Development of Regional Airports

WITPRESS
WIT Press publishes leading books in Science and Technology.
Visit our website for the current list of titles.
www.witpress.com

WITeLibrary
Home of the Transactions of the Wessex Institute, the WIT electronic-library
provides the international scientific community with immediate and permanent
access to individual papers presented at WIT conferences. Visit the WIT eLibrary
at http://library.witpress.com
This page intentionally left blank
Development of Regional Airports

Theoretical Analyses and Case Studies

Edited by:

M.N. Postorino
University of Reggio Calabria, Italy
Editor: M.N. Postorino
University of Reggio Calabria, Italy

Published by
WIT Press
Ashurst Lodge, Ashurst, Southampton, SO40 7AA, UK
Tel: 44 (0) 238 029 3223; Fax: 44 (0) 238 029 2853
E-Mail: [email protected]
http://www.witpress.com

For USA, Canada and Mexico

WIT Press
25 Bridge Street, Billerica, MA 01821, USA
Tel: 978 667 5841; Fax: 978 667 7582
E-Mail: [email protected]
http://www.witpress.com

British Library Cataloguing-in-Publication Data

A Catalogue record for this book is available
from the British Library
ISBN: 978-1-84564-143-6
Library of Congress Catalog Card Number: 2009926709

The texts of the papers in this volume were set

individually by the authors or under their supervision.

No responsibility is assumed by the Publisher, the Editors and Authors for any injury
and/or damage to persons or property as a matter of products liability, negligence or
otherwise, or from any use or operation of any methods, products, instructions or
ideas contained in the material herein. The Publisher does not necessarily endorse
the ideas held, or views expressed by the Editors or Authors of the material contained
in its publications.

© WIT Press 2010

Printed in Great Britain by MPG Book Group, Bodmin and King’s Lynn.

All rights reserved. No part of this publication may be reproduced, stored in a retrieval
system, or transmitted in any form or by any means, electronic, mechanical,
photocopying, recording, or otherwise, without the prior written permission of the
Publisher.
Contents

Introduction 1
CHAPTER 1
Economic aspects of regional airport development 9
Kenneth Button
1. Introduction 9
2. Transport and economic development 10
3. The economic impacts of airports 11
4. Factor movements 13
5. Finance and ownership 17
6. Retaining air services 19
7. Conclusions 22
References 24
CHAPTER 2
Development of regional airports in EU 27
Maria Nadia Postorino
1. Introduction 27
2. Main transport policies to assure easy mobility within EU 29
2.1 The airport capacity problems 31
2.2 The single European sky (SESAR) project 33
2.3 Trans-European network transport project 35
2.4 Air-rail competition/integration 37
3. Airports in the EU perspective 40
4. Low-cost carrier and regional airports 46
5. Conclusions 49
References 50
CHAPTER 3
The development of regional airports in Asia 53
Yu-Chun Chang
1. Introduction 53
1.1 Characteristics of Asian air transport industry 53
1.2 Trend of major international airport development 56
1.3 Regional airports in Asia 56
2. Aviation policy 59
2.1 Different approach to deregulation 59
2.2 Aviation deregulation in Southeast Asia 60
2.3 Capital city airports reform 61
3. Air transport development between China and Taiwan 63
3.1 Traffic between Taiwan and China 63
3.2 Airports in China 64
3.3 Direct flights between Taiwan and China 67
3.4 Airport policy in China 67
4. Low-cost carriers and regional airports 69
4.1 The rising of low-cost airlines 69
4.2 Environmental characteristics of southeast Asia 69
4.3 Interaction of low-cost carriers and regional airports 72
5. Conclusions 74
References 76
CHAPTER 4
Air demand modelling: overview and application to a developing
regional airport 77
Maria Nadia Postorino
1. Introduction 77
2. Demand modelling and airport catchment area 79
3. Demand model classification 83
3.1 Time series approach 88
3.2 Discrete choice models 92
4. An overview of the Italian airport system 94
5. Application to a test case 96
References 106
CHAPTER 5
Theory and practice in modelling air travel choice behaviour 109
Stephane Hess
1. Introduction 109
2. Air travel choice behaviour 110
3. Guidance for good modelling practice and contrast with status quo 111
3.1 Guidelines 111
3.2 Limitations of practical research 112
4. Data issues 113
4.1 Revealed preference data 113
4.2 Stated choice data 114
5. Model structure 115
5.1 Basic concepts 115
5.2 Correlation between alternatives 116
5.3 Variation in behaviour across respondents 119
5.4 Discussion 120
6. Empirical example 120
6.1 Description of data 120
6.2 Model structure and specification 120
6.3 Model results 121
7. Summary and conclusions 123
Acknowledgements 124
References 124
CHAPTER 6
Practical airport demand forecasting with capacity constraint:
methodology and application 127
Eric Paul Kroes
1. Introduction 128
2. Approach 128
3. Demand forecasts 130
3.1 Traveller choice module 131
3.2 Observed base year OD matrix 134
3.3 Growth factor model 135
4. Supply forecasts 135
4.1 Forecasting demand and supply 136
5. Capacity constraint 137
6. Case study 1: Amsterdam airport capacity planning 138
7. Case study 2: The impact of air flight tax in the Netherlands 140
8. Conclusion 145
Acknowledgements 146
References 146
CHAPTER 7
Regional airports and opportunities for low-cost carriers in Australia 149
Andrew Collins, David A. Hensher & Zheng Li
1. Introduction 149
2. An overview of airline activity and regional airports in Australia 150
2.1 Airline activity 150
2.2 Regional airports 153
3. Establishing a framework in which to investigate the role of regional
airports and airline activity 155
4. Model data 157
4.1 Sourcing of data for model estimation 157
4.2 Overview of available data 160
5. Results of base year models 165
6. Scenario assessment – introduction of low-cost carriers 168
7. Conclusions 172
Acknowledgements 173
References 173
Subject Index 175
This page intentionally left blank
Introduction
Regional airports development: background

The importance of the air transport system is widely recognized not only as a
mean to quickly cover long distances, but also as an economic engine for all
communities. Aeronautical industries, airports and airlines are some of the main
actors that, each one in its own field, compete for the air transport market.
The capability to cover long distances in reduced time with respect to other
means of transport is one of the most successful factors that have contributed
to the growth of the air transport system; as an example, the technological
developments and the increase of civilian routes between Europe and North
America have caused the end of intercontinental travels by ships, that at the
beginning of the 21st century represented the only means of transport to go from
the old to the new continent.
The continuous development and the results obtained by the aeronautical
industries have increased the level of safety, the comfort and the efficiency of the
system.
Today, the air transport system provides the only means that can be used to
reach remote or inaccessible areas, or regions not well served by land transport
systems; it is irreplaceable for medical emergencies and humanitarian assistance;
it is a source of integration between different communities as well as a source of
economic development for the region served by air carriers.
In short, the air transport system plays not only an important role as means of
transport but also as a social connector and economic engine.
The main actors of the air system are airport planners, air companies (mainly
grouped in full carriers vs. low-cost companies) and users (passengers and/or
freights) that can produce important demand levels at airports, thanks to their
travel choices.
In the air transport system, airports are the nodes of the system, while routes
are the links among nodes. In many countries, and especially in European
countries, location and characteristics of the airports are defined at central level,
while links are established thanks to the services offered by the air carriers.
Airports are important elements of the air transport system because they
represent interchange nodes among land transport systems and the air transport
system, and also because they are the air traffic control centres. The main
characteristics of an airport depend on the expected number of passengers and
movements, the performed function and the kind of routes being offered.
2 DEVELOPMENT OF REGIONAL AIRPORTS

In terms of passengers and movements, airports can be defined as:

– primary, if the number of yearly passengers is greater than five million per
year;
– regional, otherwise.
The EU also classifies airports as:
– community airports, if the number of passengers is greater than ten million
per year;
– national, if the number of passengers ranges from five to ten million per year;
– large regional, if the number of passengers ranges from one to five million
per year;
– small regional, if the number of passengers is less than one million per year.
Furthermore, according to ANNEX 14 (ICAO), an airport can be classified
depending on the airport traffic density as:
– light: if the number of movements during the peak hour is less than 15 for
each runway, or if for all the runways it is less than 20;
– medium: if the number of movements during the peak hour ranges from 16 to
25 for each runway, or if for all the runways it ranges from 20 to 35;
– heavy: if the number of movements during the peak hour is greater than 26
for each runway, or if for all the runways it is greater than 35.
Depending on the performed function, an airport can be classified as:
– hub
– feeder.
A hub is an interchange node for an air carrier that offers a hub-and-spoke
service. Particularly, flights coming from multiple origins converge into the
airport (hub) from which new flights start toward multiple destinations (spokes).
The hub-and-spoke structure has been used to increase the number of served
destinations (in fact, starting from a generic origin one can reach whichever
destination going through the hub) and to obtain greater load factors for each
aircraft (in fact, point-to-point services to guarantee the same number of
destinations from each origin could not have a satisfactory level of demand from
the point of view of economic convenience). In this way, the production costs
can be reduced while the quality of offered services can be increased.
Then a hub airport is not necessarily a primary airport (following the
previous definition), but an airport where an air carrier has established the hub of
its hub-and-spoke service.
The choice of an air carrier to locate its own hub at a given airport depends
on different factors as:
– key position with respect to the potential users;
– location near a big city or a metropolitan area able to generate high volumes
of traffic flows other than those in transit;
 REGIONAL AIRPORTS DEVELOPMENT: BACKGROUND 3

– airport high capacity and efficient management in terms of take-off and

landing systems, in order to reduce delays and congestion effects;
– passenger terminals able to guarantee an efficient transfer from an aircraft to
another one.
The management of the airport is crucial for the success of a hub-and-spoke
system; in fact, when airports are not able to support and manage the more and
more increasing volume of traffic, users experience negative effects as delays,
flight cancellation and baggage loss.
A feeder can be defined as an airport supporting the hub or in other words as
the spoke of a hub-and-spoke system; however, flights can also be guaranteed by
other companies, thanks to the alliances and code-sharing networks. Nowadays,
feeder services at many airports classified as ‘regional’ are flanked by point-to-
point services often offered by low-cost companies both toward national and
international destinations.
Inside the hub vs. feeder airport classification, another distinction can be
made between mega-hubs (as in Europe, London-Heathrow, Paris Charles de
Gaulle, Frankfurt am Main), with a large catchment area that makes them
attractive also for point-to-point trips but are too expensive for many low-cost
companies; and secondary hubs (as in Europe, Barcelona, Copenhagen, Lisbon,
Manchester, Rome, Vienna, and so on) which operate as both feeder airports for
mega-hubs and smaller hubs for certain regions.
Finally, depending on the kind of routes being offered, airports can be
classified as:
– first level airports: intercontinental and international links covering distances
greater than 3,000 km; airports supporting this kind of routes are equipped to
work with high traffic levels and are generally hubs for many companies;
– second level airports: international links covering distances less than 2,000–
3,000 km;
– third level airports: national and international links covering distances in the
range of 500–700 km.
A particular characteristic of the air system is its continuous development both in
terms of aircraft technology (larger and faster aircrafts, greater flight autonomy
also for smaller aircrafts, lower fuel consumption, large use of computer and
automatic systems to guarantee both the efficiency and the safety during the
flight, and so on) and service organization (hub-and-spoke rather than point-to-
point service, air company alliances, frequent flyer programs, business models,
and so on).
Among these factors, low-cost companies and regional jets (a kind of aircraft
used for medium-short distances, characterised by high cruise speed as regards to
the aircraft class, low pollution level and reduced landing/take-off distances)
have characterized the air transport system in the last years.
They have as a common aspect the potential re-evaluation of regional (third
level) airports. In fact, low-cost carriers prefer to use regional airports providing
point-to-point links rather than hub-and-spoke services, changing thus the
4 DEVELOPMENT OF REGIONAL AIRPORTS

tendency of the last decades. Similarly, regional jets can be considered the
preferential kind of aircraft for regional airports, because they are characterized
by low capacity in terms of offered seats per aircraft (and therefore they are
competitive as loading factors in low-level demand areas) and by reduced
distances required for landing/take-off operations; low-level demand and short
runways are common factors of regional airports.
The main differences among low-cost carriers and full service (or traditional
or flag) carriers are the following:
– low-cost carriers do not use Computer Reservation Systems (CRS) as selling
channel, but they prefer internet or phone calls;
– check-in, handling and maintenance are generally outsourced services;
– on-board services (as meals, magazines, and so on) have been reduced or
eliminated (for this reason low-cost companies are also called ‘no-frills’);
– fleet is generally formed by the same kind of aircraft (most low-cost
companies use Boeing 737 aircraft);
– the offered routes are generally medium-short and point-to-point inside a
continent (e.g., EU or USA);
– low-cost companies are independent air carriers, i.e. any network/code
sharing among companies;
– they often operate at regional airports where airport fees are low, capacity is
generally high and there are attractive slots; then, air companies can reduce
operational costs, avoid delays and maintain short turn-round times.
A general consequence of the low-cost business model used by the different air
carries has been the reduction of the air fare and the re-evaluation of some less
classical destination areas, which have thus improved their attraction as tourist
destinations. Furthermore, in most cases they have helped the development of
regional airports as well as the development of the nearest territory and the
reduction of congestion at some hubs due to the different distribution of the
traffic volumes.
The considerable increase of the air transport demand, mainly inside Europe
and the USA, but also nowadays in some Eastern countries as China and India –
where more low-cost companies started successfully their services – produces
congestion problems both along the routes and above all at the main airports.
In fact, it is well known that mega-hubs and hubs gather the most part of the
traffic volumes while the remaining airports have a lot of residual capacity that
can also be used to reduce hub congestion. Congestion is a significant, negative
consequence of the hub-and-spoke structure because it produces an increase of
the overall trip time (both for users and air companies); higher probability of
baggage loss; delayed or cancelled flights; and higher values of pollution due to
the aircraft and land vehicle movements, as well as to the induced land traffic to
and from the airport.
On the other hand, the development of regional airports can have as a
consequence not only a reduction of congestion at the main hubs but also the
socio-economic development of the areas close to the airport.
 REGIONAL AIRPORTS DEVELOPMENT: BACKGROUND 5

In this context, and in order to re-evaluate the role of regional airports, the
understanding of both the demand characteristics from/to regional airports and
the techniques of the air transport supply simulation is a fundamental step to
verify the effectiveness and efficacy of the system structure as regards to the
different involved actors (air companies, users, and society).
The overall framework taking into account the different aspects of the air
transport system simulation is depicted in Figure 1.
Air transport demand depends on air services that in turn are offered at a
given airport depending on the number of passengers, i.e. the demand level at
the airport.
There is an evident feedback between the future level of air services and the
passenger airport choice. The airlines will provide air services at airports only if
there is a financial convenience; this means a sufficient number of passengers
choosing that airport in order to make feasible a given level of air services.
Similarly, the number of passengers choosing a given airport depends on the air
service level at that airport. The explicit simulation of this interaction is
important to verify the airport development potential (within a larger airport
system) and then which are the more suitable policies to support such
development.
When demand at an airport and offered services are mutually consistent, the
system reaches an equilibrium that can then be evaluated in order to obtain the
system performances as regards airport managers, passengers, airlines and society.
The simulation of the demand–supply interactions and then the demand
distribution among the various available airports should take into account the
(limited) capacity of the system (Figure 2) and the possible constraints that can
exist and contain the capacity (prescriptive or physical constraints, urban areas in
the neighbourhood, and so on).
A first case is when the predicted demand/supply values and the corresponding
traffic flows (in terms of passengers, and thus aircraft movements) do not satisfy
the current airport capacity. If the airport capacity can be increased (consistently
with possible external constraints as urban areas in the neighbourhood, noise
constraints, available space to physically extend the airside, and so on) a new
situation can be simulated and evaluated by taking into account the monetary costs
required to increase the capacity.
A second case is when the predicted demand/supply values and the
corresponding traffic flows do not satisfy the current airport capacity as before,
but the airport capacity cannot be increased as required because the necessary
interventions are inconsistent with the external constraints. In this case, the
realization of a new airport can be considered and the evaluation should take into
account the overall airport system in terms of demand, supply and interactions
between airports.
Because the passenger’s airport choice depends on the air services at that
airport and because the airlines services are linked to the profitability of their
services, modifications of the cost structure, the available services and the
operational rules can modify the passengers’ choices to use a given airport and
the choices of airlines to provide a given service.
6 DEVELOPMENT OF REGIONAL AIRPORTS

Figure 1: Layout of the air transport system simulation.

Then, a suitable model should be able to simulate how a given set of policies can
influence, on the whole, traffic levels at each airport. This means, to an
increasing level of detail, the simulation of the future scenarios in terms of
economic activities and land use, as well as the development of land transport
systems and the consequences in terms of traffic distribution. Particularly, high-
speed trains are expected to be competitive with the air transport services for
medium-distance trips and they should be taken into account because of the
effects they produce on the overall traffic levels at a given airport. For example,
high-speed train services between a medium-size city, also served by a regional
airport, and a main urban area served by a hub airport will have a great influence
on the air traffic level at the regional airport, both in terms of direct trips and
transfer trips, especially if the high-speed train also serves the hub airport at the
main urban area. In fact, passengers can use railway instead of air services for
city-to-city trips (direct trips), but also to reach the hub and then choose the best
airlines to complete the travel (transfer trip). The regional model should take into
account this kind of effects and interactions between competitive modes both in
terms of predicted demand levels and airport distribution in a given region.
 REGIONAL AIRPORTS DEVELOPMENT: BACKGROUND 7

Figure 2: Capacity constraints and possible limits to airport development.

The above description, also depicted by Figures 1 and 2, provides a theoretical

modelling background; actually, many problems can arise when the different
parts have to be modelled; for example, air service at airports is an airline
decision; the hypothesis of boundary condition stability to simulate future
developments can be unrealistic in some contexts; modelling of the airport user
choice process as well as relationships between air supply and demand to
simulate equilibrium is a really complex task; and so on.
Following the previous considerations, the main goal of this book is to
provide a summary of the key aspects related to the simulation and analysis of
the potential development of regional airports.
In order to give an outline of the main tendencies and policies in many
countries, the EU (Chapter 1) and Asian (Chapter 2) situations are described. Then,
after an overview concerning the main economic aspects linked to the development
of regional airports (Chapter 3), several approaches used in the literature to
simulate the air demand at airports are described, with an application to a regional
airport in South Italy (Chapter 4). Air demand at airports is the result of a choice
process involving also the choice of airlines operating at the airport itself and the
choice of the access mode, depending on the airport accessibility characteristics:
8 DEVELOPMENT OF REGIONAL AIRPORTS

these aspects are discussed in Chapter 5, together with a case study conducted in
Greater London. An operational methodology, developed and applied in the
Netherlands to simulate the air demand–supply equilibrium at airports, is described
in Chapter 6, thus facing practical and theoretical aspects. Finally, the crucial role
that can be played by low-cost carriers is explored for the Australian regional
airports (Chapter 7), by considering key factors as air fares and number of
competitors on each route.

Acknowledgements
I wish to thank Kenneth Button, Yu-Chun Chang, David Hensher, Stephane
Hess, Eric Kroes, as well as Andrew Collins and Zheng Li for their invaluable
contribution and for having supported with great availability this editorial
project. Furthermore, I would like to thank Carlos Brebbia who believed in this
book and gave me the opportunity to realize it, as well as his editorial staff,
particularly Elizabeth Cherry and Brian Privett, for their help in many stages of
the editorial process.

Maria Nadia Postorino

Department of Computer Science, Mathematics, Electronics and Transports,
University ‘Mediterranea’ of Reggio Calabria, Reggio Calabria, Italy
1

Economic aspects of regional airport

development
K. Button
George Mason University, USA
Abstract
Transport has important effects on the nature of local economies and on their
overall development. The focus here is on the particular role that air transport,
and in particular airports, can play in the regional economies. Public policy has
always been interested in ensuring that adequate and appropriate air transport
infrastructure is available to serve regions, be it for strict economic reasons or for
social and political purposes. Where to invest, the form of investments and its
scale have long been within the remit of economic analysts, and more recently
the matters of finance and air service retention has attracted mounting interest.
The role that economics can play in better understanding many of these issues is
considered here.

Keywords: airports; airlines; economic growth theory; market instability; low-cost

airlines.

1 Introduction
Transport has always played an important role in economic development. It is no
accident that the dominant cities of antiquity grew on major trade routes, often
on seaports or on rivers that were the major carriers of inland trade. Railway
stations served a similar role in the 19th century and more recently good road
access has been seen as important. Airports have also played a major role in
many parts of the world in allowing economies, sometimes national, and also at a
regional level, realize more of their economic potential.
Air transport is a relative new mode but has grown rapidly. It is particularly
important for long- and medium-distance movements of people and high-
value/low-volume commodities. It serves a particular purpose where for economic
10 DEVELOPMENT OF REGIONAL AIRPORTS

or geographical reasons it is not viable to construct the fixed track infrastructure

needed for road or rail transport. In many cases, it is seen as essential to ensure
social and political cohesion in large countries, or those with terrain not suited to
surface modes. It is a very flexible mode of transport not strongly tied to a
network of inflexible tracks or infrastructure,1 and, although the larger ones can
be expensive, airports are relatively cheap to build and operate compared to an
extensive rail or road system. The tendency to name airports after local
dignitaries also makes them attractive for politicians.
But there are some downsides too, one of which is the difficulty of retaining air
services; we pay particular attention to this later. They are often monopolies and
this can lead to the exercise of market power by their owners combined with laxity
in management – X-inefficiency. State-owned facilities, where rent seeking is not
usually the norm, are particularly prone to the latter.2 Air transport is also noisy and
airports can involve considerable land-take and surface access challenges.
Consequently, rather than just focus on the potential economic benefits for a region
of having air transport access, issues of economic institutions are considered,
and the ways they may facilitate the provision of airport infrastructure and its
optimal use.

