you-go’ by taxpayers
and road users in total, road freight does not pay anything
like its full share of road maintenance/repair costs) remains
in place. Indeed, road freight costs per ntk are assumed
to rise after the modal shift to rail, reflecting a loss of
economies of scale.
One might question whether the cost savings associated
with greater use of rail will be passed on to customers,
or instead be retained by the rail freight businesses.
On this matter, Access Economics would contend that:
> There are overwhelming grounds for concluding that
substantial efficiency advantages—and therefore
cost savings to users—are both available now and/
or realisable in future in relation to inter-capital rail
freight. Provided service quality can be improved
substantially, these can be exploited by customers.
Apart from political resistance to the reforms proposed
in the PJPL report, these gains are still relatively ‘low-
hanging fruit’ in the microeconomic reform context.
> The nature of the inter-capital rail freight production
function, with high fixed costs in relation to ‘below rail’
infrastructure, in particular, suggests economies of
scale are a significant component of these potential
efficiency gains.
> Access Economics concludes that the strong, still-
subsidised, and ongoing competition from road freight
services, in particular, will be sufficient to ensure that
most if not all of the estimated cost advantages will be
passed on to inter-capital rail freight customers.
THE FUTURE FOR FREIGHT > 103
appendix 3: national economic benefits of cost savings on inter-capital rail freight
3. Specifying The Modelling Economc ‘Shock’
3.1 The Need To Quantify An Economic Shock
National estimates of inter-capital rail reform are estimated
using Access Economics’ AE-CGE model. This requires
the specification of a ‘shock’ to the status quo to drive
the modelled economy to a new equilibrium position. The
difference between the ‘before’ and ‘after’ model solutions
represents the long term costs or benefits of the ‘shock’ in
terms of conventional national accounting variables (ie, real
GDP, real consumption, employment).
The rest of this section of the report addresses this task.
3.2 Road To Rail: Modal Shift Assumptions
The driving force behind the modelling is the estimated cost
advantage of $393 million per annum in 2014, associated
with the modal shift of inter-capital city freight towards rail,
as presented in the PJPL report. As outlined in the PJPL
report, the modal shift is predicated on a robust reform
program to improve the performance of rail and prevent
what otherwise is expected to be a continuing decline in
market share.
The cost advantage associated with this modal shift is
relative to PJPL’s forecast 2014 ‘business as usual’ (BAU)
scenario. Under the BAU scenario, overall growth in the
volume of inter-capital freight (road and rail) is assumed
to be 4.5 per cent per annum between 2004 and 2014,
consistent with BTRE forecasts.
The growth of rail freight between 2004 and 2014 is also
based on BTRE forecasts. However, where the BTRE have
forecast falling absolute volumes on some short corridors,
the PJPL analysis has held volumes constant.4 Overall,
while the BTRE forecasts rail growth of 2.7 per cent per
annum over the period, the BAU scenario (the base against
which the benefits of reform are measured) incorporates
growth of around 3 per cent per annum.
While rail volume increases in absolute terms, there is
a continuing decline in rail’s share of inter-capital city
freight task, from around 35 per cent now, to around 30
per cent in 2014. With the economy’s greater reliance on
more expensive road freight under this ‘business as usual’
scenario, the cost of inter-capital freight per ntk increases
slightly relative to 2004.
Consistent with the PJPL report, this scenario is contrasted
with the case where a robust reform program for rail
is implemented, inducing a shift of around 14 billion
ntks from road to rail by 2014. This more than doubles
the volume of rail freight between 2004 and 2014 and
increases rail’s market share on a volume basis in 2014
from 30 per cent to 50 per cent.
Under ‘rail reform’ after 2014, it is assumed that rail
maintains its share of the inter-capital city freight task on
East-West corridors (growing at the economy-wide rate of
3% per annum). However, from 2014 onwards, capacity
constraints on the North-South corridors mean that the
overall volume of freight on these corridors is maintained in
absolute terms. Similarly, in the BAU scenario, net volume
in absolute terms is held constant on the overall North-
South corridors past 2014. This compares to the BTRE
assumption that rail freight on the North-South corridors will
decline in absolute terms.
Cost savings flow to the economy from the greater reliance
on rail freight, with this less expensive form of transport
reducing the cost of freight per ntk relative to the ‘business
as usual’ scenario.
As noted earlier, the basis for rail’s cost advantage
compared to road includes that associated with ‘non-
externality’ benefits (rail operations and rail infrastructure)
plus lower externalities (greenhouse, health, congestion and
time benefits).
3.3 Estimated Operator Cost Savings (‘above rail’)
The PJPL report provides details of the cost advantage of
rail over road at the operator level. In particular, the report
compares road transport costs of a 38 net tonne b-double
(such as fuel, tyres, driver costs, etc), with similar rail
operator costs (fuel, crews, maintenance, etc).
