the future for freight|2005> - home | ara · 2015-04-23 · intro para the future for...

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



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.


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.



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


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


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?


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.





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.


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.



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.





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.





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.





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.



4 Laird P., Land Freight External Costs in Queensland, 2002



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


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.


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.



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.



5 Department of Transport and Regional Services, AusLink Green Paper, 2002



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.



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



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



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.



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.



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.



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.





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.


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.



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.



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.



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.



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.





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



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



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



’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



Benefit on existing volumes:

Incremental cost savings = $8/'000 ntk

Existing task = 16.5b ntk



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.



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.



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.





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;



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).


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.



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.



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


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.


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.>


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.



$ per '000 ntkEXHIBIT A1.1: TOTAL COST COMPARISON—POST RIC COST REDUCTION


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


*Assumes 50% reduction in RIC’s costs and 100% growth in RIC’s intermodal volume

**Across 2014 volume shifted to Rail of 14b ntks


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


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.





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.



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.



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).



19 National Road Transport Commission, Updating Heavy Vehicles Charges: Technical Report, September 1998



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


- 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


EXHIBIT A1.7: BELOW ROAD—COMPARISON OF COST ESTIMATES



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.



20 Bureau of Transport and Regional Economics, Working Paper 35, Roads 2020, 1997



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



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.



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)



$ 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


1.10



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



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).



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.


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.





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



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.



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.



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.





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.



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



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


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


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).



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).


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.



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).


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).



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.



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).



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.



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.



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.



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).



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).



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.



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.



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.



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


sales ($m) 1220 1980 3200


Business as usual' 2014


sales ($m) 791 2717 3508


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.



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.



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.


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


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


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

the future for freight|2005> - home | ara · 2015-04-23 · intro para the future for...

Documents