2 Transport and economic development

Airports are seldom, if ever, a strict catalyst for local economic development. An
area is only likely to enjoy substantial and sustained economic growth if it has
inherent potential. Airports are also not free-standing entities; they require
complementary surface infrastructure even if developed at appropriate locations.
Economists are not good at understanding why economic growth takes place, the
types of industry that drive it at any particular time, or the way that it is
distributed geographically. If they did then the planned economies that were
created in the post-Second World War era would be thriving. Consequently
picking ‘winning’ investments in regional airports is challenging.
Traditional neo-economic (exogenous) growth theory focused on factor
endowments as the driver of growth. A region would grow if it had a suitable
mix of economic factors that could be exploited. If it did not have the factors that
was needed then there was the possibility of importing them – labour being the
most obvious, but also some raw materials. In free market conditions, labour
would move to the regions where it was needed, attracted by the high wages
there that reflect scarcity rents. Capital and raw materials would, in turn, move to
regions with excess labour attracted by the returns offered from combining with
labour. In the long run everything would balance out and all regions would grow
at the same pace. Overall in this environment, aggregate growth could only occur
as more factors became available (e.g. a higher birth rate or human survival rate,
or because of technical progress which was seen as a random event). Policies
aimed at maximizing the mobility of factors of production were about all that
governments could advance.
The intellectual works initially of Kaldor [2], but more recently of Lucas
[3] and Romer [4, 5], coupled with the empirical analysis of Barro and Sala-i-
 ECONOMIC ASPECTS OF REGIONAL AIRPORT DEVELOPMENT 11

Martin [6, 7], on economic convergence, have brought into question whether
region’s economies do converge, and, even if they do, whether the rate of
convergence is significant. This is the ‘new growth theory’. The argument
moved to whether growth was, in part, endogenous within regions leading to
divergences in economic performance – the richer getting richer at the expense
of the poorer who were entrapped in a downward spiral of poverty. Basically,
richer regions attracted only the most highly skilled and energetic workers from
the poorer regions, whilst capital was retained because of factors such as
economies of scale (Kaldor’s argument), or because the knowledge base of these
regions stimulated technical progress (Lucas’s and Romer’s argument) which
attract investors’ interests.3
In terms of transport, the traditional view was that appropriate transport was
needed to allow for migration to take place, and countries used to provide
subsidized passage for immigrants. The new growth theory treats it somewhat
differently; it is seen as a possible way for less developed areas to become more
attractive to investors and as a mechanism to bring in skilled labour. This does not
mean that investment in an airport, or expansion of an existing one, will, per se,
lead to local economic development, but it can allow the region to more fully
exploit its potential.

3 The economic impacts of airports

Airports do not just appear in markets, they are planned and their operations are
carefully regulated by local, national and often, in the case of major hubs,
international bodies. There are positive and negative external economic effects,
combined with numerous non-economic influences that determine whether an
airport is built, its scale and its operational design. They also have features that
give them advantages in combining with airlines to provide transport in specific
circumstances, and for the carriage of particular types of traffic. Geography, for
example can be particularly important in determining the number of airports
required and their type. Table 1 [9] shows the number of airports that European
countries have by size. It is noticeable that geographically large countries with
sparse populations and difficult surface terrain tend to have numerous small
airports and few large ones, for example Finland and Norway.
At the macro-level, there have been intensive debates about the role of
infrastructure in enhancing the economic productivity of a country or a region,
and this has influenced policy choices. Early aggregate econometric work in the
1980s [10] indicated a correlation between infrastructure, mainly transport
investments and productivity growth at the national level, but subsequent
analysis at more disaggregate regional, urban and project level showed much less
impact [11].
There are few explicit studies that have sought to undertake a full regional
economic impact assessment of an airport [12, 13]. The task is not an easy one.
Measuring local economic impacts of airport investment is challenging and
studies have often over-estimated them. The main problem is to assess the
opportunity costs involved in the airport investment. Even if the traffic forecasts
12 DEVELOPMENT OF REGIONAL AIRPORTS

and resources needed are corrected forecast, the inputs that are entailed have a
cost elsewhere in the economy and, unless all markets are perfect, this can be
underestimated in a partial equilibrium impact analysis. There are also issues of
counterfactuals; what would have happened to the regional economy if new
airport infrastructure had not been provided [14]?
Table 1: Commercial airports in selected European countries (2004).
Country Passengers
Less than 100,000–1 1–5 More Total
100,000 million million than 5 airports
million
Austria 0 4 1 1 6
Denmark 7 3 1 1 12
Finland 10 10 0 1 21
France 37 20 8 3 68
Germany 18 10 14 6 48
Greece 16 15 6 1 38
Iceland 10 3 0 0 13
Ireland 9 2 1 1 13
Italy 11 4 9 2 36
Norway 25 13 5 1 51
Spain 4 16 10 6 36
Sweden 21 20 2 1 44

The local implications of investing in airport infrastructure can be divided into

four types, each with different temporal horizons.
● Primary effects. These are the short-term benefits to a region from the
construction of an airport – the design of the facility, the building of the
runways, the construction of the terminals and hangars, the installation of air
traffic navigation systems and so on – and the resultant income and
employment multipliers associated with this.
● Secondary effects. These are local economic benefits of running and
operating the airport – employment in maintaining the facility, in handling
the aircraft and passengers, in transporting people and cargo to and from the
terminal and so on. These secondary effects can be extremely important for
some local economies in terms of employment, income and, for local
government, taxation revenue.
● Tertiary effects. These stem from the stimulus to a local economy resulting
from firms and individuals having air transport services at their disposal.
These differ for those living in hub cities, compared to those on a spoke or
having no major carrier. Hubs offer more direct flights favoured by business
travellers. But the hub also benefits those on the spokes because without a
hub-and-spoke structure many would find it difficult to travel long distances
at all. Hubs allow interconnectivity. In the United States, over half of the
15,000 city pairs served by a major carrier have less than one passenger
per day.
 ECONOMIC ASPECTS OF REGIONAL AIRPORT DEVELOPMENT 13

● Perpetuity effects.4 These reflect the fact that economic growth, once started
in a region, becomes self-sustaining and may accelerate. An airport can
change the entire economic structure of a region – it can shift its production
function. This type of dynamic economic impact of an airport is the most
abstract and the most difficult to quantify. It has been little researched but
examples can be seen in the transformation of small agricultural island
economies into tourist destinations and the growth of high-technology
regions around major airports such as Dulles, Washington and Logan, Boston
in the United States.
Getting a good handle on the magnitude of these effects is not easy, and relies on
three broad methodologies: questioning local businesses and stakeholders,
economic multipliers and input–output analysis, and econometric modelling. All
have their weaknesses. For example, econometric analysis requires prior
specification of the causal relationships while input–output analysis tends to be
static in its orientation and is heavily data reliant. Additionally, most analysis
tends to focus on the first three categories of impact (primary, secondary and
pecuniary), and neglect any estimation of a perpetuity effect. In part, this is
because most impact studies are linked to planning horizons that are relatively
short term. Despite all the caveats, there is ample evidence that airports in the
appropriate place can stimulate high-technology development5 and tie regions
into the international economy [16].

4 Factor movements
While regional airport analysis has generally been concerned with making more
and better use of local resources, transport also plays an important role in moving
productive resources between regions. The classic economic assumption of
factor endowments is less applicable in the 21st century.
Traditionally, the most mobile factor of production has been capital, and this
mobility has grown as transport has developed. Raw materials have also been
moved to allow their combination at places of manufacture. Even in primitive
times, stone was often moved long distances by water for the construction of
religious and ceremonial buildings. Mass migrations of people also occurred,
often as the result of conquest or as people have sought to find food or jobs.
More recently, air transport has been instrumental in allowing new forms of
migration as well as the continuation of some more traditional forms. This has
involved, for example movements of large numbers of highly skilled workers to
places of demand. The movement of Indian computer engineers and the like to
meet demands in the United States during the 1990s is an example. While there
has always been a tradition of skilled and specialized workers migrating to find
work – the mediaeval cathedrals of Europe were the result of such mobility – the
speed, extent and duration of modern migrations is of a different scale and form.
Temporary migration is now more common as labour can easily and relatively
cheaply return to home after a period of time. This was not possible before the
14 DEVELOPMENT OF REGIONAL AIRPORTS

advent of the steam ship and railways, but even then it was expensive and time
consuming. Air transport has changed this.
The importance of air transport, and regional airports, in facilitating labour
mobility has been little explored [16]. Where it has been examined, it has tended
to look at larger airports and permanent migration. This is a major issue, it is the
large airports that tend to be the gateways for mass migration over long
distances, and it is these airports that handle most legal issues associated with
migration.6 The gradual institutional freeing up of labour markets within the
European Union, together with the availability of ‘cheap’ labour from many of
the former transition economies (the former Soviet satellite states), has provided
an opportunity for medium distance, short-term migration [17].
The growth of the service sector, the emergence of more flexible
manufacturing techniques, the reduction in trade-barriers and the creation of
more sophisticated financial markets has led to greater mobility in production
and in capital movement. In particular, labour markets have become more
malleable and workers more mobile. This has been explored to some extent in
general terms – for example since the admission of Poland to the European
Union in 2004 some 465,000 Polish workers have joined the United Kingdom
labour market7 – but mainly from the perspective of its economic implications
for national and regional GDP or in terms of the social problems that may arise.
There has been little done recently, however, looking at the role of transport
innovations and migration.8
Air transport facilitates easier migration of workers, makes it possible for
short-term migration and allows migrants to maintain contact with their home
country. Within larger countries, such as the United States, there has also been a
growth in long-distance ‘week-day’ migration as spouses working in different
parts of the country reunited at weekends and work apart during the week. As
Abella [18] has pointed out, ‘International migration of skilled persons has
assumed increased importance in recent years reflecting the impact of
globalization, revival of growth in the world economy and the explosive growth
in the information and communications technology.’
Our interest here is to look at how the new air transport environment has
affected two particular features of the labour market: first, complete (or at least
long term) migration of workers and second, and to a smaller degree, extra long-
distance commuting behaviour.9 The emphasis is on the primary effects of air
transport in facilitating migration. It is not concerned with the secondary effects
of air transport in terms of the relocation of businesses or stimulating particular
industries, such as tourism, that may subsequently lead to changes in levels of
labour migration, or in the role of air transport in leading to retirees and others
simply moving home.10
Labour migration takes place for a multiplicity of reasons and changes in air
transport markets are unlikely to have a large aggregate effect. Where there may be
important impacts are in particular sectors of the labour market or in particular
geographical corridors. This has often coursed debates about the implications of
migration on the local economy, not least in places such as the United Kingdom
where the influx of migrants been sudden and unexpected. There is also a growing
 ECONOMIC ASPECTS OF REGIONAL AIRPORT DEVELOPMENT 15

interest amongst economists in the wider role of transport in factor mobility for
designing policy responses to conditions where endogenous growth seems to
dominate.
Air transport seems to be in many cases a facilitator of these changes. Labour
migration, in its scale and in its changing composition, including greater
emphasis on circulation and temporary migration, has in many cases been shaped
by changes in the availability, frequency and costs of air travel. It makes the
initial migration itself more viable and, by facilitating cheap return trips, reduces
the longer term social costs of being away from kith and kin even when the
migrant locates considerable distances from traditional gateways.11
The reforms in air transport regulation have overcome many of the previous
limitations of air transport as a significant form of mass mobility; costs were a
significant barrier to air travel as were the frequency and convenience attributes.
Low-cost airlines, and their knock-on effects on the legacy carriers, have
changed this. As a result they have impacted on labour markets in several ways,
but mainly through reducing travel costs and increasing accessibility.
Effectively, they reduce the transaction costs of international labour migration
and, all else being equal, by shifting the balance between the costs and returns of
migration have contributed to the increase in factor mobility.
For individuals, the cost of being away from home is high (mental and
physical stress, the cost of separation, etc.), for others the cost of travelling may
be more important. For all, air transport lowers total migration costs. Some can
visit relatives more often. Others can at least afford getting to their destination.
There is also the induced demand for migration that is made possible by lower
air transport costs. Agricultural Mexican workers could not, for example, move
to Canada to work temporarily if it were not for air transport because obtaining
transit visits to travel via the United States can prove very difficulty for low-
income individuals.
There seems to be a change in attitudes towards flying as low-cost services
have grown. Low-fare services available from a local airport seem to be
changing consumers’ perceptions about flying generally and consequently are
having an effect on travel patterns. In many cases, as with Ryanair in Europe that
serves numerous small airports with radial structures of routes, it is not simply
about vacations and visiting a second home but also seems to stimulate people to
apply for jobs abroad and may facilitate working far from home. Wizz Air, the
Hungarian air carrier, is a leader among several low-cost airlines transporting
planeloads of Poles, Hungarians and others to Western Europe with one-way
fares starting at less than €20 (about $26), including taxes [20]. Nearly one
million East Europeans have moved to the United Kingdom, Ireland, Sweden,
Germany and other member countries after the European Union expanded from
15 to 25 nations in 2004.
In particular, there has been increased air traffic between several of the new
European Union member countries with significant migrant flows into the United
Kingdom on routes where there have been expansions of low-cost carrier
activity; not only Wizz but also Centralwings (a subsidiary of Lot Polish
Airlines), the Slovenian carrier SkyEurope Airways and others. Just taking
16 DEVELOPMENT OF REGIONAL AIRPORTS

Poland as an example of service growth, in 2000 there were 5 scheduled services

between Poland and the United Kingdom; by 2006 this had grown to 27
scheduled services linking 12 Polish cities and 12 United Kingdom airports [21].
While the causality between changes in the airline market and labour
migration patterns is not all unidirectional, workers are increasingly participating
in labour markets far from home and airlines have responded by creating an
informal new travel category alongside the traditional business, leisure, and
‘visiting friends and relatives’ traffic breakdown. Airlines often call this ‘ethnic
traffic’ to reflect the cultural diversity of these travellers. Many carriers have
even adapted their business models to cater for these ethnic travellers because of
the relative reliability and predictability pattern of their demands that offset the
relatively cheap fares paid. Ethnic travellers are for instance highly regarded by
low-cost airlines like Wizz and SkyEurope Airways.
While official statistics do not capture this particular sub-class of traveller,
one can glean some indication of the growth in this ethnic traffic, at least in
Europe, by looking at the conventional visiting friends and relatives (VFR); most
of the growth in VFR being in migrants making visits to their homeland. Table 2
takes the two primarily low-cost United Kingdom airports, Stansted and Luton,
and compares the number of inbound passengers for 2000 and 2005. As can be
seen, VFR traffic grew by 198% over the period to become the largest single
component of inbound traffic. At the national level, a similar picture emerges
with VFR traffic growing from less than 2.5% of European Union passengers in
1997 (when there were 15 member countries to about 15% by 2005; albeit with
25 members).
Table 2: Inbound passengers from the European Union to the United Kingdom
using Stansted and Luton airports.
Passenger Passengers Passengers 2000– Percentage Percentage
type 2000 2005 2005 of total in of total in
(million) (million) change 2000 (%) 2005 (%)
(%)
Business 0.9 1.8 98 22 17
Leisure 1.6 4.0 150 39 38
Visiting 1.6 4.8 198 39 45
friends and
relatives

There have been significant changes in demographic patterns over the past 40
years that affect the context in which residential/work place decisions are made.
Portes [22] suggested the existence of transnational communities – a community
that spans borders – boast their key assets, shared information, trust and contacts.
This has led to a more flexible economic structure that often involves longer
distances between home and work and between spouses for part of the time. This
type of behaviour is not migration in the traditional sense, but involves an
extended, regular commute, and often, spending parts of the week away from
home.12
 ECONOMIC ASPECTS OF REGIONAL AIRPORT DEVELOPMENT 17

While explicit data on the extent of long-distance commuting involving air

travel is small,13 some indication of its importance can be seen indirectly in the
high grow rate of extended-stay hotels in the United States and, albeit slower,
elsewhere. These hotels, that trace their ancestry back to the opening of
Marriott’s Residence Inn in 1974, are normally defined as hotels where guests
stay for more than five nights and that do not offer ancillary services such as
bars, restaurants and porterage. It is estimated that up to 20% of United States
hotel stays are over five days and extended stay hotels account for most of this
type of accommodation (240,000 rooms). While not all are regular ‘commuters’
and not all use the low-cost air services to travel to their accommodation, many do.

5 Finance and ownership

From a policy perspective, while there are good reasons to ensure that adequate
transport is available to facilitate economic development, there are market
imperfections that can impede this happening. One serious set of constraints in
most countries is that associated with planning requirements. These can delay the
construction of a facility and affect its design and operations. We will not deal
with these issues here, because they vary considerably between countries, but
rather focus on issues of finance and efficient operations. Financing can pose
particular problems in poorer countries, or even regions within more prosperous
nations if they do not have an adequate tax base, or access to more sophisticated
money markets.
Airports are relatively expensive build and, because of their monopoly status
in many locations, are often not operated efficiently. To reduce the X-
inefficiency, there have been moves in many countries to make airport operators
more commercially oriented. The national approaches aimed at injecting more
commercial pressures into the provision of airport services have varied. Much of
the interest that is now emerging is either the privatization of some aspects of
airport activity or engaging the private sector in some partnership arrangement
with the state. In part this is because an airport is effectively a composite entity
comprised of units offering a variety of services – land access, parking,
concessions, terminals, runways, ground handling, fire and response units,
security, etc. Commercialization does not have to be applied to all these
activities, and may be pursued in a piece-meal way if politics or economics
dictate.
Approaches differ according to the state of the local air transport market that
is, in turn, often linked to the stage in economic development of the region
concerned. For example, airports can vary in terms of their potential revenue
flow from different sources – e.g. slot fees and concession – and this can affect
the degree of privatization or deregulation that is possible and the form it is most
likely to take.
Much depends on the state of the regional air transport market. The forecasts
of relatively slow longer term growth in air traffic in and between the developed
countries (e.g. estimated at about 3.6% a year to 2025 within North America,
3.4% within Europe and 4.5% between North America and Europe) means that
18 DEVELOPMENT OF REGIONAL AIRPORTS

their major airports will increasingly become dependent on commercial or non-

aeronautical revenues to enhance their revenue stream. This in turn can pose
problems in terms of regulation as has already been seen in the debates over the
imposition of the price-capping regime used in the United Kingdom.
While still relatively small, the protected growth of many air markets
involving regions of developing countries offers the potential for increased
airside revenue in situations where there are potentially fewer social constraints
involving such things as noise and land-take on building additional capacity or
where there is already adequate capacity. The scope for raising significant
commercial income is much less, however, because of lower initial traffic bases.
It also suggests, though, that the regulatory regime overseeing a privatized
airport system needs to be less sophisticated because it only has to deal with
airside issues, making the potential for various forms of regulatory capture
smaller.
The problem for regions of the poorest developing countries, however, is that
even though their air traffic flows may in aggregate be growing, this is from a
low base and they still seldom generate sufficient revenue to cover the full costs
of operations. Airports are essentially decreasing cost entities for which cost
recovery can be difficult, especially in a situation where there is competition
from other airports. This makes pure privatization options less tenable and the
need for outside assistance from aid agencies or from government more relevant.
Private/public partnerships offer another alternative [23].
The wide variety of circumstances around the world has led to a diversity of
approaches to commercialization of airports. Some have involved complete
divestiture of former state assets to the private sector, albeit with some oversight
of how the airport is operated, but in other instances the withdrawal of the state
has been less complete.
The management contract approach retains government control, but involves
contracting out specified elements of airport services – parking, hotels, retail
concessions, etc. – for periods of time. This normally involves some form of
auction. The system is in line with the notion of ‘competition for the market’.
Long term contracting involves giving over the operational side of an airport,
sometimes including investment commitments in additional capacity, for an
extended period with the authorities retaining a degree of strategic control. The
financing required normally entails bringing inspecialized international
companies with the expertise to manage an airport, or system of airports,
together with finance houses that can provide the necessary support for large
scale service activities. These types of concessions are widespread in South
America, where local expertise and finance is limited, but there is reluctance for
the state to divest itself of aviation assets.
Many developed countries have also pursued similar philosophies when
expanding the involvement of private enterprise, but falling short of the complete
divestiture [24]. Many United States airports have adopted various concessionary
schemes – for example Boston, Pittsburgh and Reagan National Airport,
Washington have entered into concessionaire agreements, for the entire
operations at Pittsburgh and for specific terminal buildings at the other airports.
 ECONOMIC ASPECTS OF REGIONAL AIRPORT DEVELOPMENT 19

At Chicago O’Hare airport, parking has been contracted out, and the Port
Authority of New York and New Jersey, that own a number of airports, has a
variety of agreements covering such things as the operation of terminal buildings
and the supply of heating and cooling at some of its facilities. There is also a
tradition in the United States of significant airline involvement in providing
check-in facilities and baggage systems.
Complete privatization of major airports is uncommon, although increasing.
In most cases, there is concern that a privately owned airport will exercise its
monopoly power to extract rent from customers. The challenges are to device
and operationalize appropriate regulatory regimes to monitor and direct these
large companies in the public interest – often price-capping is deployed. The on-
going debates about full privatization concern such things as whether single
airport, or as with the United Kingdom’s BAA, systems of airports, should be
privatized and when they are privatized what should be regulated; should it be all
airport activities or just those directly aviation related?
In their overview of various governance options, Carney and Mew [25] focus
correctly on the government being involved in both seeking to improve the
efficiency of their airports and trying to direct the gains to particular groups
rather than leaving management with full autonomy in their actions. This
involves complexities that, while common to businesses in developed countries,
are unfamiliar in many parts of the world. As a result, this has added to the
growth in international firms specializing in airport management, including
ownership, to allow the development of common, best-practice methods of
operation while at the same time being innovative in creating bespoke models for
different circumstances.

6 Retaining air services

Airlines are by definition mobile, and simply attracting an airline service to an
airport does not guarantee that it will continue services or that it will always offer
the same range of services. In the world of deregulated airline markets, the
commitment of carriers to serve particular airports is governed by the
profitability of the services that it offers. There are some fixed costs of starting a
new service and of withdrawing there are stranded costs, but these are relatively
small. Airlines can lease or rent most things required to operate a service,
including planes, crew, catering, slots, gates, space on reservation systems,
baggage handling and check in facilities. Withdraw of a service can also mean
transfer of assets if they are owned by an airline, or the assets being leased to
another airline.
In contrast to this flexibility, industry and individuals seek continuity of
services in their decision-making. When deciding where to locate, a company
usually commits considerable sums that are not fully recoverable in the short to
medium term, or at least not without considerable discounting. The sunk costs
elements are even greater when there is specialized plant and equipment
involved. Similarly, with individuals, when selecting a place to live
consideration is given to the quality of transport services in the area and the
20 DEVELOPMENT OF REGIONAL AIRPORTS

larger accessibility of the area in national and international terms. Moving from
an area, particularly one that has gone into an economic downturn can be costly
to individuals that have bought, or taken out long-leases on homes and can
have serious social implications for families that have to abandon personal
ties. In effect, airlines are much more flexible in terms of the services that
they offer than many of the businesses and households that make use of their
services.
This problem is compounded because the competitive scheduled airline
service is highly unstable. Figure 1 shows the operating margins in the main
airline markets of the world. Elementary economic theory tells us that when
there are no fixed-costs then competitive bargaining between suppliers and
customers will ensure that prices are kept to the minimum level that that do not
allow suppliers to recover all costs over the long term. When there are no fixed
costs, the marginal cost of meeting customer demand represents the entire costs
of production. The problems come when there are fixed costs. In this case
markets can be very unstable.