104 < AUSTRALASIAN RAILWAY ASSOCIATION
appendix 3: national economic benefits of cost savings on inter-capital rail freight
4 PJPL report (page 73).
In 2014, the estimated freight cost advantage of ‘above rail’
over ‘above road’ presented in the PJPL report amounts to
around $297 million per annum.
3.4 Estimated Infrastructure Cost Savings (‘below rail’)
Similarly, the PJPL report provides details of the estimated
infrastructure costs of rail and road (‘below rail’). Significant
cost reductions are assumed on the North/South corridor,
consistent with the efficiencies expected from the merger of
the Rail Infrastructure Corporation (RIC) and the Australian
Rail Track Corporation (ARTC).
Despite a slight cost disadvantage in terms of below rail
capital costs compared to below road, the cost advantage of
below rail overall amounts to around $116 million.
Excluding externalities and changes to subsidy levels, the
value of rail’s raw cost advantage over road is around $413
million in 2014 (expressed in 2004 dollars).
3.5 Impact of Subsidy Arrangements
Our modelling also incorporates the impact of Government
financial subsidy arrangements. The rate of subsidy to rail
reduces as rail operators benefit from higher revenues, due
to cost reductions and the greater volume of freight carried
on rail.
Financial subsidies are assumed to be ‘fully passed forward’
thereby reducing the price paid by freight users. Both the
reduction in industry costs and the lower subsidy per ntk is
reflected in the price of rail freight. That is, while the 2014
rail reform scenario assumes the rail industry achieves
efficiencies that reduce costs, some of this cost reduction
is offset by a lower government subsidy per ntk which also
feeds into prices.
As such, comparing the ‘business as usual’ and ‘rail
reform’ scenarios, while there is a reduction in the price of
rail freight to end users, a large part of the modal shift is
driven by non-price factors such as improved rail industry
performance and customer service.
While the lowering of subsidy rates is a cost to freight
users, it is a saving to Government of some $105 million.
As well as incorporating the impact of the subsidy on
freight prices throughout the economy, AE-CGE takes into
account the reduced cost to the economy of raising taxation
(deadweight loss) to fund the subsidy, in order to maintain
Budget balance.
3.6 Industry Allocation Of ‘Non-Externality’ Benefits
In estimating the economy-wide effects of inter-capital rail
reform, the value of inter-capital road and rail freight used
was allocated across industries.
The latest available ABS Input-Output (IO) data for Australia
is for 1998-99. We have derived a more up-to-date IO table
(47 products by 47 industries) using the ABS 2000-01
supply-use tables.
In this study, inter-capital freight is restricted to inter-capital
movements of containers, usually referred to as inter-modal
freight, and for the most part it excludes (i) bulk and non-
container loads, and (ii) freight moved along inter-capital
routes but only part of the way between capitals. Inter-
capital road freight accounts for only a small proportion
of total road services. Road services also include activities
such as bus and taxi transport, local deliveries, rural freight
services, and the transportation of bulk commodities.
Inter-capital rail services similarly account for a small
proportion of total rail services, which also includes
activities such as passenger trains, rural and bulk services.
In part guided by data provided by PJPL, we have split
the IO-based road transport industry into inter-capital
freight and ‘other’ freight, and similarly the IO-based rail
and pipeline transport industry has been split into inter-
capital freight and ‘other’. Inter-capital freight has then
been allocated across industries in accordance with the
supply-use tables and estimates provided by PJPL.
Major users of inter-capital freight include the food and
beverage, construction and wholesale industries.
THE FUTURE FOR FREIGHT > 105
appendix 3: national economic benefits of cost savings on inter-capital rail freight
3.7 Estimated ‘Externality’ Cost Savings
As noted in section 2 above, the PJPL report includes
estimated ‘externality’ benefits in the form of reduced
economic costs/damage/health problems.5 These are
estimated at around $85 million in 2014 (or around $6 per
thousand ntks).
The estimated benefits from a modal shift to rail comprise:
> Reduced accidents and health-related problems (less
noise, pollution, etc.).
> Reduced congestion on roads (with time savings, etc).
> Reduced greenhouse gas-related benefits.
These do not represent benefits that can be privately
captured by the inter-capital rail freight industry. But they
are potentially significant economic and social benefits to
the economy more broadly.
Modelling these benefits in a quantitative fashion in
the conventional general equilibrium/national accounts
framework is difficult. For example:
> This national accounts framework may be relatively
good at recording and measuring gross flows where
market transactions are involved (akin to a P&L
statement).
> But it does not account very well for all net flows or
for balance sheet effects.
> For example spending on health adds to GDP, even
though that spending is largely preventing or correcting
health damage (resulting at best in no change in the
economy’s health balance sheet, or, as often, simply
reducing the extent of deterioration in health).