Figure 1: Operating margins of airlines (1988–2006).

Notes: (i) A lack of a bar indicates a missing observation and not a zero operating
margin, (ii) memberships of the various reporting bodies vary over time
and thus the reported margins reflect the associated carriers at the time of
reporting.
Sources: Boeing Commercial Airplane, Association of European Airlines, Air Transport
Association of America, International Air Transport Association.
The traditional view of fixed costs was developed when the bricks, steel and
mortar of industrial plants had to be paid for. The world has changed, and with
service industries, and especially those like airlines that provide scheduled
services, the fixed costs are somewhat different. While airlines do use expensive
hardware, as we have seemed, this does not imply that this in itself imposes a
fixed cost. Fixed costs for airlines take an entirely different form.
 ECONOMIC ASPECTS OF REGIONAL AIRPORT DEVELOPMENT 21

An airline is committed to a particular scheduled service some six months or

so before the flight – it is committed to have a plane, crew, fuel, gates, landing
and take-off slots, etc. In deregulated market they engage in price discrimination,
charging passengers different fares to try to extract as much revenue as possible.
In general, this means that lower fares are offered initially when a fight is some
way off because leisure travellers are willing to pay less for a seat and are more
flexible in their scheduling and will seek lower fares if available. They are
caught early by the airline. Towards the time of take-off, fares rise as last minute
travellers, often business travellers, seek seats. These people are less sensitive to
fares because meeting a last minute business deadline can make or break a deal,
and tax deductions are normally allowed for the offsetting of higher fares.
The problem is that with a fixed schedule in a competitive market, the
various airlines set take-off times for each destination at about the same time.
This leads to intense competition to fill seats and forces fares down to the levels
that do not allow all the costs of individual’s services to be met. It is worth filing
a seat once it is there with anyone willing to pay for the additional costs of
handling.14
Airlines have sought to minimize this problem in a variety of ways. Many
offer frequent flier programmes to retain customers, seek to control information
and booking systems, focus on routes where, either because they are new or for
some institutional reason, they legally have a degree of market power, try to find
ways of getting state subsidies, concentrate services on major ‘fortress’ hubs
where they control gates and slots, form cartels as strategic alliances so as to
widen their market presence and reduce costs, and others seek to limit
competition by keeping costs low – the ‘low cost business model’. For a time
these policies may work but, for example, fortress hubs have proved vulnerable
to point-to-point services and as load factors have increased so the values of
frequent flier miles have diminished – there are simply fewer services they can
be redeemed on.
The inability of airlines as a whole to recover their full costs makes
continuity of service provision uncertain, and particularly so during any period of
economic recession or slow growth. The low-cost business model has been
touted as a possible solution to this and it is true that in Europe there has been
significant regional economic development around major low-cost airports such
as Frankfurt Hahn served by Ryanair. But, the low-cost model itself is not stable
and there have been numerous failed low-cost airlines. Table 3 gives examples in
Europe covering a relatively short period. While Ryanair and Southwest (in the
United States) have prospered, they enjoyed clear first-mover advantages in their
respective markets.
There have been efforts on the part of industry to ensure continuity of services
from regional airports. In Europe, for example, Chambers of Commerce such as
that in Strasbourg have sought to subsidize services but have run foul of European
Union laws regarding the use of public money. In the United States, direct private
sector has been more active in retaining air services.15
22 DEVELOPMENT OF REGIONAL AIRPORTS

Table 3: European low-cost carriers that ceased to exist (2003 to 2005).*

Aeris BuzzAway Hellas Jet
Agent Dream Air Hop
Air Bosnia Duo Jet Magic
Air Andalucia Europe DutchBird Jetgreen
Air Catalunya EastJet JetsSky
Europe Air Exel EU Jet JetX
Air Freedom Europe Exel Aviation Group Low Fare Jet
Europe Air Fairline Austria Maersk Air
Air Littoral Fly Eco Now
Air Luxor Fly West Silesian Air
Air Madrid Flying Finn Skynet Airlines
Air Polonia Free Airways Spirit Of Balkan
Air Wales Fresh Aer Swedline Express
Airlib Express Germania Express V Bird
BasiqAir GetJet Poland VolareWeb
BerlinJet Go Fly White Eagle
Bexx Air Goodjet Windjet
* Most of these airlines operated for a period and then went into bankruptcy. Some such
as Go Fly and BuzzAway merged with successful low-cost airlines. In a few cases, the
airline was registered but never offered actual services.

7 Conclusions
Much of the modern world, and particularly so in the high income, market
oriented economies, is based on mobility and accessibility. There are few cases
of significant economic development in regions that are spatially remote and are
not tied into modern transport networks. The development of an airport in a
region is not a panacea that will suddenly bring economic prosperity. It has to
attract airlines’ services, and services that are likely to be durable over the
foreseeable future. To do that the region has to have intrinsic features that can
be exploited by air access – airlines essentially seek to exploit these local
advantages by moving the people and goods associated with them.
What we have argued here is that it is inadequate to simply consider the
airport infrastructure in isolation when assessing the role of airport investment in
regional economic development. The normal economic impact work conducted
as part of the initial planning of an airport is seldom adequate, particularly when
the role of air transport has changed so much in recent years. The way an airport
is financed and operated affects the nature and quality of air services that it
provides. Simply having a set of buildings and a runway of itself, may prove
ineffective if the financial structure does not allow adequate maintenance and
there is no incentive for management to operate the airport efficiently. Further,
no airport can be treated in isolation from the airlines that use it. The economic
 ECONOMIC ASPECTS OF REGIONAL AIRPORT DEVELOPMENT 23

market for airline services is a complicated one and one that is prone to serious
instabilities.

Notes
1. Air navigation systems are required for traffic control and advice but these are more
akin to traffic management (signals, signs, etc.) on roads than the roads themselves.
2. There is also a tendency when private airports have their returns regulated for them to
over-invest. This Averch–Johnson effect [1] is commonly seemed as ‘gold-plating’ by
airlines.
3. A survey of the literature on regional economic growth theories is contained in
Button [8].
4. These may be thought of as producing an upward shift in a region’s production
function rather than just a movement along it.
5. Button and his co-workers [15] found that in the United States a city with a major
hub-airport had, on average, an extra 12,000 high technology goods in the region.
6. There is also the issue of the extent to which migrants remain at a gateway and ‘push-
out’ existing workers that are forced to migrate internally to other parts of the country.
7. This is an estimate and includes short-term migration. The United Kingdom Office of
National Statistics (web site: http://www.statistics.gov.uk/) provides periodic data on
migration but given the flexibility of the European labour market, this only offers
broad approximations of trends. For example, it is weak regarding duration of
employment and on the movement of self-employed labour.
8. In the past, there has been considerable interest in the role of transport in facilitating
labour migration, for example in terms of examining the role of shipping development
in the forced migration of slaves from Africa to the Americas and of convicts from the
United Kingdom to Australia, and of voluntary migration as facilitated by such as the
transcontinental American railroads.
9. Migration is normally defined as individuals living outside their county of birth for
more than a year. This data misses illegal migration, although it is sometimes
estimated, and a number of groups, such as students, who may live in a country for an
extended period but are not considered immigrants.
10. Low airfares and services to secondary airports in Europe have stimulated, for
example, an extensive migration of retirees from cooler countries such as the United
Kingdom to Spain, Portugal, France and Italy. For example, in 2005, 224,841 British
people were registered in Spain, 136,766 having an EU residence card, and some
75,000–100,000 were estimated by the Spanish government to reside there.
11. Saxenian’s study [19] provides details of labour structures in California and their
travel to and from home.
12. Overall there are over 400 million long distance business trips, 16% of all business
trips, made a year in the US, but many of these cannot be defined as commuting since
they are irregular and with individuals going to a diversity of destinations. Of trips
of 100 miles or more, 13% were made by air in 2001, with an average distance of
2,080 miles.
13. Most studies of United States long-distance commuting normally define it as
commutes of over 90 minutes or 50 miles. Air travel makes little impact on most trips
of this length but cuts in for much longer commutes of several hundred miles or more.
14. This generic problem is seen to exist when there is an ‘empty core’ [26, 27].
15. In Wichita, Kansas, 400 businesses raised $7.2 million to attract carriers. Air Tran
started operations in May 2002 with services to Atlanta and Chicago’s Midway
airport. The agreement included up to $3.0 million to cover losses in its first year and
24 DEVELOPMENT OF REGIONAL AIRPORTS
$1.5 million in the second. Similarly, Pensacola, Florida raised $2.1 million from 319
businesses to attract Air Tran while companies and individuals in Stockton, California
bought $800,000 of prepaid tickets to attract American West.

References
[1] Averch, H. & Johnson, L.L., Behavior of the firm under regulatory
constraint. American Economic Review, 52(5), pp. 1052–1069, 1962.
[2] Kaldor, N., The case for regional policies. Scottish Journal of Economics,
17(3), pp. 337–348, 1970.
[3] Lucas, R.E., Making a miracle. Econometrica, 61(2), pp. 251–272, 1993.
[4] Romer, P.M., Endogenous technical change. Journal of Political Economy,
98(5), pp. S71–S102, 1990.
[5] Romer, P.M., The origins of endogenous growth. Journal of Economic
Perspectives, 8(1), pp. 3–22, 1994.
[6] Barro, R.J. & Sala-i-Martin, X., Convergence across states and regions.
Brookings Papers on Economic Activity, 1, pp. 107–182, 1991.
[7] Barro, R.J & Sala-i-Martin, X., Convergence. Journal of Political Economy,
100(5), pp. 223–251, 1992.
[8] Button, K.J., The economist’s perspective on regional endogenous
development. Regional Endogenous Development, eds. R. Stimson &
R. Stough, Cheltenham, Gloucestershire: Edward Elgar, 2008.
[9] Williams, G.A., Comparison of Airports in the EU, Bedfordshire, UK:
Department of Air Transport, Cranfield University, 2005.
[10] Aschauer, D.A., Does public capital crowd out private capital? Journal of
Monetary Economics, 24(2), pp. 171–188, 1989.
[11] Button, K.J., Infrastructure investment, endogenous growth and economic
convergence. Annals of Regional Science, 32(1), pp. 145–162, 1998.
[12] Green, R.K., Airports and economic development, Real Estate Economics,
35(1), pp. 91–112, 2007.
[13] Butler, S.E. & Kiernan, L.J., Measuring the Regional Economic
Significance of Airports, Washington, DC: Office of Airport Planning and
Programming, Federal Aviation Administration, 1992.
[14] Debbage, K.G., Airport runway slots: Limits to growth. Annals of Tourism
Research, 29(19), pp. 933–951, 2002.
[15] Button, K.J., Lall, S., Stough, R. & Trice, M., High-technology employment
and hub airports. Journal of Air Transport Management, 5(1), pp. 53–59,
1999.
[16] Button, K.J. & Taylor, S.Y., International air transport and economic
development. Journal of Air Transport Management, 6(4), pp. 209–222,
2000.
[17] Button, K.J. & Vega, H., The effects of air transport on the movement of
labour. GeoJournal, 71(1), pp. 67–81, 2008.
[18] Abella, M.I., Skilled Labour Migration from Developing Countries:
Analysis of Impact and Policy Issues, Geneva: International Labour
Organisation, International Labour Office, DFID Paper Series, 2002.
 ECONOMIC ASPECTS OF REGIONAL AIRPORT DEVELOPMENT 25

[19] Saxenian, A., Local and Global Networks of Immigrant Professionals in

Silicon Valley, San Francisco: Public Policy Institute of California, 2002.
[20] Michaels, D., ‘Ethnic’ routes, growth market for airlines: Cheap travel for
immigrants. Wall Street Journal, 7(March), p. A1, 2007.
[21] United Kingdom Civil Aviation Authority, No-frills Carriers: Revolution or
Evolution? A Study by the Civil Aviation Authority, London: Civil Aviation
Authority, 2006.
[22] Portes, A., Global villagers: The rise of transnational communities. The
American Prospect, March–April, pp. 74–77, 1996.
[23] Button, K.J., Air transport infrastructure in developing countries:
Privatization and deregulation. Aviation Infrastructure Performance: A
Study in Comparative Political Economy, eds. C. Winston & G. de Rus,
Washington, DC: Brookings Institution, pp. 193–221, 2008.
[24] Button, K.J., The Implications of the commercialization of air transport
infrastructure. The Economics of Airline Institutions, Operations and
Marketing 2, ed. D. Lee, Oxford: Elsevier, pp. 171–192, 2007.
[25] Carney, M. & Mew, K., Airport governance reform: a strategic management
perspective. Journal of Air Transport Management, 9(2), pp. 221–232,
2003.
[26] Button, K.J., Liberalising European aviation: Is there an empty core
problem? Journal of Transport Economics and Policy, 30(3), pp. 275–291,
1996.
[27] Button, K.J., Does the theory of the ‘core’ explain why airlines fail to cover
their long-run costs of capital? Journal of Air Transport Management, 9(1),
pp. 5–14, 2003.
This page intentionally left blank
2

Development of regional airports in EU

M. Nadia Postorino
University ‘Mediterranea’ of Reggio Calabria, Italy

Abstract
A suitable development of regional airports in large areas, such as continental
areas, can be obtained if there exist appropriate policies addressed to achieve
this goal. In the specific case of the European Union, some specific programmes
at European level, such as the Single European Sky and the Trans-European
Network programmes, will have many implications for the development of
regional airports as well as the relationships between regional airports and low-
cost air carriers. Regional airports can play an important role to reduce the
congestion peaks at main airports, particularly in the light of an integrated
transport network where many modes coexist and can integrate or substitute
each other. Particularly, high-speed rail can improve the airport accessibility by
increasing its catchment area or represent an alternative to the air mode.

Keywords: airport capacity; high-speed rail; low-cost air carriers; regional

airports; SESAR; TEN-T.

1 Introduction
In the last years, the air transport system has grown quickly and the planning
policies had to be adjusted to guarantee a suitable development of both air
services (as courses, frequencies, fares and airport services) and infrastructures
(as airports and land transport access facilities).
The deregulation process, started in 1978 in the USA with the Deregulation
Act, and then in Europe with the ‘Packet for aviation liberalization’, has led to a
further reorganization in the air field and such a process is still in progress.
The relatively recent foundation of the European Union (EU) has given rise
to a set of problems concerning the Air Traffic Control (ATC) system along
routes that are some of the most congested in the world, mainly due to the lack of
28 DEVELOPMENT OF REGIONAL AIRPORTS

homogenous procedures among the several European Nations. To this aim, the
EU has started an ambitious programme to create a Single European Sky
(SESAR project) in order to have a good ATC along the routes and to identify
Air Traffic Management (ATM) policies able to improve capacity and develop
an efficient and sustainable European air transport system by maintaining a high
level of safety. ATM environmental efficiency is one of the key objectives of the
SESAR programme as the EU gives a great importance to the reduction of the
environmental impact of aviation, in particular greenhouse gas emissions and
noise. At the same time, safety is the other important goal to be achieved: in fact,
although flying is one of the safest travel mode, as existing technical systems
reach the limits of their capacity, the risk of accidents grows, thus making crucial
the development of a new generation of ATM systems able to satisfy the needs
of safety and efficiency of flight operations, in the air as well as on the ground.
Congestion and delays at the airports are another important problem, and
environmental considerations on the really crowded European continent often
prevent the larger airports from expanding. In fact, most important airports are
quite near to cities and produce important negative effects due to both air
services (as noise and pollution) and airport land accessibility (as road
congestion). As air traffic in Europe is expected to increase substantially in the
next 20 years, the above problems are also expected to worsen if suitable policies
for a sustainable development are not implemented. Furthermore, as collected
data show, the cost of the European ATC system increases as air traffic
increases, and hence, all things being the same, also costs are expected to grow
in the future.
In the USA, the harmonization of the aviation ATC procedures overcomes
the EU problems, but the system is congested as well. In particular, the territorial
extension of the USA as unique country and the significant development of the
aviation sector with respect to alternative land systems (as railways services)
have made the air transport system the most used in the country. Low-cost
carriers have given a further impulse to its diffusion, thus making the air
transport system the most preferred means of transport to move within the USA.
The great interest for regional airport development both in the EU and the
USA is then mainly due to their potential role in reducing the congestion along
the main used routes and also reducing the negative impacts at the hub airport
nodes, both for the air transport users and the overall community.
The use of regional airports has many advantages: (1) satisfactory airport
capacity availability; (2) opportunity to promote the socio-economic
development of decentralized areas and regions; (3) reduction of negative
impacts due to the less traffic levels; (4) development of a spreader air service
network.
In this light, many policies have been introduced by the EU, specifically
addressed to promote the re-valuation of regional airports in the several EU
nations. As stated by the European Commission (EC), ‘Regional airports are
important to the development of an integrated European air transport network. In
this respect, it would be desirable to unlock existing latent capacity at regional
airports provided that Member States respect Community legal instruments
 DEVELOPMENT OF REGIONAL AIRPORTS IN EU 29

relating to state aids. Global Navigation Satellite Systems could play a

significant role for increasing capacity and flexibility of operations at those
airports without increasing the cost of local infrastructure. Member States should
endeavour to improve the accessibility of such airports by rail and road to allow
them to act as reliever airports.’ (‘An action plan for airport capacity, efficiency
and safety in Europe’, Brussels, Jan 2007.)
At the same time, the continuous development of low-cost air carriers has
given a further impulse to the re-valuation of regional airports. In fact, low-cost
air carriers prefer to use regional airports for many reasons as follows: reduced
aviation costs at the airport, competitive taxi and turn-around times, availability
of airport capacity due to low traffic congestion and so on. Furthermore, they can
also benefit from start-up, within the rules of compatibility established by the
Articles 87–89 of the EC Treaties and the Community guidelines on financing of
airports and start-up aid to airlines departing from regional airports. Particularly,
Article 87(3)(a) and (c) apply in relation to undertakings established or investing
in disadvantaged regions. The Guidelines specifically refer to its application in
relation to aid for undertakings financing the building of a hangar in such
regions. However, Article 87(3)(c) cannot be used to justify operating aids
(except the case where an undertaking is established in an eligible region and the
measure counterbalances particular difficulties).
To summarize, the EC recognizes the importance of regional airports to
develop an integrated European air transport network because their existing
latent capacity can reduce the congestion peaks at main airports; anyway, in
order to design an effective air transport network the EC also recommends the
improvement of their accessibility (rail and road access).
Thus, the use of regional airports can bring many advantages not only to the
air companies, airports and users, but also to the whole community both in terms
of social benefits (increased accessibility from decentralized or peripheral
regions) and potential economic expansion.
In the following, after an overview of the European main transport policies,
addressed to assure quick and easy trips within the EU (Sections 2), the main
challenges of the European airports in terms of capacity, charges and state aids
are considered (Section 3); the main relationships between regional airports and
low-cost air carriers in the European context are also summarized (Section 4);
and finally, some main conclusions are discussed (Section 5).

2 Main transport policies to assure easy mobility within EU

As well known, the EU is a relatively recent institution: born in 1950 with only
six nations, today is formed by 25 nations within the European continent,
following a significant enlargement in 2004 (Figure 1). Furthermore, two nations
(Bulgaria and Romania) have already signed the Accession Treaty while another
two (Croatia and Turkey), and several others, have requested for the EU
membership.
30 DEVELOPMENT OF REGIONAL AIRPORTS

Figure 1: The European Union.

The progressive enlargement of the EU has posed many questions concerning the
way towards the harmonization of procedures and policies, mainly in the field of
air transport where the integration of the existing control systems is crucial for
the safety and efficiency of the air navigation.
Many transport policies have been implemented in order to define a Trans-
European Transport Network (TEN-T) thus enabling European citizens to move
along the EU in a quick and efficient way.
For the air transport system, key elements to meet the goals of safety and
efficiency can be identified in the following: (1) better effective use of existing
airport capacity; (2) improvement of the airport environmental capacity as well as
the planning framework for new airport infrastructure; (3) integration among
several transport modes (co-modality) to assure fast, efficient and safe trips; (4)
development and implementation of cost-efficient technological solutions.
Two main transport policies are here discussed, concerning the air traffic
development and the fast railway system. Following the EC design of an
integrated, co-modal, spread transport system, they are related to each other
because air transport and railway systems are competitive on some routes but
complementary on some others. Improvement of co-modality is one of the basic
targets recommended by the EC that can also be achieved thanks to available
Community funds.
Before discussing the implication of competition-complementarities between
high-speed lines and air transport, an overview of the airport capacity problem is
presented and then the already recalled SESAR project and the Community
multiannual TEN-T programme are briefly described, with particular reference to
the implication for the development of regional airports.
 DEVELOPMENT OF REGIONAL AIRPORTS IN EU 31

2.1 The airport capacity problems

The airport capacity is an important matter for the EC, as it will not be able to
meet the demand in the next future and then it could be the most constraining
factor on the overall air transport chain. Furthermore, new problems concerning
the sustainability of airports, particularly the airport noise management, will
further constrain the capacity following the concept of environmental capacity as
‘the capacity of the receiving environment, both human and non-human, to
tolerate the impacts of airport activity’ (Upham et al. [1, 2]). Given that acoustic
impacts are the most relevant for the airport activities, the airport acoustical
capacity is also defined as ‘the maximum number of aircraft movements, per
RWY direction, within a specified lapse of time, generating a noise level at most
equal to the one prescribed by laws for a particular area’. Then, aircraft noise is a
major constraint to airport development.
To overcome the environmental constraints, noiseless aircraft are now
operating to meet the government requirements of improving noise climate
around airports (e.g. Directive 2002/30/EC), even if many airlines continue to
use old noisy aircraft. Anyway, as air traffic is expected to increase significantly
in the next future, this measure will soon be ineffective and then noise operating
restrictions on aircraft should not be applied as a first alternative, but only after
consideration of the benefits to be gained from other elements.
The Balanced Approach to Noise (as fixed during the ICAO Assembly in
2001) identifies an international approach to tackle airport aircraft noise
problems in an environmentally reactive and economically responsible way. It
includes four main elements:
● reduction of noise at source;
● land-use planning and management;
● noise abatement operational procedures;
● noise operating restrictions on aircraft.
Then, both the increase in demand and the restrictive constraints to guarantee
environmental sustainability can decrease the efficiency of airports, often
considered as one of the engines for the economic growth of a region, thus
undermining the overall competitiveness of the European economy.
As recent traffic estimates report, over 60 European airports will be heavily
congested and the top 20 airports will be saturated at least 8–10 hours per day by
2025, if any efficient policy to increase the capacity levels is undertaken. The
worsening of the available capacity levels, mainly at hub airports, has serious
impacts on the overall efficiency of the European air transport industry. In fact,
critical congestion at airports means not only that airline schedules cannot be
assured, but also that environmental and safety costs increase.
The actions that can be undertaken to overcome these problems can be
identified in the more efficient use of the existing runways and the support for
new infrastructures as well as a suitable balance between market-driven solutions
(as market mechanisms for slot allocation) and regulatory measures (as Single
European Sky and airport safety management). In any case, aviation security
32 DEVELOPMENT OF REGIONAL AIRPORTS

must be always assured whichever the actions undertaken to increase airport

capacity.
In this overall perspective, regional airports are a key element to the
development of an integrated European air transport network, because they can
benefit from existing unused capacity and, if well linked to the main land
transport connections (both roads and railways), can be a suitable alternative to
more congested main airports.
Furthermore, they can overcome the slot assignment problems, very crucial
for coordinated airports. In the current EC Regulation – Article 2(a) [3], a ‘slot’
is defined as ‘the permission given by a coordinator in accordance with this
Regulation to use the full range of airport infrastructure necessary to operate an
air service at a coordinated airport on a specific date and time for the purpose of
landing or take-off as allocated by a coordinator in accordance with this
Regulation’. According to the Regulation, Community airports with a serious
loss in capacity should be designated as ‘coordinated airports’ on the basis of
objective criteria after a capacity analysis has been conducted, taking into
account environmental constraints at the airport itself. In other words, in a
coordinated airport time windows are assigned by a coordinator to airlines for
taking-off/landing (slot allocation); so, air carriers can access airport facilities for
landing and taking-off at specific dates and times for the duration of the period
for which the permission is granted. Only emergency and humanitarian flights as
well as government flights are excluded.
According to Article 3 of the EC Slot Regulation, in any specific time period
coordinated airports identify the number of slots that are offered and they are
divided into arrival and departure slots. Such a partition takes into account the
impact of other operational constraints, such as aircraft size and capacity, and
maximum possible passenger throughput. The slot management and assignment
can have significant consequences on the capacity of coordinated airports, and
such a process should be based on assignment criteria inspired by transparency
and unbiased principles.
In congested airports, delays in arrival/departure of aircraft means that slot
availability is no more assured and then delays can extend to several flights with
increased costs both for users and airlines. Delays predictability is then crucial
both for airlines and airports in their operation management. Generally, airlines
consider in their schedule a ‘buffer time’ to take into account unexpected delays
in arriving or departing aircraft. Anyway, this enlarged time means also an
inefficient use of the time resource and then a worsening in the overall airport
capacity. As an example, 5 minutes of buffer time means a cost of about €1000
million in the use of airline and airport resources.
In terms of slot allocation systems, at airports where capacity limits exist, a
slot coordination committee allocates slots on the basis of some criteria, mainly
on the basis of ‘grandfathering’ rights. It means that slots are assigned to
incumbent airlines according to their prior use of these slots. This method has
been strongly criticized and some countries have introduced the rule ‘use it or
lose it’, that is airlines lose their slot if they do not use it above 80% of the time.
This measure tries to avoid that incumbent airlines can take strategic action in
 DEVELOPMENT OF REGIONAL AIRPORTS IN EU 33

order to discourage the entry of new competitors. The ‘grandfathering’ slot

allocation system perpetuates the market presence of an air carrier at an airport
without ensuring that the air carrier will use them in the best way (i.e. offer
competitive services to the largest possible number of passengers).

2.2 The Single European Sky (SESAR) project

As introduced before, the SESAR project is aimed at designing and

implementing a single ATM Master Plan (including technology and operational
aspects), based on the Interoperability Regulation; in fact, air transport is
international by nature, so an ATM global interoperability is needed, for
economic and technical reasons, and above all for safety reasons.
The SESAR programme consists of two phases:
● the definition phase (2005–2007);
● the implementation phase (2008–2020).
The definition phase has been designed and co-funded by both the EC and
Eurocontrol (the European organization for air navigation safety) in order to
define a European ATM Master Plan. In turn, such a plan consists of:
● a technology roadmap, with target dates for development and introduction of
specific systems;
● a proposed system architecture;
● a detailed funding and implementation plan.
The implementation phase will consist of two steps:
● during the first step (2008–2013) the technologies underlying the new
generation of systems should be developed together with functional
enhancements in areas such as evolution of automated support tools and task-
sharing between ground and aircraft;
● during the second step (2014–2020) the new systems will be used at large
scale as well as their enhanced functional capacities. It is expected that the
new ATM system will treble current capacity, increase safety by at least a
factor of ten and will have an operating cost significantly less than today.
In order to extend the potential of a single sky in a larger environment, SESAR is
thought as a worldwide project. In fact, many agreements are expected with third
countries and Partner States. Cooperation agreements with third countries will
help to synchronize the SESAR technological and operational choices with other
modernization initiatives. On the other hand, Partner States will be invited to
become members of the SESAR Joint Undertaking, to contribute to the
technological programme, as well as other potential Single Sky activities.
To reach the goal of a Single European Sky, the European technical and
organizational infrastructure needs to evolve; the main aspects involved in the
identification of a Single European Sky, particularly in what concerns the
European ATM, are as follows:
34 DEVELOPMENT OF REGIONAL AIRPORTS

● the separation of regulatory activities from service provision, and the

possibility of cross-border ATM services;
● the integration of airspace into Functional Airspace Blocks (FABs), defined
in line with operational traffic flows and no longer constrained by national
borders;
● the crucial role of the EC in setting common rules and standards, covering a
wide range of issues such as flight data exchanges and telecommunications.
The SESAR programme is also addressed to develop new technologies aiming at
further increasing the safety and efficiency of airport operations. In particular,
both goals are linked to the separation between aircraft and the relating control
devices that can help in assuring and monitoring such a separation.
Separation between aircraft depends on wake vortexes that in turn depend on
the aircraft dimensions. As well known, the increase in separation will increase
the safety level, but will decrease the airport capacity. Then, an optimal
agreement between safety and capacity exigencies has to be achieved. New wake
vortex prediction and detection devices will enable to safely reduce separation
minima between aircraft while new generations of airport airside management
tools will enable the optimization of ground movements. Finally, new sensors
will enable remote tower operations. Then, the final goal to achieve both safety
and efficiency in airports can be obtained.
Furthermore, Advanced-Surface Movement Ground Control Systems (A-
SMGCS) surveillance and control functions should be implemented throughout
European airports, since safety for the aerodrome airside includes not only the
infrastructure but also operations and management.
As known, a SMGCS can be used not only to control the ground circulation
in order to guarantee suitable safety standards, but also to optimally manage the
ground movement (Postorino et al. [4]), that is by reducing possible excessive
spaces among aircraft and addressing the moving vehicles along the best path in
order to optimize the whole system in terms of capacity. An A-SMGCS must
also answer to requirements such as surveillance, routing, guidance and control
functions.
Finally, to conclude this section a brief overview about Galileo is needed.
Galileo is the future European satellite navigation system, planned to be
operational by 2013 and is to be included in the Trans-European Network (TEN)
(see Section 2.3) as a major technological project for the industry. It will be
formed by a global constellation of 30 satellites, distributed over three planes in
medium Earth orbit, and ground stations providing information concerning the
positioning of users in many sectors. It is expected to give a great contribution to
the development of technologies, organization methods and industrial
components which can ensure the safety and fluidity of air transport in the next
20 years.
As the White Paper [5] (Part 4 ‘Managing the Globalisation of Traffic’)
indicates, the Galileo programme represents a key need for global transport
programming and for removing the European dependence on GPS (USA) and
GLONASS (Russia) satellite navigation systems. Actually, Galileo is aimed at
 DEVELOPMENT OF REGIONAL AIRPORTS IN EU 35

ensuring interoperability with the previous two systems, but it is also an

independent system, in the sense that it will not be subjected to any limitation as
introduced in the other two systems because of military considerations.
Galileo will support Intelligent Traffic Systems, mainly in order to provide
navigational functions, real time information to users and optimized control to
urban and interurban transportation. Positive impacts of the provided navigation
and information services are expected, particularly: decreased travel times,
increased capacity of transport links and increased safety.
The availability of an additional satellite constellation offers significant
benefits to aviation in terms of improved performance and robustness,
particularly resistance to unplanned interferences, greater availability of
satellites, robustness to control segment failures, improved availability and
continuity.

2.3 Trans-European Network Transport project

The first plan of a TENs was born at the end of the 1980s, with the idea of
making available a free movement of goods, persons and services among the
Europe. This is possible only if the various regions and national networks that
form the single market are suitably linked by modern and efficient infrastructure.
The construction of a TEN-T (Figure 2) is an important element for the
economic growth and the creation of employment in the EU (European
Commission [6, 7]).

Figure 2: Trans-European Transport Network (TEN-T).

Source: European Commission (ec.europa.eu/ten/transport/maps/axes_en.htm).
36 DEVELOPMENT OF REGIONAL AIRPORTS

Furthermore, as transport has always been a key element to guarantee economic

and social cohesion, the identification of a TEN is crucial for the creation of the
internal market. Given that the EU process towards the unification is still in
progress, it is important to establish the interconnection and interoperability of
national networks as well as access to such networks.
In this context, airports are not merely air transportation infrastructures, but
rather they should be considered as inter-modal nodes in a larger multimodal
transportation network. Actually, for the vast majority of travellers, airports are
multimodal interchange nodes. Similarly, many freight transporters are using air
transportation as one of the links of their trip chain and, again, airports are
serving as modal interchange points.
As airports are generally located far from city centres, passenger or freight
movements involving air travel require the use of other transportation modes to
reach the airport. Generally, two secondary trips can be identified: the first one to
access the airport to initiate the air transportation component of the overall trip,
and the second one to move from the airport at the end of a flight to the final
destination. In that sense, all airports are multimodal interchange nodes.
Anyway, more in-depth analyses can prove different levels of complexity in the
interchange role of an airport. In fact, in some cases the trip to/from the airport is
simply the access and egress from the air transportation network, as air travel is
the primary mode. In other cases, there may be more than one primary mode
(e.g. local transportation to an airport, a flight, an intercity rail journey and local
transportation to the destination).
Furthermore, the growth of low-cost air services at some secondary airports
has significant implications for the local surface transportation. Many of the
airports used by low-cost carriers are generally far from the final destination city
(e.g. Frankfurt/Hahn is about 100 kilometres from Frankfurt). In a very small
number of cases there are limited local train services but, in general, access and
egress is by road transportation. This can lead to congestion on routes that do not
have the capacity to cope with the additional traffic flows, cause road damage on
pavements not originally designed for such flows and impose an environmental
burden on areas that were previously relatively quiet.
In order to speed up the integration progress among the several EU nations
and to improve accessibility, the EC has provided a large amount of specific
funds that, together with the Structural and Cohesion Funds, could be used as
leverage for national funding. Furthermore, specific attention has been addressed
to cross-border projects in order to obtain a more effective integration among
neighbouring nations.
The TEN-T actions are periodically revised to take into account new
exigencies and projects already finished. Main important corridors have to link
both Northern and Southern European regions as well as Eastern and Western
countries. Together with the land network extension, both in terms of road and
railway infrastructures, also the ‘Motorways of the Sea’ are key targets within
the TEN-T framework. They should ensure regular, high-capacity ferry routes
between the main ports in the EU, in order to improve the efficiency and
reliability of freight transport and then provide suitable alternatives for congested
 DEVELOPMENT OF REGIONAL AIRPORTS IN EU 37

land routes on roads. Their benefits are expected in the economic and social
fields in terms of improved cohesion between the involved countries: enhanced
accessibility; reduction in transport costs and times; improvement of quality;
creation of employment; access to new markets.
One of the most important consequences of the TEN-T actions is the
expected increase in the accessibility among the several EU nations and, inside
the same nations, among regions. Conventional accessibility indicators measure
the total effect of both geographical location (periphery vs. centre) and quality of
transport provided by the transport system. Decentralized regions have
disadvantages linked to their geographical position, but the inhabitants of these
areas should reach relevant destinations with the same travel speed as the people
in the central regions.
The TEN-T’s actual effects on accessibility are still unknown, because some
projects can increase the accessibility of some regions, but they do not improve
the accessibility of other neighbouring regions. Anyway, a more integrated land
network could be a successful element to increase the accessibility to airports,
mainly regional airports, thus starting the development of such airports and a
more effective distribution of demand over the EU territory.
Anyway, the land transport infrastructure facilities required for airports
depend on how the airport network develops and then which is the more
suitable European integrated network. Development of airports in turn depends
on the development of the airline market and the policies adopted by local and
national governments.

2.4 Air-rail competition/integration

As addressed by the EC, air and rail transport systems should become more
complementary. Actually, there is a need for efficient co-modal infrastructures
and modal split for airport access should also be improved. The EC exhibits
particular attention towards projects concerning inter-modal infrastructures that
are encouraged by means of the European Cohesion Policy and continue to be
eligible for financing under the European Regional Development and Cohesion
Funds, still available for the period 2007–2013. The Commission also invites
Member States to support the development of inter-modal inter-changes at
airports (rail links to and railway stations at airports), which promote efficiency
of both rail and air transport.
Anyway, the point of view in most adopted policies (both at national and
over-national levels) is to establish competing high-speed rail services to meet at
least some of the current and future demands on some links, in order to substitute
air services with rail services.
At least three interfaces between air and rail exist that can produce positive
impacts on the environment:
● links to the city, that can produce decongestion of road traffic and better air
quality around airports;
● links to the region, that can have the same benefits as above and the
additional benefit of the airport catchment area expansion;
38 DEVELOPMENT OF REGIONAL AIRPORTS

● link between the airport and major metropolitan areas by means of high-
speed rail with the same benefits as above and additional potential for short
haul slots to be freed for long-haul flights, which for airports and air carriers
represent higher slot productivity.
At the moment, the railway European network can be classified as conventional
rail and high-speed rail (Figure 3).

Figure 3: The European main railway network (2008).

Conventional rail can play an important role to connect secondary and regional
airports; where such links lack, they should be planned by Member States and
Community funding can also be done to realize these infrastructures.
As defined in the European Directive on Interoperability, high-speed rail is a
system where trains run at a maximum speed of at least 250 kilometres/hour. The
boundary between high-speed and conventional rail can also be unclear, as some
conventional trains in the UK and Sweden achieve better average speeds than,
for example, high-speed trains in Germany even though the latter achieve better
maximum speeds.
The opening of high-speed lines, such as Paris-Lyon and Madrid-Seville, has
enabled rail transport to obtain significant market shares on routes where time-
sensitive passengers would previously have travelled by air. Anyway, due to the air
liberalization process, the expansion of low-cost airlines has produced on some
routes, particularly in Germany and the UK, prices for air transport similar to or
below rail transport fares that could reverse the switch in market share. In addition,
the construction and maintenance of high-speed rail lines require significant public
 DEVELOPMENT OF REGIONAL AIRPORTS IN EU 39

funding. But the role of high-speed lines is more complex, because they can act not
only as competitors against air transport, but also as complementary mode to
increase the catchment area of an airport, mainly hub or primary airports. Then,
conventional rail can increase the connections between secondary and regional
airports enabling them to split air traffic according to their policies; on the other
hand, high-speed rail can be seen as a key factor to obtain significant decrease in
air congestion, as they can replace air links on some routes. Finally, they can
increase the accessibility at the main airports and provide additional demand from
a larger catchment area mainly for long-haul trips, as for short trips high-speed
trains can replace the air mode. As an example, two European airports, Frankfurt
and Paris CDG, have high-speed rail stations at the airport and there is the potential
for rail and air services to complement each other rather than compete. Instead of
taking a short distance flight to the airport, in order to connect on a longer distance
flight, passengers can travel by high-speed rail to/from the airport. As integrated
transport system network should also provide fare integration, on some routes
passengers can buy tickets which include both a rail sector and the air sector.
However, the attractiveness of such an integrated system to air passengers is
limited if they cannot check-in their luggage at the station and obtain a single
electronic ticket for the combined journey, due to the severe measures undertaken
by the EC in terms of security. On the other hand, the absence of integrated air–rail
tickets can be seen as an obstacle to the further development of air/rail inter-
modality in terms of services and passenger interest.
Finally, air/rail inter-modality cannot be considered as a primary way to
decongest airports as the decongesting effect amounts to 1 or 2 years of air traffic
growth. However, it is useful to achieve a greater efficiency of the transport
system and in particular of airports where the environmental burden will be
reduced.
To encourage rail as a complement for air, attention should be given on
improving the attraction on the rail product for both point-to-point and transfer
journeys.
The principal drivers for passengers when choosing a mode of transport are
relatively stable and consistently important across Europe. As investigations on
passenger behaviour show, users will choose a transport mode among a set of
available modes that match their trip exigencies when time, fare, frequency,
access and also information, ticketing, languages, service integration and other
issues, offer them an advantage with respect to the alternative modes. Then,
users will choose the rail option with respect to air if its characteristics are
overall better. Inter-modal development should therefore understand the
passenger market choices.
The main factors determining air and rail market share can be identified in
the following:
● the operating costs of each mode;
● the expected trend in market share and operating costs in the next 5–10 years;
● the security aspects concerned with the introduction of through baggage
handling and e-ticketing;
40 DEVELOPMENT OF REGIONAL AIRPORTS

● reliability;
● terminal accessibility.
For the last two points, as an example, the excellent reliability of the high-speed
trains in Spain was a key factor determining the high market share
of the Madrid–Seville rail service. Similarly, the good location of the airports on
the Madrid–Barcelona route contributed to high air market share; in contrast, the
relatively poor location of the airports on the Paris–Marseille route contributed to
high rail market share.
The choices of passengers between air and rail are then driven by the
combination of the relevant attributes as times and costs, where times include
trip time, waiting time at terminals, access/egress times and so on. In terms of
costs (as air/rail fares, parking or bus tickets to reach the terminal and so on) the
belief here is that rail is generally less costly than air. This is true in principle,
mainly if the comparison is between conventional trains and full-service (or
traditional) air carriers, while the difference is much less significant (and in some
cases it is reversed) if the comparison is between high-speed rail services and
low-cost air carriers.
For example, for the London–Paris and London–Edinburgh routes, the rail
mode is on average the cheapest option (Steer Davies Gleave Report [8]).
However, while rail fares on the London–Paris route were nearly as high as those
of the main traditional operators, on the London–Edinburgh route, low-cost air
carriers offer on average better fares than the rail operator. This could be the
main reason for the low rail market share on this route, as airlines on the
London–Edinburgh link offer relatively lower fares.
As investments in multimodal interchange nodes and the connecting road and
rail links have a long time-horizon, then the actual potential for competitiveness
vs. integration should be carefully evaluated.
To conclude this section, another important project that can further increase
the land accessibility among neighbouring nations within the EU is the
innovative European Rail Traffic Management System (ERTMS) that introduces
digital technology for the European rail infrastructure. As for the air transport,
there are different signalling systems in the cab of the train for the various
national networks, but ERTMS standard train protection system will greatly
simplify and speed up the technical interoperability of cross-border transports.
Further advantages will be the increase in safety standards to a high common
level throughout the EU as well as the increase in the capacity utilization of the
existing rail network (ERTMS [9]).