> Spending involved in cleaning an oil spill adds to GDP,
even though the environmental balance sheet, at best,
is simply restored to its former state by such activity.
> Even the treatment of insurance claims is badly
handled. Replacement of a car that has been ‘totalled’
with another new car is recorded as increased
production in the period in question, even though the
stock of cars has not changed. (Actually, the solution
to this defect is ridiculously simple, but it has not been
implemented. However, that’s another story.)
Accordingly, for modelling purposes, Access Economics
has treated the externality benefits in an indicative manner
to provide a broad first approximation of their impact, given
the defects of the national accounts framework (and, almost
certainly, the sizeable error margins around the estimated
value of externality benefits as well).
The treatment used in this report is as follows:
> Given that traffic accidents dominate the value of
externality benefits, the cost saving is assumed to
have the effect of lowering the value of expenditure on
health services.
> About two thirds of that saving is assumed to represent
a cost saving to the public sector, and one third to
represent a cost saving to the private sector (roughly
reflecting the proportions of expenditure on health by
both sectors at present).
> For the public sector, the lower expenditure on
health generates a small public sector surplus that,
under the budget closure assumption used (see
section 4.2 below) generates a small reduction in
personal and company income tax rates. This has the
effect of increasing real consumption by increasing
disposable income.
> For the private sector component, the lower health
expenditure is reallocated to all goods and services,
in line with consumer preferences.
> The first of these effects increases real disposable
incomes, which finances increased real consumption
and, to some extent (depending on import leakages),
real GDP.
106 < AUSTRALASIAN RAILWAY ASSOCIATION
appendix 3: national economic benefits of cost savings on inter-capital rail freight
5 See Appendix 1 of the PJPL report for more details regarding the externality estimates.
> The second allows minor efficiency gains from
greater choice as well, also at the margin lifting real
consumption and real GDP.
Access Economics acknowledges that this is an
approximate and simplistic treatment of much more
complex effects, but it is the most feasible approach given
the limitations of general equilibrium models and the
underlying national accounts/input-output data.
To the extent that health care, for example, following a
traffic accident, can be regarded as largely compensating
individuals for their loss of well-being or health, this
approach could be regarded as a reasonable approximation
of the welfare benefits of reduced health expenditure,
which frees up funds for expenditure elsewhere in line
with consumer preferences. (It also highlights the inherent
defects of the purely ‘flow’ expenditure-based analysis
captured by the national accounts as presently structured—
and, necessarily, reflected in our AE-CGE model.)
4. Modelling Analysis
4.1 The General Modelling Approach
4.1.1 The Long Run Modelling Approach
Starting from an initial equilibrium solution, the AE-CGE
Model computes a new equilibrium solution as a result of
changes applied to the model. The values in the model
database correspond to annual flows.
4.1.2 The Cost-Savings ‘Shock’
The cost saving ‘shock’ of $393 million in 2014 is the
product of:
> the estimated modal shift of around 14 billion ntks
by 2014 to rail, above the BAU benchmark; and
> rail’s underlying cost advantage over road (around $28
per thousand ntk including externalities).
The objective of the exercise is to determine the difference
between ‘rail reform’ and ‘business as usual’ in 2014,
with emphasis on ongoing improvements in real GDP and
real consumption. We have taken the 2014 result as a
reasonable proxy for the long run equilibrium in our CGE
model. For our purposes any economy wide improvements
during the transition period from 2004 to 2014 have
been ignored.
Non-externality cost savings of $308 million have been
allocated to freight users as discussed in section 3.6 above.
The remaining $85 million has been assumed to reduce
health care expenditure, as discussed in section 3.7 above.
4.2 Market And Other Closure Assumptions
Important standard assumptions underlying long run
general equilibrium modelling results include the following:
> The exchange rate for the $A adjusts to maintain a
fixed balance of trade.
> Income tax rates for individuals and businesses adjust
to ensure that government revenue equals government
current expenditure plus transfers. The public sector
budget balance is fixed.
> In the more extreme long term formulations of the
AE-CGE model, aggregate employment is held constant
with wages adjusting to clear the labour market. More
often, we also allow a small labour supply response to
higher wages (a positive labour supply elasticity with
respect to real wages of 0.2).
4.3 Closure Assumptions Used In This Report
Two different sets of closure assumptions were modelled
as follows:
> Run #1. Fixed trade balance, fixed budget balance,
markets clear, fixed labour supply.
> Run #2. Fixed trade balance, fixed budget balance,
markets clear, +0.2 labour supply elasticity.
Access Economics prefers the second of these model runs,
including in the context of this report, because in the long
run we would expect some responsiveness of labour supply
to real wage changes.
THE FUTURE FOR FREIGHT > 107
appendix 3: national economic benefits of cost savings on inter-capital rail freight
These runs provide the results that are reported below.