3 Airports in the EU perspective

The role of airports in Europe is a matter of discussion, mainly because of their
double nature: as important node of air network (but also as node of inter-modal
networks) and as source of employment and potential economic engine for a
region.
Solutions to increase airport safety, capacity and efficiency while remaining
environmentally friendly are the main goals of the EUROCONTROL Airport
 DEVELOPMENT OF REGIONAL AIRPORTS IN EU 41

Operations Programme (APR), first launched in January 2002 and then, due to its
success, in January 2007.
As recent data provide, in 2006 in Europe there were about 2100 airports,
with a great part being regional and secondary airports. They are spread on the
territory, so potential passengers can reach them on average in 90 minutes. The
cities closest to European busiest airports have between 4 and 46 airfields within
100 kilometres from the city centre.
The largest European airports get 80% or more of all departures within 100
kilometres (EUROCONTROL [10]), the usual distance of these airports from the
city centres being about 20–25 kilometres. In terms of distribution on the
territory, large cities are far from large airports no more than 50–150 kilometres
in Northern Europe, while in Southern Europe the separation among the main
cities, and then large airports, is larger. As an example, the main airport serving
Madrid is 13 km from the city centre, but the next airport with more than 100
departures/day is 290 kilometres away, at Valencia.
In terms of ownership, many airport types can be identified, due to different
historical and political perspectives in the EU Member States and the Accession
countries. Several ownership categories can be recognized, but there are also
mixed ownership (public/private) whose difference is in the proportion of
national and local government involvement. To give an overview, the airport
ownership types can be identified in the following:
● national government;
● regional/local government owned;
● other state sector ownership;
● chamber of Commerce;
● privatized airport/airport group;
● mixed type: majority government ownership;
● mixed type: minority government ownership.
Within the different airport categories, regional airports are a potential challenge
for the European air network. In fact, they can provide additional capacity in
areas where major airports have become congested, but given their inherent
small nature they may not be making sufficient revenue to cover their costs.
These two points are discussed in the following.
Regional airports have become progressively more important in the European
aviation network as air demand has increased and many new airlines, particularly
low-cost air carriers, have begun to operate at small airports. Then, from one
hand regional airports have increased access to aviation markets for a larger
proportion of the population that otherwise could not use the air mode. On the
other hand, they have increased competition among airlines because of their
attractive potential for many air carriers, thus leading to lower fares, increased
frequencies and more served destinations. More factors can be identified as
crucial for the progressive development of regional airports, as the growth and
expansion of the number of regional carriers, the rise of the low-cost carrier, the
conversion of military airfields into commercial service airports and the
increased use of larger regional jets.
42 DEVELOPMENT OF REGIONAL AIRPORTS

Anyway, the liberalization process has produced remarkable changes not only
in the airline industry but also in the airport industry, although most of them have
maintained their profitability from both location and, in some cases, monopoly
rents. For regional airports, however, the question is more complex. Although the
number of movements in such airports has increased significantly as well as the
number of passengers and/or goods, however the absolute numbers are quite low
and then they may have insufficient revenue to cover their costs.
According to the airport European data, there is significant variation across
both cost and revenue performances. Some airports have high costs but also high
operating revenues. Efficient airports can be considered those providing
economically efficient levels of output; anyway, for airports the output is a
combination of aircraft movements and handled passengers, so different airport
rankings are obtained when comparing cost per passenger vs. cost per
movement.
Another aspect concerns the potential competition among airports and then
which is their effective role with respect to the air services generally they offer.
As nodes of the air network, airports provide links between locations at given
times and costs (also due to the airline offer) and they compete for passengers by
competing for airlines and airline services. The passenger airport choices are
based on the combination of airport and airline characteristics (see also Chapter
5). Airlines, in turn, will choose the airport where to start the service on the basis
of many factors including network fitness and airline business model.
Regional airports will compete on comparable full price over their catchment
area in order to attract a level of demand able to increase the revenue, both due to
aviation and non-aviation services and, as some analyses show, there is an
increasing level of competition between regional and secondary airports as well
as between the secondary airports themselves.
To start a convenient process of airport competition, there should be effective
available alternative airports and the willingness of airlines to compete with the
air services offered at the current dominant airports. But also passengers should
be willing to move from the dominant airport to the potential competitive one, on
the basis of the air services they can obtain at the latter. Airport competition has
the potential to expand further if factors such as attractiveness of the low-fare
airlines and of the airports they serve combine together to become appealing for
passengers.
The main attractive factors for airlines are:
● the reduced airport turnaround times (e.g. 25–30 minutes on average) that
enable airlines to increase the productivity of planes and their crews when
compared to the slot-constrained hub airports;
● the lower airport charges they can obtain at secondary and regional airports.
Other factors that encourage airlines to introduce air services for new city-pair
markets are:
● potential traffic and yield at each end of route;
● availability of slots;
 DEVELOPMENT OF REGIONAL AIRPORTS IN EU 43

● existing competition offered by other airlines on route and indirect routings;

● operating costs not dependent on airport operators.
Attractive factors for passengers to move towards a new airport are not only
reduced air fares (generally offered at secondary or regional airports), but also
reduced delays (as these airports do not suffer from traffic congestion) and direct
point-to-point services (that avoid waiting time for transfers at an intermediate
hub airports). Finally, when a delay occurs at smaller airports it is generally
worse than at larger airports, but delays at smaller airports are relatively
infrequent (actually most of the total delay takes place at large airports) and this
may represent a compensation for the lower frequencies generally offered to
passengers.
The choice of passengers to move towards an airport and airlines to start new
services can be helped by the airport that can play an important role in stressing
the attractions of the airport infrastructure and surrounding area for business and
leisure travellers, but also can influence such decisions directly by offering good
facilities (parking, surface access, check-in, lounges, shopping, gates, and
sometimes handling services), and suitable cost of landings and take-offs.
When two or more airports share a similar catchment area, the airline decision
to introduce the service at all and to use a given airport will strongly depend on the
airport offer. Airports close to the same urban area are in the best position to
compete with each other, given their proximity to the commercial and residential
centres of the city; in this case, great relevance assumes the identification of
their overlapping catchment area, where two or more airports (or, better, airlines
operating at those airports) compete for the same air demand (see also Chapter 4).
In terms of connection and frequency, scheduled traffic at an airport grows by
adding connections and by increasing frequency on a small proportion of those
connections. Two kinds of offered air services can be identified as hub-an-spoke
vs. point-to-point.
The hub-and-spoke system is used by many full-service air carriers as the
preferred one in order to increase the load factor on average and to gain a
monopolistic power at airports. A hub-and-spoke system is characterized by a
high spatial network concentration, a time coordination of flights at the hub,
according to a ‘flight wave’ concept, and the integration of the air services at the
hub (as baggage transfer). The coordination at hubs means that delays in only
one flight can have consequences on many other flights (e.g. crew or aircraft
unavailability). An airport can play the role of ‘hub’ for a hub-and-spoke
network if it is centrally located with respect to the served market and there
is a significant demand starting/ending trips at hubs. Furthermore, it should
guarantee suitable facilities to accommodate large aircraft and the baggage
transfer, as well as all the accessory facilities required by airlines in doing a hub-
and-spoke service.
Point-to-point services are simply routes serving a given city-pair where there
is a significant demand to guarantee the accomplishment of the break-even load
factor. In this case, passengers can fly from an origin airport to a destination one
without scheduled transfers at intermediate airports and then without loss of
44 DEVELOPMENT OF REGIONAL AIRPORTS

accessories time (e.g. due to baggage recovery and waiting time). Generally, the
used aircraft are smaller at the first time and larger if demand increases after the
service has started; the flown distances are within 400 kilometres from medium-
sized airports, while smaller airports can still have shorter flights.
There is a general agreement that the trend is towards smaller aircraft for
intra-European services if market trends will continue. That could imply more
point-to-point services and the growth in the use of smaller, regional airports. A
point-to-point increase in passenger travel would require the extension and
upgrading of smaller airports and a redesign of the network.
Anyway, the question is whether the changes in the market structure
observed in the airline industry will be permanent, or they are simply transitory
stages on the path towards a different, long-run equilibrium. Two opposite
beliefs are proposed: some airport network planners state that the determinants of
the hub-and-spoke system are not set to decrease in the near future and the
current factors promoting the growth of point-to-point services may weaken in
the next future; on the contrary, some others seem convinced that point-to-point
services will predominate in the long run. This latter point of view is also
encouraged by some studies carried out in the USA on the development of the
US airport system that predict a huge increase of point-to-point links by 2020
with almost all passengers being offered direct flights to their destinations, an
improvement in services and a reduction in the number of takeoffs and landings
relative to the volume of transport services.
Some final considerations about the tendencies concerning the development
of airports in Europe, and mainly regional airports, is the question of State aids,
funding for transport links and regulation measures.
State aids are intended as local, regional or national government measures to
assist airports (and airlines), and they may fall within Article 87(3)(a) of the EC
Treaties if their location is within an eligible region, thus being particularly
relevant for regional airports. EC Guidelines on regional airport State aid affirm
that regional airport operators have to inform the Commission about any public
finance they receive from regional or national authorities, as well as by means of
the EU funds. Notified aid is then examined by the Commission and if there is
doubt that certain investment at a regional airport constituted unauthorized State
aid, the Commission can start a formal investigation against the concerned
airport.
In the context of subsidies for airlines that realize a public service for some
specified routes which they would not normally operate on economic basis, there
are some specific indications. Such subsidies can be allowed in two cases:
● when other transport systems cannot guarantee a suitable accessibility to
some regions, the local or central authority identify some scheduled air routes
to serve airports located in peripheral areas or developing regions, or
alternatively low-demand routes provided that such measures can be
considered crucial for the economic development of a given geographical
area (public service obligation, as defined in Council Regulation 2408/92,
 DEVELOPMENT OF REGIONAL AIRPORTS IN EU 45

currently under revision); such subsidy then serves to reimburse a carrier

selected by a public tender for performing the required public service;
● where the subsidy has a social character granted to specific categories of
passengers on a route granted without discrimination on the grounds of the
carrier operating the services.
Another aspect considered in the Guidelines concerns an airline that can receive
significant financial advantages by obtaining from the public authority, or the
entity that operates the airport on behalf of the authority, exclusive concessions
for a market price lower than their actual market value. Although the exclusive
concession may be given to an airline for a price lower than its market value, if
the airline pays no fee for the exclusivity or the fee is lower than would be paid
under normal commercial conditions, that should be considered an aid element.
State aids to airports may have as consequence a distortion in competition,
both among airports and airlines. In fact, direct or indirect subsidization of
airports can distort competition between airports themselves, while subsidization
of the airports could indirectly benefit airlines, thus distorting competition
between them as the aid enables airports to facilitate specific airlines by means
of, for example, lower airport charges.
Funding for transport links, in order to connect airports to the land transport
systems, falls within the EC Guidelines on finance for infrastructure, open to all
potential users without discrimination. As already discussed, these improved
transport connections may extend an airport catchment area, thus improving its
competitiveness; funds from the Community TEN-T can also be used by airports
for improving inter-modal links following the wide Community transport policy.
This kind of funding does not constitute State aid as the funds derive from the
central EU funding, and then the construction of a high-speed rail link, rail-air
station or a people-mover link to an airport terminal is unlikely to be considered
a State aid.
Finally, regulation should be confined to those activities in which the airport
has constant monopoly power, as airport services essential for downstream users
that cannot be duplicated without considerable costs. The airside system,
included aprons, and the passenger and freight handling terminals are generally
regarded as such services. On the contrary, non-aviation activities and ground
handling are seen as activities in which, at a given extent, the airport might have
some monopolistic power. As the air liberalization process extends to several
features, also ground handling was liberalized in 1996, thus enabling third party
providers to enter the market. Ground handling services should not be regulated,
but central infrastructure services such as package handling systems should be
part of the regulated activities. In countries where airports do not offer ground
handling (as in the UK) this does not represent a problem, but in the majority of
the European Nations (as Italy, Spain, France, Austria, Germany) airports
provide ground handling services and then the way towards the effective
liberalization will be longer.
46 DEVELOPMENT OF REGIONAL AIRPORTS

4 Low-cost carriers and regional airports

The European deregulation has generated a massive expansion in a segment of
the airline industry known as ‘no frills’ or ‘low-cost’ airlines. There has been a
significant entry of new airlines offering single-class services on a point-to-point
basis. Low-cost air carriers have risen quickly in the last years, thus making the
air transport system more accessible to many users, mainly to young and low-
income people, but today also businessmen are attracted by the competitive fares
proposed by low-cost air carriers. While the market share of these low-cost
carriers was still relatively low in 2000/2001, the number of successful carriers
increased noticeably.
In Europe, the growth of low-cost and regional carriers has led to a greater
use of secondary and local airports with few or no international connections prior
to deregulation. On the other hand, it can be argued that the expansion of the
low-cost carrier is mainly linked to the available overcapacity at regional
airports.
The low-cost models adopted by air carriers to reduce their operational costs
in order to maintain a high competitiveness in terms of air fares, can be different
among the different companies (Gillen and Lall [11]), but some common factors
can be identified, also with respect to full-service air carriers, mainly in the
commercial, organizational and technical fields.
In terms of commercial aspects, low-cost carriers achieve significant savings
by selling their tickets via Internet or in same cases by means of payment call-
centres. Furthermore, the ticket is just a reference code (sent by e-mail after the
flight has been purchased) that has to be presented at the check-in point together
with the passport in order to receive a boarding card, often without seat
reservation. Finally, low-cost air carriers generally do not adopt fidelity
programmes (as frequent flier programmes).
In terms of organizational aspects, low-cost air carriers have reduced all the
services not included in the transport function, as on-board meals, magazines
and drinks that are not included in the fare and can be bought on the airplane
during the flight. Cabin crew and ground staff are also reduced and many
services considered unessential for the development of the company are
outsourced. Furthermore, they offer medium distance flights, generally intra-EU
for the European case: on average, flights cover about 800 kilometres. The
service is organized on a point-to-point base, thus avoiding baggage and
passenger transfers and maintaining a low turnaround time. In fact, smaller
airports can offer only limited opportunities to transfer to other flights, while
major airports may suffer from delays which can make connections unsure.
Generally, frequencies are not high because low-cost air carriers frequently offer
air services between low-demand, secondary airports, although, in many cases,
with a great potentiality (as Hahn for Frankfurt, Beauvais for Paris, Charleroi for
Brussels, Bergamo for Milan). In this light, secondary, or regional, airports have
great advantages for low-cost air carriers and also for passengers in terms of
reduced waiting time in the baggage claim area and walking time inside the
terminal as well as less congestion inside the airport. The main disadvantages for
 DEVELOPMENT OF REGIONAL AIRPORTS IN EU 47

passengers of regional airports are the higher access/egress times between the
city-centre of their actual destination and the airport itself.
In terms of technical aspects, low-cost carriers have low turnaround time in
order to use more efficiently their aircraft. As an example, turnaround times for
low-cost carriers are about 25–35 minutes, while for full-service carriers they
range from 45 to 60 minutes. Then, generally low-cost carriers can use their
aircraft 30% more than full-service carriers thus realizing more flights per
aircraft. Furthermore, for intra-EU flights the number of flown hours per aircraft
is on average 9 per day for full-service carriers and about 10–12 for low-cost
carriers. Finally, full-service carriers generally use a mixed fleet while low-cost
carriers use a unique type of aircraft (generally B-737). This enables the air
carriers to spend less for crew training and to have significant savings in terms of
aircraft maintenance.
The first significant example of low-cost air carrier in Europe is Ryanair. At
the beginning, Ryanair tried to offer the same air services as the full-services
airlines it faced (as British Airways and Aer Lingus), but with lower fares in
order to be more attractive for passengers.
However, the high operational costs produced a significant loss in the period
1985–1991 (apart from 1987) by convincing the airline managers to transform
Ryanair in a cash driven airline with a reduction of the offered services and the
adoption of strategies to limit costs of the kind usually identified as ‘low-cost’
models. Among the different measures, Ryanair outsourced all the passenger
services considered unessential for the mission of the company. Outsourced
services have advantages and disadvantages: from one hand, they enable the
company to obtain a quick economic expansion by offering services to
passengers without taking on staff working only on that specific aspect; on the
other hand, the relationships with the subcontractors are time consuming and
source of stress for the company, because a problem with one of the
subcontractors can produce discomfort to passengers. After, Ryanair adopted and
improved the low-cost/low-fare business model originally developed in the US
by Southwest Airlines, by going progressively towards a business model based
on single fleet type, use of uncongested airports, low turnaround times, point-to-
point services and on-line bookings.
As Ryanair started its services between the UK and Ireland as direct
concurrent of Aer Lingus (see, e.g. Figure 4) and then between the British Isles
(from airports such as Stansted, Luton, Manchester, Newcastle, Glasgow and
Dublin) and continental Europe, the impact of the low-cost segment was greatest
in UK, where the number of passengers carried by Easyjet, a new low-cost entry
following a similar low-fare approach, and Ryanair was 84% of the total British
Airways passenger.
The rise of further low-cost airlines gave (and still is giving) a higher profile
to European secondary airports and created new markets for air travel. For
example, within Britain congestion at London airports encouraged the growth of
many regional airports. Scotland Edinburgh Airport, for instance, increased
significantly its traffic after being identified as a low-cost destination by some
low-cost carriers.
48 DEVELOPMENT OF REGIONAL AIRPORTS

Figure 4: Aer Lingus and Ryanair passenger traffic on Ireland–London routes.

In contrast to Ryanair, EasyJet Airline has largely avoided secondary airports.

Apart from Liverpool in northwest England and the airline home base at London
Luton, EasyJet has concentrated on serving primary airports. Routes are
guaranteed on major destinations as Barcelona, Nice, Paris, Zurich, Geneva and
Amsterdam, and in the future of the company there could be a change towards
higher fares as EasyJet managers expect that people will pay more to travel to
main airports.
Nowadays, low-cost carriers are expanding services on continental Europe.
Ryanair is developing Frankfurt-Hahn and Charleroi as bases while EasyJet is
developing Geneva. The trend at the major airports has been towards a continued
pressure for additional slots but in some cases there has been a contraction in the
served destinations, perhaps because some airlines focused on their most
profitable routes due to some current economic situations.
An important aspect that has promoted the use of regional airports is the
opportunity for low-cost companies to obtain subsidies in the form of discounted
airport charges. In fact, some regional airports have undertaken this way to
stimulate traffic and then to increase the revenue. However, discounts offered to
airlines by airport operators may be considered State aid if they discriminate
against particular airlines. In some cases, the analysis of the circumstances by the
EC have stated that ‘discounted landing fees would not constitute State aid where
they were applied for a limited duration and were available to all airlines
operating from the airport, subject to their fulfilling objective criteria (i.e.
starting a new service to a new destination)’ (Manchester Airport case, EC). In
other cases, agreements between airport operators and airlines or the independent
application of a discriminatory policy by the airport operator have led to
discounted airport charges and, therefore, Articles 81 and 82 of the EC Treaties,
as opposed to Article 87, have been used to challenge discriminatory airport
charges. Such situations distort competition for airports that cover their costs and
may lead to inefficiencies for passengers and airlines. This was the key issue in
the Charleroi Airport–Ryanair case, where, following the EC, aid was made
 DEVELOPMENT OF REGIONAL AIRPORTS IN EU 49

available only to Ryanair. Anyway, a recent sentence of the European Court has
reversed the EC judgment.
One of the most controversial questions about the great success of Ryanair in
Europe mainly concerns the negotiation of the airport charges realized by the
managerial staff of the airline. Many Ryanair competing air carriers have
denounced the application of advantage fees to Ryanair by airport operators and
have asked the EC for a clarification. Actually, data for eight provincial airports
in UK indicate that Ryanair has negotiated airport charges with at least a two-
thirds discount on the Stansted charge.
Finally, regional airport development has been often associated with
passenger traffic. However, the growth in the air freight transport has led to a
similar re-valuation of the role of regional airports as the emergence of passenger
low-cost carriers. In fact, smaller airports, some of them being the former
military airports with only basic service functions, now serve as specialized
freight airports. Compared to primary and secondary airports with freight and
passenger services and international connections, they are located further away
from the larger cities, with far fewer connections to the supra-regional railway
and road network. But the same measures (and the same considerations), as for
regional airports used by passengers, can be adopted for these freight-specialized
airports.

5 Conclusions
Many policies have been undertaken by the EU to promote the development of
the transport system within Europe, following an integrated approach as stated
by the TEN-T project.
Significant progresses have also been realized in the air transport system, also
due to the urgency to achieve integration and harmonization among the different
aviation procedures in the several European Nations.
The considerable congestion at the main airports and the liberalization
process, enabling low-cost carriers to offer passengers lower-fare services, are
the main factors underlying the current development of regional airports.
Available capacity and demand potential have been key factors for the
progressive growth of more regional airports, also helped by the EU general
policy addressed to environment friendly transport strategies.
The significant growth of point-to-point flights offered by low-cost carriers is
also associated with the use of smaller airports, often far from metropolitan
centres or even metropolitan areas.
Low-cost airlines have also played a determinant role for the growth of both
air traffic and regional airports. At least to some extent, their expansion has been
made possible by overcapacity at regional airports. Furthermore, as there is little
extra cost involved in giving low-cost carriers access to unused capacity
available at airports, airport charges might also be low.
The question that can be posed is whether this will be the tendency in
Europe. In other words once spare capacity has been used up, airport costs in the
regional airports might increase substantially. Then, additional capacity can be
50 DEVELOPMENT OF REGIONAL AIRPORTS

provided on the basis of the demand for airport services at fares which cover the
full costs.
On the other hand, spare capacity is also due to competition among airports;
airports with larger capacities can obtain economies of scale while potentially
competing airports may outperform one another. Anyway, without a right planning
of the airport system the risk is to produce overcapacities often due to the
authorities themselves that try to stop the development of other potential
competing airports in order to attract business and the associated tax revenues.
Overcapacities, in turn, require government subsidies thus creating uneconomical
cycles. A right planning policy and a great coordination among local authorities
can avoid over-investment at regional airports, even if this could produce higher
airport costs for the carriers using these airports but also lower public expenditure
on subsidizing them. Investment decisions require either greater centralization or
mechanisms for cooperation between local authorities which would otherwise be in
competition.
In any case, as statistical data provide, the growth of regional airports, also
supported by the EU policies, seems unavoidable by now, but the increasing
importance of smaller airports (often decentralized with respect to the actual
destination city) in the overall airport network could lead to a general worsening
of the road transport intensity of the surface transport deriving from air travel.
Then, it is fundamental that the EU faces the problem of the right integration
among all the different transport modes, in order to achieve the expected result: a
multimodal connectivity in a really integrated transport network.

References
[1] Upham, P., Thomas, C., Gillingwater, D. & Raper, D., Environmental
capacity and airport operations: Current issues and future prospects. Journal
of Air Transport Management, 9, pp. 145–151, 2003.
[2] Upham, P., Raper, D., Thomas, C., McLellan, M., Lever, M. & Lieuwen,
A., Environmental capacity and European air transport: stakeholder opinion
and implications for modelling. Journal of Air Transport Management, 10,
pp. 199–205, 2004.
[3] Regulation (EC) No 793/2004 of the European Parliament and of the
Council of 21 April 2004 amending Council Regulation (EEC) No 95/93 on
common rules for the allocation of slots at Community airports, Strasbourg,
21 April 2004.
[4] Postorino, M.N., Barrile, V. & Cotroneo, F., Surface movement ground
control by means of a GPS–GIS system. Journal of Air Traffic
Management, 12, pp. 375–381, 2006.
[5] White Paper, European Transport policy for 2010: Time to decide,
European Commission, September 2001.
[6] European Commission, Multi-annual indicative TEN-T programme (MAP)
for the period 2007–2013, COM(2004)475, 14 July 2004.
[7] European Commission, Multi-annual indicative TEN-T programme (MAP)
for the period 2007–2013, COM(2006)245, 25 May 2006.
 DEVELOPMENT OF REGIONAL AIRPORTS IN EU 51

[8] Steer Davies Gleave, Air and Rail competition and complementarity, Final
Report prepared for the European Commission DG TREN, 2006.
[9] European Rail Traffic Management System (ERTMS), http://ec.europa.eu/
transport/rail/interoperability/.
[10] EUROCONTROL, A place to stand: Airports in the European air network.
Trends in Air Traffic, Vol. 3, 2007.
[11] Gillen, D. & Lall, A., Competitive advantage of low-cost carriers: Some
implications for airports. Journal of Air Transport Management, 10,
pp. 41–50, 2004.
This page intentionally left blank
3

The development of regional airports in Asia

Y-C. Chang
National Taiwan Ocean University, China

Abstract
Countries in Asia are more diverse politically, economically and culturally than
those in North America and Europe, with each of them differing in their
approach to deregulation. This chapter reviews and analyses the regional airports
development in Asia. It concludes that Asian airport operators will be facing a
challenging time ahead with air traffic growth at a reduced pace and volatility in
fuel prices affecting airline business. In addition, the LCCs continuing growth in
Asia, consumer need for point-to-point travel and scarcity of secondary airports
in Asia have opened up many opportunities for airport operators and investors.
Keywords: Taiwan; China; ASEAN; low-cost carriers.

1 Introduction
1.1 Characteristics of Asian air transport industry

Over the past few decades, Asia has grown more rapidly than any other part of
the world economy. Its trade and industrial transformation has been built upon
openness and integration into the world economy, with a continual process of
reform and liberalization in a succession of Asian economies. The strong
economic growth led to a rapid increase in the demand for air travel in the region
of Asia.
Total international scheduled passenger traffic to, from and within the region
grew by 7.4% per annum on average between 1985 and 2005. The global share
of Asia-Pacific international scheduled passenger traffic increased from 25.5% in
1985 to 28.3% in 2005. The airlines of Asia-Pacific region are expected to show
the highest growth in passenger traffic, at 5.8 % per annum through to the year
2025, while the airlines of North America are anticipated to show the lowest
growth, around 3.6% per annum. As a result of the projected growth in passenger
54 DEVELOPMENT OF REGIONAL AIRPORTS

traffic, the airlines of the Asia-Pacific region are expected to increase their share
of world passenger traffic (in terms of passenger-kilometres) by about 6.5%
points to 32.52%, the highest among all the regions, with their share of total
international scheduled passenger traffic increasing to about 33.7% (see Figure 1,
International Civil Aviation Organisation [11]).

Figure 1: Share of total international scheduled passenger traffic by each region.

Table 1 reveals that regional platforms enjoyed the highest year-on-year growth
on a percentage basis, up 7.3% against a 5.6% rise for Intercontinental hubs and
5.4% for Secondary hubs. Regional platforms in Asia/Oceania were still above
the world average and well ahead of their counterparts in North America and
Western Europe, owing mainly to the explosion of domestic traffic in China
and India but also, and increasingly, in Indonesia and Vietnam (Air Transport
World [4]).

Table 1: 2006–2007 Airport traffic growth by continental and airport type.

Unit: %
Airport type World Western North Asia/Oceania
Europe America
Intercontinental 5.6 5.5 2.3 8.2
hubs
Secondary hubs 5.4 6.1 4.1 6.3
Regional 7.3 6.7 4.2 10.7
platforms
Source: Air Transport World [4].
 DEVELOPMENT OF REGIONAL AIRPORTS IN ASIA 55

According to ‘World Airport Traffic Analysis 2007’ report, China and India
continued to pull demand with a 70% share of the region’s traffic volume
increase in 2007 (Figure 2). China’s airports handled some 395.4 million
passengers, up 16.7%. It accounted for 54% of the region’s growth or 17% of the
world growth. Throughput at Indian airports rose 19.9% to 97.6 million whereas
traffic in the region’s second-largest market. Vietnam and Philippines recorded
significant growth rates, up 25.8% and 20.4% respectively. The strong economic
development, particularly in China and India, with an emerging middle and
upper class has really stimulated air traffic demand. In addition, domestic
markets are fuelling country and regional growth. Domestic traffic accounted for
54% of China’s total growth in 2007 (Figure 3). In India, the imbalance between
domestic and international travel was even more pronounced; domestic traffic
accounted for 96% of the growth in passenger throughput in 2007. In Vietnam,
63% of the 3.9 million additional passengers in 2007 flew within the country
whereas in Indonesia domestic travel soared 118%.

Figure 2: Distribution of Asian airport traffic evolution by country 2007 vs. 2006.
(passengers in millions)
56 DEVELOPMENT OF REGIONAL AIRPORTS

Figure 3: Split of mainland China growth per destination.

1.2 Trend of major international airport development

Given the rapid economic growth in Asia-Pacific, the region has a large number
of busy airports, with twelve recording passenger traffic levels in excess of 20
million in 2007 (Table 2). Nine of the airports were ranked among the top 50 in
the world in terms of passenger volume. In order to meet the growing demand
and become regional hubs, substantial investment has been carried out in the
region in recent years. New Kansai Airport opened in 1994, Hong Kong Chek
Lap Kok Airport in 1998, New Kuala Lumpur Airport in 1999, Taipei CKS
second terminal in 2000 and Seoul Incheon Airport in March 2001, Japan
Nagoya Airport in 2005, second Bangkok International Airport in 2006, and
Singapore Changi airport terminal 3 in 2008. In particular, Beijing’s enormous
Terminal Three opened in time for the Olympic Games in August 2008.
In these major international airports, Seoul Incheon Airport is the hub of
Skyteam alliances in Asia; Singapore Changi and Bangkok Airports are the
regional hubs for Star alliances; and Hong Kong Airport is the hub for Oneworld
alliances.

1.3 Regional Airports in Asia

Due to the large number of airports in Asia, identifying the definition of regional
airports becomes a difficult task. This chapter uses the Air Transport Intelligence
Database [3] with the criteria, International Passengers less than one million, and
Domestic Passengers more than 80 millions in 2007. The result comes out with
47 airports as showed in Table 3. Among these nations, Indonesia is the first
country to liberalize its air transport sector. Indonesia is expected to have the
fastest annual growth in international traffic during the 1990s, followed by
Malaysia (Dempsey [10]). The government is planning to construct new airports
at Medan and Padang in Sumatra, and in central Lombok.
 DEVELOPMENT OF REGIONAL AIRPORTS IN ASIA 57

Table 2: Major international airport characteristics in 2007.

Total Main hub
Airport Passenger(000) Runways(metres)
Movements airlines
1*3,800
Beijing Capital 53,737 399,697 Air China
1*3,200
Shanghai
28,930 187,045 China Eastern 1*3,300
Pudong
1*3,600
Jakarta 31,987 248,482 Garuda Airlines
1*3,660
Tokyo Narita 35,530 195,074 Japan Airlines 1*4,000
All Nippon
Osaka Kansai 16,592 125,637 1*3,500
Airways
Korea Airlines 1*3,600
Seoul Inchon 31,422 213,194
Asiana Airlines 1*3,200
Malaysia 1*4,000
Kuala Lumpur 26,534 193,688
Airlines [14] 1*4,019
Philippine 1*2,225
Manila 20,468 188,797
Airlines 1*3,409
Singapore Singapore 1*4,000
36,702 223,488
Changi Airlines 1*3,260
China Airlines 1*3,660
Taipei CKS 23,426 160,120
EVA Air 1*3,350
1*3,700
Bangkok 41,210 265,763 Thai Airways
1*3,500
Hong Kong 46,995 305,010 Cathay Pacific 2*3,800
Source: Air Transport Intelligence Database.
In these regional airports, some are popular tourist island airports, like Haikou in
China, Bali in Indonesia, Jeju in Korea, Okinawa in Japan, Langkawi in
Malaysia, and Phuket in Thailand. Because of their excellent weather and
beautiful beach, these island airports attract lots of passengers each year. Bali is
the most famous tourist island in Indonesia. Its Ngurah Rai – Bali International
Airport started operation in 1931 by using a grass runway. The runway was
upgraded during the period of 1941 to 1947 and lengthened to 1600 metres. It
was named Tuban Airport as it’s located in Tuban Village. Haikou Meilan
International Airport is located on the southern island province of Hainan, which
is partly-owned by the Hainan Airlines Group. It maintained a double-digits
growth in passenger throughput, aircraft movements, and cargo throughput in
2008, and achieved the highest growth rate among the civil aviation airports in
China. Phuket International Airport ranks second in Thailand in terms of
passenger and cargo volume. Phuket Island and nearby provinces are popular
resort areas and Phuket International Airport enjoys a large number of
passengers from all over the world, and plays an important role in promoting
travel and tourism in the southern region as well as throughout the kingdom. The
airport is served by ten airlines, carrying more than 2.9 million passengers and
58 DEVELOPMENT OF REGIONAL AIRPORTS

12,000 tons of cargo on 27,000 flights (data of 2007). The airport is located 32
km from downtown Phuket.
Table 3: Asia regional airport characteristics in 2007.

Total Runways
Total
Country Airport passenger
movements length*width
(in millions) (metres)
Haikou 7.265 63,416 Length: 11811
Changsha 8.070 82,041 2600*45
Dayong 1.517 14,148 2600*45
Jinghong 1.808 17,506 2200*45
Nanjing International 8.037 82,391 3600*45
3000*60
China New Chitose 18.361 98,827
3000*60
Shanghai Hongqiao
22.633 187,045 3300*58
International
Shenzhen Baoan
20.619 181,450 3400*45
International
Xi’an Xianyang International 11.383 119,404 3000*45
Ahmedabad 3.038 39,280 3489*46
Amritsar International 0.677 7,463 3289*45
Gauahati 1.334 29,183 n.a.
India Goa 2.528 22,817 3430*46
Jaipur 1.263 24,434 n.a.
3164*46
Nagpur 0.839 15,880
1573*46
Ngurah Rai-Bali
5.888 62,689 3000*45
International
Banda Aceh 0.550 7,231 n.a.
Bandung-Husein
0.361 6,397 n.a.
Sastranegara
Medan 5.004 53,795 2625*45
Indonesia Padang 1.753 14,995 n.a.
Palembang 1.660 17,059 n.a.
Pekanbaru 1.839 22,818 n.a.
Pontianak-Supadio 1.379 17,898 n.a.
Surabaya-Juanda 3.488 87,587 3000*45
Ujung Pandang-Hasanuddin 1.600 51,698 n.a.
2500*45
Tokyo Haneda 66.823 331,818 3000*60
3000*60
Japan
Fukuoka International 17.903 71,456 2800*60
Kagoshima 5.591 n.a. 3000*45
Okinawa Naha 14.951 123,596 3000*45
 DEVELOPMENT OF REGIONAL AIRPORTS IN ASIA 59

Table 3: Asia regional airport characteristics in 2007 (continued).

Total Runways
Total
Country Airport passenger length*width
movements
(in millions) (metres)
Length: 3000
Korea Jeju International 12.297 93,285
Length: 1910
Penang 3.174 39,265 3352*45
Bintulu 0.381 7,093 n.a.
Kota Kinabalu
4.537 52,047 Length: 2987
International
Kuching 3.236 37,348 n.a.
Malaysia
Kunming Wujiaba 15.729 148,185 3800*45
Langkawi 1.123 10,828 3810*45
Senai Airport 1.325 38,509 3800*46
Subang - Sultan
0.090 44,302 3780*45
Abdul Aziz Shah
Macau Macau International 5.499 53,386 3360*45
Diosdado Macapagal 3200*45
Philippines n.a. n.a.
International 3200*60
Kaoshiung 2600*45
Taiwan 5.717 67,149
International 2700*60
Phuket International 5.704 41,719 3000*45
Bangkok Don Muang 3700*60
4.805 87,064
International 3500*45
Chiang Mai
Thailand 3.291 32,758 3100*45
International
Chiang Rai
0.754 7,743 3000*45
International
Hat Yai International 1.390 16,687 3050*45
Hanoi Noi Bai Length: 3200
Vietnam 6.371 44,353
International Length: 3800
Sources: Air transport intelligence database.

2 Aviation Policy
2.1 Different approach to deregulation

Although Asia has been the fastest growing air travel market in the world during
the last two decades, air transport deregulation and liberalization in the region
has been slower than in North America and Europe. Countries in Asia are more
diverse politically, economically, and culturally than those in North America and
Europe, with each of them differing in their approach to deregulation. Some
countries, with strong economies and successful, well-established national
airlines, developed countries with small populations, and countries that are
60 DEVELOPMENT OF REGIONAL AIRPORTS

isolated, have been strong advocates of liberalization, while others have been
fearful of the consequences.
Singapore, the country with the smallest populations in this region, is keen to
pursue more liberalized Air Services Agreements. Some countries have allowed
a second national carrier to fly international routes in order to meet the rapidly
rising demand and provide a better service. This has been the case in several
countries, such as Hong Kong, Indonesia, Korea, Philippines, Taiwan, and
Thailand. As the pressure for deregulation continues, China Civil Aviation
Authority has restructured its domestic market. It merges its carriers into three
groups, Air China, China Eastern, and China Southern, in order to strengthen its
national airlines.
In Japan, the national government announced its Asia Gateway Plan,
intended to liberalize Japanese skies, cut ticket prices and stimulate trade. Japan
said Asia Gateway is the most radical liberalization of the highly regulated
Japanese skies in decades. Tokyo wanted to open access for foreign carriers at
regional airports, expand operations at Haneda airport and reduce limits on
carriers flying from it. The country’s aviation officials hope the policy will spur
an Open Skies agreement with the Association of South East Asian Nations
(ASEAN).

2.2 Aviation deregulation in Southeast Asia

The aviation deregulations in Southeast Asian countries are still progressing.

Similar to EU, the trade circle within Southeast Asia is becoming smaller. To
promote economic cooperation among Southeast Asian countries, ASEAN was
established in 1976. The founding members were Malaysia, Thailand, Singapore,
Indonesia, and the Philippines; Brunei (1984), Vietnam (1995), Burma (1997),
Laos (1997), and Cambodia (1999) followed. Figure 4 is the map of ASEAN
countries. The ASEAN region has population of around 500 million, total area of
4.5 million square kilometres, combined gross domestic product of almost
US$700 billion, and total trade of about US$850 billion (About ASEAN [5]).
The European Union (EU) established the first deregulation open skies policy in
the world. The same question of open skies in Southeast Asia has been a topic of
discussion at past ASEAN summits.
According to ASEAN’s air transport deregulation policy, the ASEAN
summit plans to open skies from capital city to capital city in 2008. In 2010, the
member countries will open skies of a ‘capital city + 1’, which means, besides
capital cities, each member country will open an airline in one additional city.
From that point on up to 2015, the member countries will achieve 100%
open skies in the ASEAN region. It proposes the formation of the ASEAN
Airline Association to organize privately-owned airlines and provide greater
accessibility. One of the largest disadvantages is the degree of deregulation in
ASEAN countries that are not unanimous, nevertheless (Patel [18]). Singapore,
Thailand, and Brunei joined the open skies pact in December 2004, and
Indonesia government has already declared open skies to ASEAN in October
2006.
 DEVELOPMENT OF REGIONAL AIRPORTS IN ASIA 61

Figure 4: Map of Southeast Asia.

Once the air transport is deregulated, the low-cost carriers (LCCs) can compete
to occupy the market without any restrictions. Furthermore, passengers are often
faithful to the airline company that emerges first. In an attempt to take the
leadership role in the air transport market, some of the main LCCs, such as
AirAsia [1] and Jetstar [12], have implored to their governments to open skies to
all ASEAN members. Hopefully, the ASEAN era of open skies will soon come.

2.3 Capital city airports reform

After some new hub airports operated in Asia, the old city airport was transferred
into regional function, including Japan Tokyo Haneda Airport, Korean Seoul
Gimpo Airport, Taiwan Taipei Shengshan Airport, Thailand Bangkok Don
Muang Airport, and China Shanghai Hongqiao Airport.

2.3.1 Japan Tokyo Haneda Airport

Tokyo Haneda Airport is easily Asia’s busiest airport, but it mainly handles
domestic flights, with only limited international charters during overnight hours
plus special shuttle flights during the day to Seoul’s primarily domestic Gimpo
airport and Shanghai’s primarily domestic Hongqiao airport. The overnight
charters began in 2001, while Gimpo flights began in 2003, followed by
Hongqiao flights in 2007. Most of the international flights serving Tokyo operate
to Narita airport, which is much further away from the Japanese capital’s main
62 DEVELOPMENT OF REGIONAL AIRPORTS

business districts. Business travellers in particular prefer Haneda and airlines

have been looking forward to the opening of the long-awaited fourth runway in
2010. Congested Narita will in 2010 also have many new slots after an extension
is completed of its secondary runway. Narita has an overnight curfew while 24-
hour operations are allowed at Haneda (Airline Business [2]).
Until 2007, international charter flights were only allowed at Haneda between
23:00 and 06:00 but departures are now allowed from 20:30 and arrivals until
08:30. Ahead of the wider opening of Haneda to international flights, All Nippon
Airways and Japan Airlines have been adding charter flights to key business
destinations such as Hong Kong. Cathay Pacific Airways has also launched
charter flights between Hong Kong and Haneda and says it hopes to eventually
serve Haneda with scheduled flights.

2.3.2 Korea Seoul Gimpo Airport

Gimpo International Airport, commonly known as Gimpo Airport (formerly
Kimpo International Airport) was the main international airport for Seoul and
South Korea before it was replaced by Incheon International Airport in 2001. It
is now the second largest airport in Korea.
Gimpo airport is located in the west of Seoul, Republic of Korea. It became
primarily a domestic airport but it also handles limited international traffic from
Tokyo’s primarily domestic Haneda airport and Shanghai’s primarily domestic
Hongqiao airport. The services are popular with business travellers as Gimpo is
closer to the centre of Seoul than Inchon. Nowadays Gimpo is giving new
pleasures to airport travellers of enjoying the complex space, called ‘Sky City’
which is, a place of culture, leisure, and shopping.

2.3.3 Taiwan Taipei Songshan Airport

Taipei Songshan Airport, nucleus of Taiwan civil aviation, is located in the
Songshan District of Taipei. It is easily accessible from the town centers. Total
space of this airport is 182 hectare (civil aviation area: 83 hectares). The east-
west runway can accommodate B-747 and smaller aircrafts. Several auxiliary
facilities including navigation guide and aid equipment, and approach lights are
installed, to facilitate low visibility and night landing. Eight air bridges are
available for passenger boarding and deplaning. This airport operates on 24-hour
basis with the exception of the 2300–0600 hour during which only limited
control flights and engine test are allowed.
Situated in Metropolitan Taipei, the Taipei Songshan Airport is the hub of
domestic air transport. As a result of the economic slowdown, the aviation
market has become depressed. Consequently, the number of passengers has
dropped by close to 10% in recent years. In the future, this Airport will offer
flights to Japan, Korea, China, and other regional destination.

2.3.4 Thailand Bangkok Don Muang Airport

Don Muang Airport used to be Bangkok’s main international airport. The
situation changed when Bangkok’s Suvarnabhumi airport opened in September
2006, the new airport became the international gateway and Don Muang was left
 DEVELOPMENT OF REGIONAL AIRPORTS IN ASIA 63

for point-to-point domestic services. In December 2007, Thailand’s airport

authority decided to re-open Don Muang airport to scheduled international
flights in an effort to relieve congestion at Suvarnabhumi airport and also to
make Don Muang more viable financially. The biggest domestic operators at
Don Muang airport are Thai Airways International and LCCs Nok Air [15] and
One-Two-Go [16].

2.3.5 China Shanghai Hongqiao Airport

Although Shanghai is not the capital city, it is a charming city full of energy, and
is an important gateway in China and a bridge connecting the outer world. To
satisfy the needs of the operation of ‘One City, Two Airports’, Shanghai Airport
Authority was established officially in February, 1998 upon the approval of
Shanghai Municipal People’s Government as a result of an important reform
on the management of Shanghai airports. From then on, both Hongqiao
International Airport and Pudong International Airport are under the uniform
management and operation of Shanghai Airport Authority. Hongqiao
International Airport is located in the western suburbs of Shanghai, only 13
kilometres away from the city centre. Currently, Hongqiao Airport mainly
accommodates domestic flights and remains the backup function for
international flights and regional flights.
Besides developing international flights, a Shanghai Hawker Pacific Business
Aviation Centre has been established at Shanghai Hongqiao airport. This
Aviation Centre is 49% owned by Australia’s Hawker Pacific and 51% owned by
the Shanghai Airport Authority. It will have a maintenance, repair, and overhaul
operation due to be completed in August 2009. The MRO will handle heavy
maintenance for aircraft including Dassault Falcons and Hawker Beechcraft
types. The company is seeking to expand its offering to include Bombardier,
Embraer and Gulfstream business jets, the hangar will accommodate a Boeing
737–800, a Gulfstream and another smaller aircraft simultaneously. The new
business aviation centre at Shanghai Hongqiao will be significant, because the
airport authority plans to transfer business jet operations from Shanghai Pudong
to the new centre at Shanghai Hongqiao. Due to its proximity to the city centre
Hongqiao should prove to be a popular choice among business travellers.

3 Air transport development between China and Taiwan

3.1 Traffic between Taiwan and China

Over the past two decades, demand for air transport between Taiwan and China
has grown rapidly, partly in response to the economic development of the two
countries, and also as a result of political factors. Passengers who travel between
the two places always transit via Hong Kong or Macau. Although Taipei and
Fuzhou are only 184 kilometres apart, air travel between these two cities is only
possible via Hong Kong or Macau. The journey takes 4 hours and 50 minutes,
and costs US$374, excluding airport taxes. The high travel cost owing to transit,
64 DEVELOPMENT OF REGIONAL AIRPORTS

along with the inconvenience of transferring flights and the additional flying
time, has become a major annoyance to most travellers.
As an alternative to frequent flying customers, the Taiwan-Hong Kong route
is one of the busiest routes in the world. The fly time between Taiwan to Hong
Kong is only 110 minutes. In addition, there are about forty-one-way flights
operated by five airlines for this route per day. Direct cross-strait flights,
currently in a high-demand market, have been approved in July 2008. The
distance between Taiwan and China is similar to that between the UK and the
European mainland. The successful experiences of LCCs in Europe could be
replicated in Taiwan and China (Figure 5) in the near future. The operational
strategies of LCCs in Europe could present a model for Taiwanese and Chinese
carriers.