4.4 Modelling Results
The results below show that there is a considerable gain to
real GDP flowing from inter-capital rail reform, of, roughly,
between $800 million and $1.1 billion per annum by 2014,
arising from the ‘non-externality’ cost savings presented
in the PJPL report. This is to be expected, given the lower
average cost to freight users of inter-capital city freight.
Transport is a pervasive business input, with the indirect
savings flowing through the value chain to benefit final
consumers of all goods and services.
Similarly, real consumption, a standard proxy for economic
welfare, is estimated to increase by between around $440
million and $650 million per annum by 2014.
The effects of the reduction in ‘non-externality’ costs under
the two different closure assumptions are shown in Table
4.4.1 below.
The results represent the real differences in 2014 between
inter-capital rail reform and the PJPL benchmark of
‘business as usual’, with values expressed in 2004 dollars.
Table 4.4.1—‘Non-externality’ Modelling results ($2004)
Run #1 Run #2
Fixed labour supply Flexible labour supply
Fixed trade balance Fixed trade balance
Real consumption ($m) 438 647
GDP ($m) 780 1,070
Employment (persons fte) 0 2,208
NPV real consumption ($b) 10 15
NPV GDP ($b) 18 24
The results in table 4.4.2 below also include the additional
benefit to society from the estimated reduced externality
costs of $85 million. In Run#2 for example, as a result of
‘externality’ benefits, the addition to real consumption is
estimated at about $155 million and the corresponding
increment to real GDP is around $143 million. Externality
benefits are dominated by a reduction in accident costs,
and all are modelled as a reduction in health costs shared
between households and government consumption as
discussed in section 3.7.
Table 4.4.2—Modelling results Including
externalities ($2004)
Run #1 Run #2
Fixed labour supply Flexible labour supply
Fixed trade balance Fixed trade balance
Real consumption ($m) 567 802
GDP ($m) 888 1,213
Employment (persons fte) 0 2,480
NPV real consumption ($b) 13 18
NPV GDP ($b) 20 27
As noted above, on balance, and having regard for the long
term nature of general equilibrium modelling, we favour the
assumptions underpinning Run #2 (because of the flexible
labour supply assumption).
108 < AUSTRALASIAN RAILWAY ASSOCIATION
appendix 3: national economic benefits of cost savings on inter-capital rail freight
4.5 Net Present Value Estimates Of Economic Benefits
The net present value (NPV) calculations for real GDP and
real consumption set out in tables 4.4.1 and 4.4.2 above
are based on the following assumptions:
> Under ‘rail reform’ after 2014, it is assumed that rail
maintains its share of the inter-capital city freight
task on East-West corridors (growing at the economy-
wide rate of 3 per cent per annum). However, from
2014 onwards, capacity constraints on the North-
South corridors mean that the overall volume of
freight on these corridors is maintained in absolute
terms. Similarly, in the BAU scenario, net volume is
held constant on the overall North-South corridors
past 2014, but is allowed to expand on the East-
West corridors.
> Consistent with the ‘comparative static’ nature of
general equilibrium modelling, the transition path to
the estimated 2014 effects is ignored for purposes of
this report (the benefits of this path are included in the
PJPL NPV estimates, however).
> The real discount rate applied from 2014 to the future
real GDP and consumption is assumed to be 7 per
cent per annum, consistent with the PJPL report.
5. Conclusions
5.1 National Benefits from Rail Cost Savings
Annual Real National Benefits, 2014
The national economic benefits from cost savings derived
from a substantial modal shift for inter-capital freight from
road to rail are large:
> The PJPL report estimates cost savings in 2014 of
about $393 million per annum, (including about
$308 million per annum in cost savings excluding
‘externality’ benefits).
> On the basis of Access Economics’ preferred modelling
assumptions, these annual savings in 2014 would be
reflected in increased real GDP of around $1.2 billion,
(or around $1.1 billion excluding ‘externality’ benefits).
> There would be corresponding increases in real
consumption (used as a proxy for increased
community welfare or living standards) of about $800
million, (or around $650 million excluding ‘externality’
benefits).
> There would also be modest employment gains of
about 2,500 jobs, (or 2,200 jobs excluding ‘externality’
benefits).
Net Present Value (NPV) Of Real National Benefits, 2014
Assuming that rail freight maintains its market share on
the East-West route after 2014, and adopting the PJPL
assumption of a 7% real discount rate, and using Access
Economics’ preferred modelling assumptions:
> The 2014 NPV for real GDP is around $27 billion, (or
$24 billion excluding ‘externality’ cost savings).
> The 2014 NPV for real consumption spending is
$18 billion, (or $15 billion excluding externality
cost savings).
THE FUTURE FOR FREIGHT > 109
appendix 3: national economic benefits of cost savings on inter-capital rail freight
Attachments
Attachment A—Detailed modelling of Inter-capital Rail
Freight Changes
Input output (IO) data
The latest available ABS Input-Output (IO) data for Australia
is for 1998-99. We have derived a more up-to-date IO table
(47 products by 47 industries) using the ABS 2000-01
supply use tables.
The supply use data contains inter-industry and final
demand values expressed in purchasers’ prices, together
with primary inputs, Australian production, imports, and
total uses of taxes and different types of margins (mainly
wholesale, retail and transport categories).
We constructed the 2000-01 IO table at basic values
and associated full tax and margins matrices, using the
corresponding IO matrices to distribute taxes and margins
data along the rows.
The current task is to examine the effects of rail reform
between 2004 and 2014. As such, data was expressed in
2003-04 terms, based on ABS growth rates between 2000-
01 and 2003-04.
Results in the report are presented in 2014 quantities.
Growth of 3 per cent per annum has been assumed to
inflate real consumption and GDP, while employment
numbers (where labour supply is flexible) have been grown
by 1.2 per cent per annum.
Rail freight is in competition with road freight for inter-
capital services. The objective is to estimate the economy-
wide effects of modal shift towards rail, relative to BAU
scenario where the volume of rail freight grows at an
average annual rate of around 3 percent between 2004
and 2014.
In this study, inter-capital freight is restricted to inter-
capital movements of containers and for the most part it
excludes bulk and non-container loads, and freight moved
along inter-capital routes but only part of the way between
capitals. Inter-capital road freight accounts for only a small
proportion of total road services which also include bus
and taxi transport, local deliveries, rural freight services,
and bulk coal, minerals, grain and liquids. Inter-capital rail
services similarly account for a small proportion of total
rail services which also include passenger trains, rural and
other non capital services, and bulk services.
The IO road transport sector has been split into ‘inter-
capital freight’ and ‘other’, and similarly for rail and pipeline
transport. For the purpose of our modelling, inter-capital
‘above rail’ and ‘below rail’ are treated as one industry.
PJPL supplied data describing road and rail costs, road
and rail freight for 2004, and the scenarios of ‘rail reform’
in 2014 and ‘business as usual’ in 2014. The rail freight
task is measured in billion net tonne kilometres (ntk).
Multiplication by price (cents per ntk) gives the freight cost
to customers.
Modelling Scenarios
We consider two options. Under ‘rail reform’, rail increases
its share of inter-capital freight, measured in ntk, to 50
per cent by 2014. This is largely achieved by improved rail
industry performance and customer service. Given the cost
advantage of inter-capital rail over inter-capital road freight,
the overall cost of the national freight task falls with this
modal shift.
It is assumed that there is some increase in the price
of road freight per ntk, reflecting a loss of economies of
scale with declining market share. No change in road user
charges has been incorporated into the modelling.
The ‘rail reform’ scenario is compared to the ‘business as
usual’ scenario where the share of inter-capital freight,
measured in ntk, declines from 35 per cent in 2004
to 30 per cent by 2014. This is the result of the inter-
capital freight task growing at 4.5 per cent per year over
this period, while rail freight grows at the slower rate of
3 per cent.
110 < AUSTRALASIAN RAILWAY ASSOCIATION
appendix 3: national economic benefits of cost savings on inter-capital rail freight
The objective of the exercise is to determine the difference
between the ‘rail reform’ and ‘business as usual’ scenarios
in 2014, with emphasis on the improvements in GDP and
real consumption.
Model closure
The model can be run with alternative macroeconomic
closure assumptions. In particular, there is a choice
between fixed employment and flexible employment
(corresponding to assumptions of flexible or fixed wages). In
the case of flexible employment, we have assumed a labour
supply elasticity of +0.2. That is, an increase in real wages
of 1% induces an increase of 0.2% in labour supply, and
the model solution ensures that this supply of labour is fully
employed. The second choice is between a fixed exchange
rate (variable balance of trade) and a flexible exchange rate
(fixed balance of trade). Based on these variations, we have
presented a range of results in section 4.4.
Specifying the Modelling ‘Shocks’
For each simulation, the following adjustments are made
in order to specify the ‘shock’.
1. Changes in the costs of the rail and road freight
industries. PJPL has provided estimates of the
composition of rail and road costs per ntk for 2004,
‘rail reform’ in 2014 and ‘business as usual’ in 2014.
Additional capital costs associated with around
$870 million of new capital investment by the ARTC
were modelled as an annualised increase in the
cost of capital, reflected in higher Gross Operating
Surplus (GOS).
Road costs are assumed to fall slightly in the ‘business
as usual’ case and rise slightly in the ‘rail reform case’,
in response to changing volumes and economies
of scale.
The changes in freight unit costs are applied to the
relevant cost components, namely the inputs from
other industries, wages, GOS, subsidies and taxes.