Hailar
Qiqihar Jiamusi

Hirbin
Mudanjiang
Changchun
Urumqi
Shenyang
Kashi
Hohhot Beijing
Tianjin Dalian
Taiyuan

Jinan Qingdao
Lanzhou
Xian

HefeiNanjing Shanghai
Pudong
Wuhan Shanghai
Chengdu
Lasa Chongqing Hangzhou Hongqiao
Zhangjiajie Nanchang
Xichang
Guiyang Changsha Fuzhou

Kunming Taipei
Guilin
Guanzhou Xiamen
Xishuangbanna Shenzhen
Nanning

Sanya

Figure 5: Routes for direct cross-strait flights.

3.2 Airports in China

Civil airports in China have been expanding rapidly since the 1990s. There were
142 airports operating in the country at the end of 2005, compared to 121 in 2000
and 109 in 1994. Twenty-five of these could facilitate aircraft of Boeing 747
size, and 113 could facilitate aircraft of Boeing 737 size (Table 4). China will
have invested RMB 140 billion in airport construction in the 11th Five-Year Plan
period which ends in 2010. By then it will have 186 airports, including three
large-composite hubs, seven large regional hubs, 52 medium-size airports and
124 regional airports.
China has undertaken a massive civil aviation infrastructure effort to build
new airports and expanded existing facilities. China’s Xiamen Airlines is
 DEVELOPMENT OF REGIONAL AIRPORTS IN ASIA 65

establishing a hub in Hunan province’s Changsha city in co-operation with the

Hunan provincial government. Within the next 12 months from April 2008 the
airline will have three aircraft stationed at Changsha’s Huanghua airport.
In May 2008, UPS announced the planned relocation of its Intra-Asia hub to
Shenzhen from Clark Airport in the Philippines. The development is expected to
cost US$180 million and become operational in 2010. The new hub is expected
to cover an area of around 89,000 square metres and will include an express
customs handling unit, sorting facilities (which will be 5 times the size of the
Clark facility), cargo handling and ramp handling operations The relocation is
expected to reduce shipment times for Asian customers by 1 day or more.

Table 4: Classification of Mainland China civil airports.

Flight
Aircraft Number
area Airport name
model of years
grade
Beijing Capital, Shenyang Taoxian, Dalian
Zhoushuizi, Harbin Taiping, Shanghai
Hongqiao, Shanghai Pudong, Nanjing Lukou,
Hanzhou Xiaoshan, Fuzhou Changle, Jinan
Yaoqiao, Xiamen Gaoqi, Zhengzhou
4E B747 25 Zhengxin, Wuhan Tianhe, Guangzhou
Baiyun, Shenzhen Baoan, Zhuhai Sanzao,
Haikou Melan, Sanya Fenghuang, Chongqin
Jiangbei, Chengdu Shuangliu, Kunming
Wujiaba, Lhasa Gongga, Xi’an Xianyang,
Urumchi Diguobao, Keshen
Tianjin Binhai, Shijiazhuang Zhengding,
Taiyuan Wushu, Huhehot Baita, Changchun
Dafangshen, Changzhou Benniu,
Lianyungang Baitafu, Xuzhou Guanyin,
Ninnbo Lishe, Wenzhou Yongqiang,
Chuanzhou Jinjiang, Zhoushan Zhujiajian,
Hefei Luogang, Nanchang Changbei, Yantai
B767
Laishan, Qiangdao Liuting, Weihai
4D B757 35
Dashuibo, Weifang, Luoyang Beijiao,
MD82
Yichang Sanxia, Changsha Huanghua,
Zhangjiajie Hehua, Shantou Waisha, Naning
Wuxu, Geermu, Liuzhou Bailian, Guilin
Liangjiang, Xichang Qingshan, Kweyang
Longdongbao, Changdu Bangda, Lanzhou
Zhongchuan, Xining Caojiabao, Yinchuan
Hedong, Hetian, Mianyang Nanjiao
66 DEVELOPMENT OF REGIONAL AIRPORTS

Table 4: Classification of Mainland China civil airports. (continued)

Flight
Aircraft Number
area Airport name
model of years
grade
Qinwangdao Shanhaiguan, Hailaer Dongshan,
Baotou, Dandong Langtou, Jinzhou, Yanji
Zhaoyangchuan, Qiqihar Sanjiazi, Jiamusi
Dongjiao, Mudanjiang Hailang, Heihe,
Nantong Xingdong, Yancheng, Huangyan
Luqiao, Yidao, Quzhou, Huangshan Tunxi,
Anqing, Fuyang, Wuyishan, Ganzhou
Huangjin, Jingdezhen, Linyi, Dongying,
Jiujiang Lushan, Nanyang Jiangying, Changde
4C B737 53 Taohuayuan, Zhanjiang, Yongzhou Lingling,
Beihai Fucheng, Yilan Caibai, Jinyang,
Guangyuan Panlong, Jiuzhai Huanglong,
Panzhihua, Luzhou Lantian, Wanzhou
Wuqiao, Zhaotong, Lijiang, Dali, Diqing,
Simao, Baoshan, Tongren, Yanan, Dunhuang,
Linchang, Wuzhou Changzhoudao, Wuxi
Shuofang, Liancheng Guangmaoshan,
Jinggangshan, Xiangfan Liuji, Liping, Anshun
Huangguoshu
Xilin Haote, Tongliao, Wulan Haote, Wuhai,
Chifeng, Manzhou Lixijiao, Changzhi
Wangcun, Liancheng Zhangxiao, Datong
Beijiazao, Chaoyang, Meixian, Enshi
Below
3C 29 Xujiaping, Huaihua Zhijiang, Nanchong
B737
Gaoping, Xishuangbanna Gasa, Dehong
Mangshi, Yulin Xisha, Hanzhong, Jiayuguan,
Qingyang, Ankang, Yining, Aleitai, Akesu,
Tacheng, Kuche, Kangerlei, Qiemo, Xingyi

Source: China Civil Aviation Authority, 2006. Analyses of Air Transport from
Statistics, 2005, Beijing, China.

Zhuhai Airport, once China’s largest white elephant airport, has made
remarkable progress since Airport Authority Hong Kong entered into a joint
venture management deal with the Zhuhai Municipal People’s Government.
Meanwhile, another Mainland airport, Hangzhou Xiaoshan International Airport,
which is 35% owned by Airport Authority Hong Kong, plans to raise US$800
million from an initial public offering (IPO) by the end of the third quarter
of this year, 2009. The Airport Authority’s share will be reduced to 25% after
the IPO.
 DEVELOPMENT OF REGIONAL AIRPORTS IN ASIA 67

3.3 Direct flights between Taiwan and China

On 4 July 2008, regular non-stop passenger flights between China and Taiwan
have begun, nearly 60 years after the two sides split following a civil war. Six
Chinese airlines and five Taiwanese airlines will be operating regularly between
the two sides, with flights permitted from Friday through Monday each week.
Under the agreement, airlines from each side are allowed to operate 18 round-
trip flights between Fridays and Mondays, between five Chinese cities and eight
Taiwanese ports. In November 2008, an expanded deal was agreed to allow for a
combined total of 108 weekly flights, which can be operated any day of the week.
The expanded agreement also allows for a total of 60 cargo flights per week and
increases to 21 from five the number of Chinese cities that can be served. Flights
are allowed to seven Taiwanese airports, namely Taoyuan International Airport,
Taichung International Airport, Taipei Shangshon International Airport, Kaoshiung
International Airport, Mukun Airport, Hualien Airport and Taidong Airport. In
China, 21 airports are allowed to fly to Taiwan, including Shanghai Pudong,
Beijing Capital, Guangzhou Baiyun, Xiamen Gaoqi, Nanjing Lukou, Chengdu
Shuangliu, Chongqin Jiangbei, Hanzhou Xiaoshan, Dalian Zhoushuizi, Guilin
Liangjiang, Shenzhen Baoan, Wuhan Tianhe, Fuzhou Changle, Qiangdao Liuting,
Changsha Huanghua, Haikou Melan, Kunming Wujiaba, Xi’an Xianyang,
Shenyang Taoxian, Tianjin Binhai, and Zhengzhou Zhengxin. It is expected that
the traffic will grow strongly in these airports in future in large part because of the
new tourist demand.

3.4 Airport policy in China

Since all airports were transferred from the General Administration of Civil
Aviation of China (CAAC) to provincial governments in mid-2004, with the
exception of Beijing Capital International Airport and Lhasa Airport, China’s
civil airports have entered a new era. After the airport decentralization, each
provincial government established a 100% state-owned airport management
group company responsible for all civil airports inside the province. However,
most of China’s airports are losing money because there are not sufficient
aircraft movements and local governments would find it very difficult to justify
supporting the loss-making airports. So, the provinces look to ‘big brothers’ to
help and advise them in the running of heir airports. The idea is that in the long-
term the new system will be experienced enough to sustain itself.
Capital Airports Holding Company (CAH), Xiamen International Airport
(Group) Co. Ltd, and Shanxi Airport Management Group are major airport
groups. CAH, established in December 2002, now has a portfolio of 25 airports,
which belong to airport management groups in Hubei, Guizhou, Jiangxi, Jilin,
and Liaoning provinces and Beijing, Tianjin, and Chongqing municipalities,
which are under the direct control of the central government. The market share of
CAH’s member airports was 30% in terms of passengers handled in 2005.
Through asset restructuring, CAH now holds 65% of Beijing Capital
International Airport Co. Ltd, which last year was the fastest growing of the
68 DEVELOPMENT OF REGIONAL AIRPORTS

world’s top 30 hotels, according to Airports Council International (ACI). It

handled 40.1 million passengers, up 17.5% year-on-year. Similarly, CAH also
holds 100% in Tianjin and Chongqing airports and of Guizhou, Jiangxi, and Jilin
Airport Management Groups. Through acquisition, CAH holds 100% of shares
in Hubei and Liaoning Airport Management Groups (Orient Aviation [17]).
In addition, Xiamen International Airport Group Co. Ltd. manages three
airports; Xiamen Gaoqi Airport, Fuzhou Changle Airport, and Longyan Airport,
with total assets of about RMB 5 billion. The total passenger throughput in 2005
reached 10 million. Shanxi Airport Management Group consolidated with
Ningxia and Qinghai groups in October 2004 and March 2006 respectively. They
have a total of eight airports with Xian Xianyang Airport as the hub airport. It is
anticipated that this move will accelerate airport growth as part of the West
China development strategy. For many years, there have been three channels of
airport financing in China. They are: the airport construction fees fund, started in
1992 and collected from passengers, financing by local government, and Chinese
bank loans. Since the transfer to local governments, there have been more
opportunities for airport financing such as commercial loans, domestic bonds,
private sector investment, domestic cross-industry investments through airport
corporatization and IPOs, foreign loans, international bonds, and investment or
lease by foreign operators (Orient Aviation [17]).
In 2002, the CAAC, the former Ministry of Foreign Trade and Economic
Cooperation and the State Development Planning Commission jointly issued
‘Rules on Foreign Investment in the Civil Aviation Industry’. Foreign investors
and airport operators expressed interest. For example, Copenhagen Airport
acquired 20% of the Haikuo Meilan International Airport Company in 2002 and
signed a 10-year consulting services contract which covered areas of capacity
utilization and expansion, airport operations and the development of commercial
activities. Haikuo Meilan is a listed company. In 2005, the airport handled seven
million passengers and is ranked tenth in China. Copenhagen received a dividend
of two million Danish krones (US$342,500) in 2004. Meilan Airport,
meanwhile, earned 151.4 million yuan in 2005. In 2001, Aeroports de Paris
acquired a 9.9% share in the Beijing Capital International Airport Co. Ltd. Last
year Beijing Airport recorded a net profit of 915.5 million yuan, an increase of
20% over 2004. In June 2005, German airport operator, Fraport A.G., agreed to
take a 25% stake in Ningbo International airport. In the same month, Singapore
Changi International Airport signed an initial agreement to spend up to 1.6
billion yuan for as much as 45% of Nanjing airport, from which Singapore
Airlines [19] has fifth freedom rights to operate cargo services to US
destinations. BAA, the world’s leading airport management company, has
expressed interest in investing in secondary airports, which have international
services and process a minimum of five million passengers annually (Orient
Aviation [17]).
 DEVELOPMENT OF REGIONAL AIRPORTS IN ASIA 69

4 Low-cost carriers and regional airports

Southeast Asia’s enormous populations, scattered land form, and growing
economies have pushed LCCs into ascendancy. In most of Southeast Asia LCCs,
shortly after initiating their businesses, quickly expanded to occupy a large
percentage in the aviation market.

4.1 The rising of low-cost airlines

Following the experiences of LCCs in the US and the EU, the wave of LCCs that
has emerged in Southeast Asia has raised expectations that the experience of
LCCs in other major markets will now be replicated in this dynamic region. The
main reason is that national airlines have often not been able to provide enough
capacity, and they have often been criticized for their poor service standards in
domestic markets (Chang and Lee [8]).
The aviation industry in Southeast Asia is always very competitive. Almost
all of these Southeast Asian LCCs have been in business for only a few years,
but they have expanded rapidly and now, particularly with the deregulation of
the industry, own a large percentage of the aviation market in the region. Direct
competition between full-service carriers (FSCs) and LCCs is intensifying across
the area. Many evolving LCCs in Southeast Asia use several strategies which
decrease its operation cost, promote efficiency, and attract passengers with low
ticket price. This kind of airline business style immediately became popular with
short-haul passengers.

4.2 Environmental Characteristics of Southeast Asia

Table 5 shows basic information of ASEAN countries. Enormous population is

the most important characteristic of each of these ASEAN countries, and the
population continues to grow. Another important standard used to describe these
countries is the Gross Domestic Product (GDP). The GDP per capita in
Southeast Asia is much lower as compared to other countries in the world. From
2005 to 2006, the GDP per capita in every country in ASEAN experienced
positive growth. Table 6 contains information that describes communication,
airports, and current airline businesses. Both the internet and cellular phone
usage rates are low in these ASEAN countries, except Singapore. Cambodia,
Burma, and Laos have a very low internet usage rate; it is even less than 1%.
The number of privately-operated airlines implicates the degree to which the
aviation industry has the freedom to operate. In Malaysia and Cambodia, there is
no state-operated airline. The governments of these two countries are the owners
of their privately-operated airlines. Brunei, Burma, and Laos have no privately-
operated airline. Therefore, the aviation industry in these three countries is less
regulated. Airports are also important to aviation industry. Runway length over
1,524 m is acceptable for Airbus 320, one of the most widely used aircrafts of
LCCs. The more airports with runways longer than 1,524 m a country has, the
propitious condition to develop LCCs it becomes.
70 DEVELOPMENT OF REGIONAL AIRPORTS

Table 5: Basic information of ASEAN countries.

GDP per Real
Population capita(US$)1 GDP
Total area Capital
Population1 growth growth
(km2) city
rate (%)2 rate
(%)2
Kuala 1,700 5.3
Malaysia 329,750 24,385,858 1.78
Lumpur
Thailand 514,000 Bangkok 64,631,595 0.68 8,300 4.5
Singapore 692.7 Singapore 4,492,150 1.42 28,100 6.4
Indonesia 1,919,440 Jakarta 245,452,739 1.41 3,600 5.6
Vietnam 329,560 Hanoi 84,402,966 1.02 2,800 8.4
Philippines 300,000 Manila 89,468,677 1.80 5,100 5.1
Bandar 23,600 1.7
Brunei 5,770 Seri 379,444 1.87
Begawan
Cambodia 2,200 6
Cambodia 181,040 Phnom 13,881,427 1.78
Penh
Burma 678,500 Yangon 47,382,633 0.81 1,700 2.9
Laos 236,800 Vientiane 6,368,481 2.39 1,900 7.2

Note: 1in July, 2006; 2from 2005 to 2006. Source: CIA: The World Factbook [9].

Southeast Asia seems less developed than other areas in the world; however, it
seems to have more tourism attractions. Many ASEAN countries continue their
focus on the tourism industry. Table 7 presents the total tourism and visitor
arrivals to ASEAN countries in 2005. Excluding Thailand and Indonesia, the
total international visitor arrivals have increased in the remaining countries,
with most of the visitors coming from other ASEAN countries. As Figure 6
demonstrates, 45% of the visitors are from ASEAN countries and an additional
27% from other Asian countries. This indicates that the main tourism market for
Southeast Asia is from the Asia region. This implies that there is a good demand
for regional airports for these visitors.
Table 8 provides basic information about some Southeast Asian LCCs. Most
of them are subsidiaries of FSCs. Because of this close relationship, the
Southeast Asian LCCs might not be operating as a true LCCs because there is
bound to be legacy influence in the operation. The parent companies do not want
to turn their subsidiaries into competitors, so the subsidiary LCCs and parents
FSCs often do not share the same routes. For example, some of the subsidiaries
only operate the domestic routes, or have only few international routes which
connect to the major cities. Passengers sometimes are forced to use FSCs and
pay more for the tickets. Non-subsidiary LCCs, such as AirAsia and LionAir
[13], however, are not only willing to compete with FSCs, but to surpass them.
 DEVELOPMENT OF REGIONAL AIRPORTS IN ASIA 71

Table 6: Communication, airports, and aviation business in ASEAN countries.

Communication Airports Aviation business

Cellular
Number of
Internet phone
Number runways State- Privately-
usage usage
of airports length operated operated
rate % rate%
>1524 m
(about)
Malaysia 41.17 59.92 37 22 0 4
Thailand 13.03 42.36 66 41 1 15
Singapore 53.91 94.76 9 7 1 5
Indonesia 7.33 12.22 159 69 3 32
Vietnam 6.95 5.88 26 24 1 2
Philippines 8.74 36.81 83 37 1 14
Brunei 14.76 36.11 1 1 1 0
Cambodia 0.30 3.59 6 4 0 8
Burma 0.13 0.20 21 19 1 4
Laos 0.33 8.17 9 6 1 0
Source: CIA: The World Factbook, 2006; wikipedia airline list, http://en.wikipedia.
org/wiki/Main_Page

Table 7: Tourism in southeastern countries.

Percentage total
ASEAN Percentage Risk of
Total visitor visitor arrivals
visitor visitors from infectious
arrivals growth
arrivals ASEAN diseases
2004~2005
Malaysia 16,431,055 4.43 12,984,646 79.03% √
Thailand 11,516,936 –1.91 3,099,569 26.91% √
Singapore 8,942,408 6.34 3,341,392 37.37%
Indonesia 5,002,101 –5.84 2,038,292 40.75% √
Vietnam 3,467,757 15.57 469,536 13.54% √
Philippines 2,623,084 12.65 179,386 6.84% √
Brunei 127,142 6.51 76,156 59.90%
Cambodia 1,421,615 25.77 219,579 15.45%
Burma 660,206 0.50 51,705 7.83%
Laos 1,095,315 18.31 794,044 72.49%
Total 51,287,619 4.35 23,254,305 45.34%
Source: CIA: The World Factbook, 2006.
72 DEVELOPMENT OF REGIONAL AIRPORTS

Others & Unspecified Africa

5% 0%
Australasia/ Pacific
5%

Europe
12%

ASEAN America
44% 6%

Middle East
1%

Asia (Excl: ASEAN)

27%

Figure 6: Visitor arrival % from each area.

Source: ASEAN official website [6].

Table 8: Basic information of some Southeast Asian LCCs.

LCCs company Started year Country Base Destinations
Kuala Lumpur
AirAsia Dec. 2001 Malaysia 46
International Airport
Cebu Pacific Manila Ninoy
Mar. 1996 Philippines 40
[7] Aquino International
Singapore Changi
Jetstar May. 2004 Singapore –
Airport
Tiger Airways Singapore Changi
Dec. 2003 Singapore 15
[20] Airport
LionAir Jan. 2000 Indonesia Colombo Airport –
Bangkok Don
Nok Air Jul. 2004 Thailand 6
Muang Airport
Bangkok Don
One-two-go Dec. 2003 Thailand –
Muang Airport

4.3 Interaction of low-cost carriers and regional airports

Low-cost carriers in the Western countries often place destinations in smaller

airports only. Most of them also need no air-bridges, have large negotiation
 DEVELOPMENT OF REGIONAL AIRPORTS IN ASIA 73

space towards some regional small airports. Southeast Asian LCCs cannot use
only regional small airports because there are only few airports that can be used
in these ASEAN countries. The LCCs set destinations at the same airports with
other FSCs. But there is still another alternative. The two main international
airports, Changi Airport Singapore and Kuala Lumpur Airport Malaysia, always
compete with each other. They both had their low-cost terminals launched in
March 2006. These low-charged terminals are opened to attract LCCs which
bring millions of tourists.
AirAsia is based in Malaysia, is by far the largest LCCs in Southeast Asia,
and it is also the first LCCs in all Asia. It has 46 destination flied from Malaysia
(Figure 7). Currently it planned a new LCCs airport for the capital Kuala
Lumpur, it will be dedicated for use and managed by AirAsia, equipped with its
own runway and air traffic control facilities. It is due to open in 2011. AirAsia
transformed from a negative-revenue FSC and has not only turned from loss to
profit, but also funded two subsidiaries, Thai AirAsia and Indonesia AirAsia. Its
success suggests that the low-cost model could be a path to the victory in the
Southeast Asia aviation competition.

Figure 7: Routes of AirAsia from Kuala Lumpur.

Source: AirAsia website [1].
74 DEVELOPMENT OF REGIONAL AIRPORTS

Macau is one of the places, where AirAsia is considering establishing a joint-

venture. The airline group currently operates to seven destinations in northeast
Asia but Macau is by far its largest destination in terms of number of flights.
AirAsia operates to Macau from four points in Malaysia and Thai AirAsia
operates to Macau from Bangkok. AirAsia is looking at establishing an airline in
the Philippines and has planned to establish an LCCs in Vietnam in joint-venture
partnership with state-owned Vietnam Shipbuilding Industry Corp. But Vietnam
already has three airlines. The government there has granted preliminary
approval for two more and has said that in the near term it will cap the number of
airlines operating in the country to five.
Towards the end of this year it is launching services to Perth in Australia
after it receives a second A330. Also this year the airline plans to launch services
to a second destination in China and a destination in India, a destination in South
Korea and a destination in Taiwan.