2. Changes in inter-capital freight prices faced by
users (that is, users in the production and investment
sectors—we have assumed that households do not
use inter-capital freight directly). It is assumed that
cost reductions for each industry are fully passed
on as changes in output prices. The change in the
cost of the freight task is a combination of changes
in quantities of rail freight and road freight (growth
and modal shift) and changes in rail and road
freight prices.
Government subsidy arrangements create a ‘wedge’
between freight costs and prices to end users. The
Government subsidy to rail per ntk is significantly
reduced under the ‘rail reform’ scenario as rail
operators accrued greater revenue from volume
growth and cost reductions. Both cost savings and
subsidies are fully ‘passed forward’ and reflected in
end prices to users. The price of rail freight falls in
net terms after taking account of cost reductions and
subsidy arrangements.
As well as incorporating the impact of the subsidy
on freight prices throughout the economy, AE-CGE
takes into account the reduced cost to the economy of
raising taxation (deadweight loss) to fund the subsidy,
in order to maintain Budget balance.
The AE-CGE model assumes that each industry uses
road freight and rail freight in a fixed proportion to
production. It does not allow for changes in these
proportions in response to changes in relative prices.
The effects on individual industries depend on which
industries use the relevant inter-capital rail and road
freight services. PJPL provided indicative estimates
of the proportions of total inter-capital ntk accounted
for by different goods in 2004. We have used this
data together with our estimates of the IO total rail
freight and total road freight margins to estimate the
corresponding inter-capital freight matrices.
The effects on freight customers in each industry
were modelled, consistent with the total values in
Table A1.1.
THE FUTURE FOR FREIGHT > 111
appendix 3: national economic benefits of cost savings on inter-capital rail freight
3. Reduced cost of externalities. As estimated by PJPL,
these are dominated by a reduction in accident costs.
We estimate the benefits of these externalities by
imposing reductions in the cost of health to households
and government final consumption. The total
externality cost of $85 million is allocated as a saving
of around $28 million to households and $57 million to
government (in $2004).
Overall, the modelling ‘shock’ was specified to be
consistent with the specific data provided by PJPL,
summarised in table A1.1 below.
112 < AUSTRALASIAN RAILWAY ASSOCIATION
appendix 3: national economic benefits of cost savings on inter-capital rail freight
Table A.1.1—Data provided by PJPL for input into AE-CGE ($2004)
Inter-capital Inter-capital Inter-capital
rail freight road freight total freight
2004
ntk (billion) 16 30 47
sales ($m) 601 1653 2253
$/thous ntk 37 54 48
Rail Reform' 2014
ntk (billion) 36 37 73
sales ($m) 1220 1980 3200
$/thous ntk 34 54 44
Business as usual' 2014
ntk (billion) 22 51 73
sales ($m) 791 2717 3508
$/thous ntk 36 54 48
Rail reform 2014 less business as usual 2014
sales ($m) 429 -737 -308
Source: Port Jackson Partners limited
Attachment B—The AE-CGE Model
Broad Description of the AE-CGE Model
The AE-CGE model is a small, long-run, non-linear,
computable general equilibrium (CGE) model of the
Australian economy. AE-CGE was initially developed
by Access Economics for the Economic Planning and
Advisory Council, Bureau of Industry Economics, Industry
Commission and Business Council of Australia in 1992.
The standard version of AE-CGE models employment,
profit, production, consumption, investment, imports and
exports for 26 separate industries. Interactions between
industries are modelled using input-output data, a measure
of the various inputs required by each industry to produce
its output. Changes in each industry are then aggregated
to provide estimates of macroeconomic variables.
The strategy underlying the design of AE-CGE was
to construct a CGE model of manageable size where
interactions within and between industries could be
modelled in reasonable detail. At the same time, the
number of industries in AE-CGE can be increased readily
to provide necessary detail in particular applications.
AE-CGE, unlike many other CGE models (such as ORANI),
solves in levels rather than percentage deviations. This
non-linear approach maintains the complex detail of the
equations describing supply and demand. This full impact,
particularly for consumption and production technologies,
can be blurred by the linearisation commonly employed in
solving larger models.
Within the business sector of AE-CGE, profit maximising
firms are assumed to demand labour, capital and the
output of other firms, to produce output. This output is
disposed of through domestic or export markets (which are
imperfect substitutes). Production supplied to domestic
markets is combined with imports (which are imperfect
substitutes for domestic supply) to satisfy total demand.
Australia is assumed to be a price taker in import markets.
Total demand consists of private consumption, intermediate
input demand, investment and government consumption.
The model distinguishes Commonwealth and overall
state/local government sectors. For each, the government
sector imposes a series of direct and indirect taxes. In the
standard version of the model, the rates of indirect taxes
are determined from input-output data while direct tax
rates are assumed to adjust to maintain budget balance.
Governments maintain real government current expenditure
in each industry (again determined from input-output data)
irrespective of price changes.
The long-run snap-shot nature of the AE-CGE model is
reflected in the assumptions about market behaviour. In
the standard long-run closure of the model, nominal gross
national expenditure (GNE) is taken as ‘numeraire’ relative
to which other nominal variables adjust. The exchange
rate is assumed to adjust to keep the overall trade balance
unchanged. Capital and labour are assumed to be fully
mobile between sectors. The total supply of labour is
assumed to be fixed in the standard version of the model,
with the wage adjusting to equate supply and demand.
Alternative approaches—reported here—allow some labour
supply responsiveness to real wage changes. The capital
stock is assumed to be flexible, with expansion/contraction
in each industry sufficient to maintain a fixed, economy-
wide, rate of return to capital.
The current implementation of AE-CGE models the
Australian economy as reflected in ABS input-output data
but scaled up inter alia to reflect the implementation of the
New Tax System as at 1 July 2001. The 2000-01 Supply
Use tables are also used to provide more recent data on
usage patterns.
Consumption expenditure is wages and profits less the
sum of taxes (which equal government spending) and the
resources—saving—needed for gross investment. This
expenditure is allocated between the outputs of the various
industries using a Klein-Rubin (or Stone-Geary) utility
system. This system allows consumption of each industry’s
output to reflect sensitivity to changes in the industry’s
output price, as described by their own-price elasticities.
For each commodity there is a fixed or ‘autonomous’ level
of consumption and a ‘discretionary’ level. The discretionary
levels of consumption adjust, subject to the constraints
imposed by the model, so as to maximise utility.
THE FUTURE FOR FREIGHT > 113
appendix 3: national economic benefits of cost savings on inter-capital rail freight
A more detailed description of the model is available
in Access Economics Computable General Equilibrium
(AE-CGE) Model Documentation.
Attachment C—Measurement Of Economic Welfare
Standard Measures of Economic Welfare:
A Short-Cut Summary
Measures of economic welfare are commonly used concepts
in economic analysis. Their precise definition is somewhat
complicated for non-economists. As a practical, observable
approximation, total household spending on consumption
of goods and services is a reasonable approximation to
economy-wide economic welfare.
For modelling purposes in this report, the net change in
Australian consumer spending is a good summary measure
of the change in welfare caused by modal shift from road
to rail.
Standard Measures of Economic Welfare:
A Little More Detail
The AE-CGE model generates estimates of a wide range
of aggregate and sectoral variables − covering production,
incomes, expenditure, trade and prices. An important issue
is how to measure the improvement in economic welfare
that results from a change, such as a shift to more cost-
effective freight transport.
A general equilibrium model, such as AE-CGE, has markets
for all the goods and services in the economy. It also
contains an explicit household utility function, allowing us to
estimate the utility associated with particular consumption
bundles. We are therefore in a position to provide more
sophisticated estimates of the overall impact of a tax
change on economic welfare—based on the ‘compensation
principle’ 6.
In practice in the AE-CGE model, welfare measures based on
the compensation principle are usually close, numerically, to
the change in aggregate real consumption caused by a shock
to the model.
It is this latter measure that is emphasised in this report.
114 < AUSTRALASIAN RAILWAY ASSOCIATION
appendix 3: national economic benefits of cost savings on inter-capital rail freight
6 See for example Johansson, P-O, An introduction to modern welfare economics, Cambridge University Press, 1991. Two closely related welfare measures based on the compensation principle are the Compensating Variation and the Equivalent Variation. The former is the amount of money that consumers would have to be given (or pay) after the change, to keep them at the same level of utility in the event that the change were subsequently reversed. The latter is the amount of money that consumers would have to be given before the change that would make them as well off as they would be if the change were in fact to proceed.
THE FUTURE FOR FREIGHT > 115
ARTC, Annual Report, 2002
BIS Shrapnel, Freight in Australia 1999-2004, 1999
Booz-Allen and Hamilton, ARTC Interstate Rail Network Audit, 2001
Bruzelius N, Measuring the Marginal Cost of Road Use—An International Survey, 2003
Bureau of Transport and Regional Economics, Electronic Toll Collection: EU updates, 2003
Bureau of Transport and Regional Economics, Information Sheet 22: Freight between Australian Capital Cities, 2003
Bureau of Transport and Regional Economics, Working Paper 109: Rail Infrastructure Pricing, 2003
Bureau of Transport and Regional Economics, Working Paper 57: Land Transport Infrastructure Pricing, 2003
Bureau of Transport Economics, Working Paper 35: Roads 2020, 1997
Bureau of Transport Economics, Working Paper 40: Competitive Neutrality between Road and Rail, 1999
Button K, Internalising the social costs of transport, 2003 (Paper to the ECMT)
Cambridge Systematics, Freight Trends and Freight Rail, 2002 (Presentation)
Canada Transportation Act Review Panel, Vision and Balance: Report of the Canada Transportation Act Review Panel, 2001
Department of Transport and Regional Services, Auslink Green Paper, 2002
Department of Transport and Regional Services, Auslink White Paper, 2004
German Institute for Economic Research, Estimation of Infrastructure Costs, 2003
Independent Review of RIC Metropolitan Maintenance Funding, 2002
IPART determination for the Regulation of NSW Electricity Distribution Networks
Laird P., Land Freight External Costs in Queensland, 2002 (for Queensland Transport)
National Road Transport Commission, 3rd Heavy Vehicle Road Pricing Determination Issues Paper, 2003
National Road Transport Commission, Technical Report: Updating Heavy Vehicle Charges, 1998
National Transport Secretariat, Strategic Freight Corridors Assessments, 2001
Perkins S, Recent Developments in Road Pricing Policies in Western Europe, 2002
Queensland Competition Authority, Decision on QR’s 2001 Draft Access Undertaking, 2001
Sinclair Knight Merz, ACT Passenger Transport Study, 2003
UK Department of Transport, NERA report on lorry track and environment costs, 2000
US Federal Highway Cost Allocation Study, 1998
appendix 4: bibliography
116 < AUSTRALASIAN RAILWAY ASSOCIATION
ARTC
Australasian Rail Track Corporation Ltd
Average cost
Total cost per unit of traffic load of providing the highway
Avoidable cost
Costs that would not be incurred in the absence of traffic
loading (Also referred to as ‘variable’ cost or ‘marginal’
cost).
B-double
Truck/trailer combination consisting of a prime mover
towing two semi-trailers.
BTRE
Bureau of Transport and Regional Economics (Formerly
the BTCE and BTE). The transport research arm of the
Department of Transport and Regional Services.
ESAL
Equivalent Standard Axle Load. Calculated for each
axle of a heavy vehicle as (Actual Axle load/Reference
load)4 and then summed to give an ESAL figure for the
vehicle. Reference load used in Australian calculations is
8.2 tonnes. Actual loads are adjusted to take account of
differences between axle configurations.
GPS
Global Positioning System. A system that uses
geostationary satellites to accurately determine the
position of a receiver unit by comparing signals from 3
or more satellites.
GTK
Gross Tonne Kilometre. NTK plus mass of vehicle used to
haul freight.
GVM
Gross vehicle mass. The combined mass of the vehicle
and any freight carried. Average Gross Mass (AGM) is often
used to allow for the fact that trucks will not be fully laden
on all trips.
HMD4
Highway Development and Management System
developed by the International Study of Highway
Development and Management Tools (ISOHDM). Uses
detailed engineering and financial models to plan and
optimise road management activities and costs for given
traffic loadings. Used by the World Bank, as well as
highways agencies and transport economists in a large
number of countries.
ISOHDM
International Study of Highway Development and
Management Tools—the organisation responsible for the
development of the HDM4 Highway Development and
Management System, used by the World Bank.
Marginal cost
The costs associated with an incremental unit of traffic
load. Excludes costs due to weathering, fixed costs, etc.
(Also referred to as ‘variable’ cost or ‘avoidable’ cost).
NRTC
National Road Transport Commission (now the NTC).
Responsible for recommending heavy vehicle charges
through the Heavy Vehicle Pricing Determination
(3rd Determination is currently under development).
ntk
Net Tonne Kilometre. Standard unit of freight task.
Equivalent to transporting 1 net tonne of freight a distance
of 1 kilometre. Used because costs increase with both
freight mass and distance travelled.
PCU
Passenger Car Unit. A measure of the ‘footprint’ of a
vehicle. 1 car = 1PCU. Trucks are considered to be
equivalent to 2-4 passenger vehicles depending on
their size.
appendix 5: glossary of terms
THE FUTURE FOR FREIGHT > 117
PMS
Pavement Management System. A combination of
engineering and economic models used to predict
required road maintenance and upgrade activities and the
associated costs. The World Bank/ISOHDM model, HDM4,
is one example.
PUD
Pick-up-and delivery costs. The cost of moving freight from
source to rail head and from rail head to final destination.
Regression analysis
A statistical technique that seeks to explain an outcome
(dependent) variable in terms of multiple predictor
(independent) variables. This analysis reveals the nature
and strength of the relationship between each predictor
variable and the outcome, independent of the influence
from all other predictors. The term typically refers to
Ordinary Least Squares (OLS) regression, which models
a linear relationship among variables.
TEU
Twenty Foot Equivalent unit. The standard unit of freight
volume. 1 TEU corresponds to a standard 6.1m container,
thus a 12.2m container is 2 TEU.
Variable cost
A cost that varies directly with road use, such as damage
caused to pavements by vehicles (Also referred to as
‘marginal’ or ‘avoidable’ cost).
VKT
Vehicle Kilometres Travelled
appendix 5: glossary of terms
>05