5 Conclusions
Countries in Asia are more diverse politically, economically, and culturally than
those in North America and Europe, with each of them differing in their
approach to deregulation. Some countries, with strong economies and successful,
well-established, national airlines, developed countries with small populations,
and countries that are isolated, have been strong advocates of liberalization,
while others have been fearful of the consequences. The economic characteristics
in Asia included developed countries (Hong Kong, Japan, Korea, Singapore, and
Taiwan) and developing countries (China, Indonesia, Malaysia, Philippines, and
Thailand). Asian airport operators will be facing a challenging time ahead with
air traffic growth at a reduced pace and volatility in fuel prices affecting airline
business. In addition, the LCCs continuing growth in Asia, consumer need for
point-to-point travel and scarcity of secondary airports in Asia have opened up
many opportunities for airport operators and investors.
The strong economic development in China with an emerging middle and
upper class has really stimulated air traffic demand. There were 142 airports that
served civil aviation in China in 2005. On 4 July 2008, regular non-stop
passenger flights between China and Taiwan have begun, nearly 60 years after
the two sides split following a civil war. Under the agreement, airlines from each
side are allowed to a total of 60 cargo flights per week. Flights are allowed to 7
Taiwanese airports and 21 Chinese Airports. It is expected the traffic will grow
strongly in these airports in future in large part due to the new tourist demand.
In 2006, China’s Total operations (international and domestic) ranked 2 in
terms of Passenger-kilometres. This placed an enormous strain on the capital
requirements of the commercial aviation sector. The Chine government
concluded that its airlines and airports need capital and operational expertise;
therefore, the CAAC opened its commercial aviation market to foreign investors.
Afterwards, Copenhagen Airport has acquired 20% of the Haikuo Meilan
International Airport Company, Aeroports de Paris acquired a 9.9% share in the
Beijing Capital International Airport Co., German airport operator, Fraport A.G.,
 DEVELOPMENT OF REGIONAL AIRPORTS IN ASIA 75

took a 25% stake in Ningbo International airport, and Singapore Changi

International Airport acquired 45% of Nanjing airport. BAA, the world’s leading
airport management company, also has expressed interest in investing airports in
China. The equity investment with Chinese airports will help these European
airports and Asian airport enter into China market.
There are lots of airports in Asia, it is not easy to collect the information of
regional airports. Among Asian nations, Indonesia is first in its dependence on
air transport. In these regional airports, some are popular tourist island airports,
like Haikou of China, Bali of Indonesia, Jeju of Korea, Okinawa of Japan,
Langkawi of Malaysia, and Phuket of Thailand. Because of their weather and
beach, these island airports attract lots of passengers each year. In addition, in
some countries, after some new hub airports operated, the old capital airport was
transferred into regional function, including Japan’s Tokyo Haneda Airport,
Korea’s Seoul Gimpo Airport, Taiwan’s Taipei Shengshan Airport, Thailand’s
Bangkok Don Muang Airport, and China’s Shanghai Hongqiao Airport.
The landforms of ASEAN are almost islands. Some archipelagic states, like
Malaysia, the methods to travel from one island to another are only by through
sea and air. In general, travel by air can save more time. Over the past decade,
many new LCCs have arisen in Southeast Asia. Currently, Southeast Asia’s
enormous populations, scattered landforms, and their growing economies of the
countries they service have pushed LCCs into ascendancy. Since the slogan of
most LCCs is ‘make all the people fly’, travel by LCCs may be more efficient.
AirAsia is by far the largest LCCs in Southeast Asia, it is based in Malaysia and
has joint venture airlines in Indonesia and Thailand. It is looking at establishing
an airline in the Philippines and planned to establish an LCCs in Vietnam.
While the 21st century has begun with an economic downturn both in the air
transport market and in general, the cataclysmic events of 11 September, the Iraq
war and SARS in some Asian countries have made this situation worse. The
aviation industry is facing the most difficult economic situation it has ever
experienced; even worse than the Gulf War in the 1990s. Although Asia-Pacific
has been the fastest growing air travel market in the world during the last two
decades, air transport deregulation and liberalization in the region has been
slower than in North America and Europe. In the meantime, the ASEAN summit
plans to open skies from capital city to capital city in 2008. In 2010, the member
countries will open skies of a ‘capital city +1’, which means, besides capital
cities, each member country will open an airline in one additional city. From that
point on up to 2015, the member countries will achieve 100% open skies in the
ASEAN region. As a result, other non-ASEAN countries in Asia should try to
cooperate with the ASEAN members, to develop the Asian market into a Single
Market, just like the EU did two decades ago. Once the air transport is
deregulated, the airlines can compete to occupy the market without any
restrictions. The regional airports will develop dramatically as a result of the
liberalized market.
76 DEVELOPMENT OF REGIONAL AIRPORTS

References
[1] Airasia, http://www.airasia.com.
[2] Airline Business, Japan expands Haneda internationalization, July, 2008.
[3] Air Transport Intelligence Database, www.rati.com.
[4] Air Transport World, World Airport Report, September, pp. 30–63, 2008.
[5] ASEAN Statistics, Tourist arrivals in ASEAN by selected partner country/
region, http://www.aseansec.org/Stat/Table29.pdf (accessed Dec. 29,
2006).
[6] ASEAN Transport Action Plan 2005–2010, ASEAN official website,
http://www.aseansec.org/16596.htm.
[7] Cebu Pacific, http://www.cebupacificair.com.
[8] Chang, Y.-C. and Lee, N., Are low-cost carriers a bargain? The
comparison of low-cost and full-service carriers in southeast Asia. Journal
of the Transportation Research Board, 2052, pp. 21–27, 2008.
[9] CIA: The World Factbook, https://www.cia.gov/cia/publications/factbook/
index.html.
[10] Dempsey, S.P., Airport Planning and Development Handbook: A Global
Survey, New York: McGraw-Hill, 1999.
[11] International Civil Aviation Organisation (ICAO), Outlook for air
transport to the year 2025, Montreal, 2007.
[12] Jetstar, http://www.jetmail.com.au.
[13] LionAir, http://www.lionair.co.id.
[14] Malaysia Airlines, http://www.malaysiaairlines.com.
[15] Nok Air, http://www.nokair.com/NokConnext/aspx/Welcome.aspx.
[16] One-two-go, http://www.fly12go.com.
[17] Orient Aviation, China: New era, new airports, June, 2006.
[18] Patel, M., Low cost airlines in Asia Pacific: Will it succeed or fail in this
region? Travel Consulting Group, 2003.
[19] Singapore Airlines, http://www.singaporeair.com.
[20] Tiger Airways, http://www.tigerairways.com.
4

Air demand modelling: overview and

application to a developing regional airport
M. Nadia Postorino
University ‘Mediterranea’ of Reggio Calabria, Italy

Abstract
Air demand forecast at airports is an important problem for the airport
management and also for the regulator that has to plan a homogeneous
development of the overall transport system. The current tendency is towards
airport privatization; then, the goal to increase the served demand is one of the
most important together with the progress of non-aviation activities. The
evolution of the air transport system both in terms of low-cost companies, that
generally use regional airports, and new technologies (as regional jets) has given
a further impulse to the development of planning methodologies able to support
decisions for an efficient distribution of resources. Regional airports can play an
important role in this new background if the most suitable developing strategies
are identified. This chapter wants to give a general overview about the problem
of the air demand modelling, both in terms of theoretical approaches and
practical problems. Models are classified with respect to different criteria, and
the most suitable models for each planning level are also identified. An
application to a regional airport in Southern Italy is also presented in order to test
some of the described approaches and to obtain practical indications about
applied models and developing strategies to be used.
Keywords: air demand; air demand model classification; airport catchment area;
time series models; random utility models.

1 Introduction
The estimate of transport demand has always been one of the most important
stages in the transport system planning process and one of the most stimulating
challenges for the analysts, because the dependence of demand on the overall
78 DEVELOPMENT OF REGIONAL AIRPORTS

socio-economic system (particularly, income and job activities on the territory)

makes it difficult to obtain reliable values. On the other hand, the supply
characteristics (as e.g. terminal capacity, parking size and runways in the case of
an airport system) as well as their performances and profitability are strongly
dependent on the predicted demand levels.
Significant overestimates or underestimates of future demand levels lead to
wrong developing policies that can generate respectively: (a) uneconomic use of
infrastructures and/or services; (b) quick worsening of the transport system
performances due to infrastructures and/or service deficiencies compared with
the actual demand levels.
Generally, the analysis and the simulation of an air transport system concern
three macro-topics:
● estimate of the air transport demand and its distribution among several
competitive airports;
● identification of the supply organization and its effects on the different actors
(community, passengers, airports, airlines);
● forecast of the air transport services and their induced effects on the air
demand as well as on the other actors working in the system.
The first two aspects are linked to the system simulation for a given scenario
(current scenario or future hypotheses). The third aspect depends on the
airline/airport decision policies, the profit analysis, the market conditions, and,
last but not least, the political decisions aiming at the system development
following social other than technical criteria.
The increase in the air transport demand in the last few decades, also helped
by the deregulation policy, has had a major effect of increasing transport services
offered by different air carriers and has resulted in increasing congestion levels
both in the airways and at airports (Graham and Guyer [1]). As an immediate
effect of deregulation, the service offered to users, in terms of trip organization
and costs, has changed rapidly and various alliances and mergers have occurred,
together with the emergence of new air carriers in the market. For users,
deregulation has produced greater benefits due to airfare decrease and the
opportunity to choose among more flights supplied by more air carriers (Cohas
et al. [2]). Thanks to deregulation, various air carriers have the opportunity to
offer their services along high-demand routes, new connections have arisen and
fare reductions have been applied (ATAG [3]). Hence, there has been an increase
in the demand level, especially for non-systematic reasons (e.g. in Europe a
reduction of 15% in airfares has produced an increase of about 10% in the
carried passengers, Italian Ministry of Transport [4]).
Demand variations depending on local aspects (such as the building of a new
airport or the expansion of an existing one) represent another important factor in
modelling air transport demand, as the use of hub-and-spoke systems means that
each pole can be potentially connected to almost any other.
Air transport demand directly affects the planning of airport terminals in
terms of ground services design (check-in/check-out points, waiting areas,
facilities as restaurants, shops and so on). In order to properly design such areas,
 AIR DEMAND MODELLING 79

both the absolute demand and its temporal distribution are required. Furthermore,
even the competitiveness among airlines plays an important role to define the
demand distribution.
The above considerations show how important the demand characteristic
analysis and its evolution in time are in order to design more effectively the
service supplied by both airport managers and air carriers.
Reliability of the demand estimates depends on the kind of model and data
availability. Models theoretically efficient in terms of forecast often require a lot
of data referred to users and both socio-economic and supply systems for a long
period of time in order to estimate not only the current demand characteristics
but also the future levels. The knowledge of current and future levels helps
planners to develop effective short-medium and long-term actions, respectively.
Thus, data availability and model reliability are the two key elements to
obtain high-quality demand level forecast, within the effectiveness limits defined
by the stability of the boundary conditions. The latter can be identified in the
socio-economic and political stability, which has a relevant influence on the
user’s decisions to make trips and particularly to travel by aircraft.
In the following, the relationship between demand and airport catchment area
is discussed (Sections 2), and a classification of the air demand approaches with
respect to different criteria is proposed together with a description of the two
most important air demand approaches (Section 3). Finally, after an overview of
the Italian airport system (Section 4) an application to a regional airport located
in Southern Italy is presented and discussed (Section 5).

2 Demand modelling and airport catchment area

A key element for evaluating the developing potentiality of airport systems,
particularly regional airport systems, is the demand forecast for each airport
serving the considered region; such a forecast should be consistent with the
airport choices made by the air users travelling from and towards the region
itself.
As it is well-known to transport analysts, demand forecast is a relevant input
for the transport system planning; particularly, in the case of a regional airport,
system forecasts of demand have a significant influence on the future functioning
of each airport as well as on the development of the airport master plans.
Generally speaking, a demand model is a mathematical relationship linking
the expected demand level (dependent variable) to one or more explanatory
variables (independent variables), whose nature depends on the kind of model
and the availability of the corresponding data.
The choice to start the air trip from a given airport can depend on many
factors as accessibility, facilities, air services and connectivity levels (i.e. the
destinations that can be reached from the airport itself). Accessibility depends on
the land network available on the region, while the other factors depend on the
airport characteristics and the airline supply.
80 DEVELOPMENT OF REGIONAL AIRPORTS

For each airport, a catchment area is usually defined. In the literature, there
are several definitions for airport catchment area, depending on the geographical
or demand considerations.
Basically, from a geographical point of view, the airport catchment area can
be defined as the area containing all the potential users and the passengers of a
given airport (Transport Canada [5]). From a demand point of view, the airport
catchment area can be defined as the number of travellers using a given airport,
where origins of travellers can be identified in a surrounding study area whose
size depends on the characteristics of the airport itself, but that does not
necessarily represent the geographical extension of the airport catchment area.
Both points of view lead ultimately to the knowledge of demand and
geographical area.
The identification of the catchment area following either the first or the
second point of view can be made by using different approaches and different
models. Basically, the geographical point of view is better satisfied by using
indicators (mainly, accessibility indexes) while the demand point of view is
better satisfied by using behavioural models.
The geographical identification can be useful for airports not built yet, and
then to estimate the airport’s potential attractiveness just in terms of accessibility
for users living in the surrounding area, given that the airport characteristics as
well as the airline supply are not defined.

Figure 1: Airport catchment area identification: Study area and traffic zone.

The demand point of view can be useful when there are competitive airports and
then it is crucial to establish which demand is considered while choosing a
particular airport based on its characteristics. In this case, a study area containing
an examined airport, and where more airports can be located, is identified and
divided into traffic zones. The probability that users living in each zone and
travelling by air choose that airport will depend on the characteristics of the
airport itself and the competitive airports, as well as the distance (or, more
 AIR DEMAND MODELLING 81

generally, the accessibility measures) between each traffic zone and the
examined airport [6, 7, 8, 9]. Then, the demand obtained at each airport
represents the catchment area from the demand point of view, while its
geographical identification potentially corresponds to the study area (Figure 1).
Furthermore, a distinction can be made between primary and secondary airport
catchment areas (Transport Canada [5]). For a given airport, the first one refers
to air travellers choosing that airport because they are ‘captive’. The second one
refers to air travellers that may choose that airport but are not captive and then
are more elastic with respect to the choice of another airport (Figure 2).
Generally, the secondary catchment area is typical both for airports where there
are low-cost airlines and for classes of users that are more price-sensitive.

Figure 2: Primary and secondary airport catchment area.

Whatever be the approach, the knowledge of the airport catchment area is

important because it represents the possible demand for the airport and then the
potential for airport development. Demand levels and airport catchment area are
then highly dependent on each other and land accessibility plays an important
role.
As it is well-known in the economic and social sciences, accessibility is the
key to development and particularly for airports. In fact, the larger the catchment
area the larger the potential demand at the airport. An important factor for a
successful airport development, mainly for regional airports, is to increase the
catchment area other than providing good services in terms of fares, destinations
and frequencies.
The simplest method to identify the geographical extension of the catchment
area is to define a threshold value with reference to one or more accessibility
indexes: the geographical area whose accessibility index is less than the
threshold value is the airport potential catchment area; in other words, it contains
the potential users and passengers of the airport. On a first approximation, the
catchment area can be identified as the geographical area not larger than a
82 DEVELOPMENT OF REGIONAL AIRPORTS

prefixed time value (as 2 hours by car, for some important European airports
(van Reeven, de Vlieger and Karamychev [10]) or 60 minutes by any land
vehicles in the USA (Milone et al. [11])).
The knowledge of the geographical origin of passengers is a useful piece of
information for the airport management in order to identify the best developing
strategies; for example to decide if it is more suitable to invest on land
accessibility rather than on airport facilities and services. Such knowledge can be
achieved by running sample surveys on departing air travellers at the airport.
The key elements that can play an important role for the development of an
airport can be identified as:
● capability to generate demand in the airport catchment area;
● capability to generate adequate demand (from the point of view of economic
convenience) for potential point-to-point links;
● capability to adapt the airport services to the need and exigencies of the
airlines;
● involvement of airlines on airport investments to improve the offered
services.
In terms of factors affecting the size of the airport catchment area, and with
reference to the primary catchment area only, the most relevant are:
● living population;
● yearly average income and average family income;
● employment level;
● sector of employment.
Generally, if the first three factors increase, the number of air travellers increases
(and then the catchment area); while the distribution of sectors of employments
is a useful indication to identify the potential air travellers by trip purpose
(e.g. business travellers).
Furthermore, the airport catchment area also has an important impact on the
financial situation of the airport itself. For example, airports surrounded by
densely populated catchment areas and increasing population with employment
levels and income in the average or over the average and employment sectors
that generate business trips, generally, have got positive financial situations. An
important role is also played by competition among airports; airports far away
from the main national airports (250 kilometres following some studies on the
Canadian regional airports [5]), where there are low-cost airlines, and that are
also at a proper distance from the potential competing airport (e.g. 90 minutes by
car) have still a good financial situation. On the other hand, airports offering
similar services and sharing the same catchment area within a radius of about
100 kilometres, probably will have both a critical and negative financial
situation.
The airport development also depends on the location of other, potentially
competitive, airports. In the current situation, airports play a competitive role
rather than a cooperative one and then the distance among airports as well as the
 AIR DEMAND MODELLING 83

services offered are crucial in terms of user choices. Generally, users are willing
to cover longer distances to obtain better trip fares, point-to-point flights, larger
choice sets of available airlines and flights to choose the best options in terms of
departure time, destination and airline reliability (Suzuki [12]).
Furthermore, the presence of fast land modes (as fast trains) that can be
competitive in terms of fares and times with respect to the destinations served at
the airports can be an additional important factor in the airport users’ choice.
Indeed, one of the EU topics of major interest is the analysis of the fast train
network influence on the distribution of traffic volumes among airports; land fast
links between city pairs are supposed to produce a decrease in the air transport
demand among the same city pairs (Button [13]). However fast trains also
represent an easy way to arrive quickly at major airports (e.g. the Inter City
Express links between Frankfurt and Paris or the international high speed trains
linking Paris and Brussels to Great Britain trough the Channel Tunnel) and then
they also contribute to improve the airport catchment area in a whole, integrated
and inter-modal transport system. The complexity of the fast train role
(competition or integration?) with respect to airports is one of the most attractive
research fields in the transport system analysis.

3 Demand model classification

Passengers demand models can be classified with reference to the transport
system representation, the mathematical formulation, the nature of variables
(Table 1).
Table 1: Demand model classification.
Multi-mode Stage models
models (discrete choice)
Competitive
Static models
Non-competitive
Zone approach
Uni-mode Stage models
models (discrete choice)
Competitive
Time series models
Non-competitive
Competitive
Static models
Non-competitive
Airport pairs Uni-mode
approach models
Competitive
Time series models
Non-competitive
First of all, a distinction can be made between air demand models providing
forecasts for specific city-pairs, corresponding to airport-pairs serving those
cities, and demand models providing forecasts for O–D pairs, corresponding to
traffic zones pairs. Whatever be the used approach, in the first case the focal
point is the analysis of the specific relationship between airports or among one
84 DEVELOPMENT OF REGIONAL AIRPORTS

airport and all the others, while in the second case the demand model is usually
part of a more general framework where demand is estimated for traffic zones
and many transport modes; for the aircraft mode, demand is also allocated to one
or more competitive airports by using suitable models (airport choice is
described in Chapter 5).
Traditionally, demand models are classified as aggregate and disaggregate
depending on the nature of data referred to demand and explanatory variables. If
the variable ‘demand’ is referred to a single user, and so the explanatory
variables, the model is said to be disaggregate, while if the variable ‘demand’,
and then the explanatory variables, are referred to a homogeneous group of users
the model is said to be aggregate.
Furthermore, models can be called: (1) descriptive or behavioural according
to whether there are or not explicit hypotheses about trip user behaviour; (2)
multi-mode or uni-mode if they allow obtaining mode shares among several
alternative modes or demand on only one transport mode, respectively.
Particularly, multi-mode models refer to the simulation of the overall transport
system, where many transport modes are generally available (e.g. train, bus, car,
aircraft), and then the demand on many transport modes can be computed. On
the contrary, uni-mode models provide forecasts of demand for only one
transport mode and then they are suitable for the simulation of a part of the
overall transport system (e.g. the air transport system).
Multi-mode models are generally stage models, where more trip
characteristics as destination, frequency, mode and so on can be simulated by
using discrete choice models (a general overview is in [14, 15]).
Uni-mode models can be classified as static if they simulate air demand at a
given time, time series if they simulate the demand trend for a given time period,
or stage models if they simulate more trip characteristics but for a mode-specific
demand.
Time series models can still be grouped as Simple Time Series (or univariate)
and Causal Modelling (or multivariate). Simple Time Series approaches, among
the most used to obtain air demand forecasts, consider the stochastic nature of an
event does not vary in time and they simulate the demand trend without
explaining the causes. In other words, explanatory variables are not considered.
On the other hand, Causal Modelling models simulate demand in terms of cause-
effect relationships, i.e. they associate explanatory variables to the observed
demand by means of mathematical relationships linking independent variables
(causes) to dependent variables (effects). Explanatory variables are generally
referred to the examined mode, but characteristics of alternative modes (and,
particularly, competitive modes as fast train with respect to aircraft) can also be
considered by using suitable variables (mainly, level-of-service variables). From
this point of view, uni-mode models can be classified, respectively, as non-
competitive and competitive.
Demand forecast can be achieved at two different levels of detail: for long-
term planning (strategic level) and for medium-short planning (tactical or
operational level), where the difference is mainly due to the amount of required
input data and the resulting level of output information.
 AIR DEMAND MODELLING 85

At a strategic level (long term), the air demand forecast should support
hypotheses about both airport development and investment plans for a medium-
long period. Then, models able to simulate the expected demand levels as a
function of the past realizations – all the boundary conditions being the same –
can be more attractive. In other words, the demand trend is analysed under the
hypothesis that the underlying conditions (as the socio-economic system and/or
the transport supply system) are evolving with the same characteristics. If
different developing scenarios in terms of socio-economic and/or transport
supply systems can be tested, then specific hypotheses about the trend of the
variables representing such scenarios should be made, in order to obtain expected
demand levels consistent with them.
Models that better support the strategic planning are time series models that
generally use aggregate information referred to the dependent and independent
variables that occurred in a suitable time period (normally, at least 10 years).
At a tactical or operational level (medium-short time), the air demand
forecast should support operational hypotheses about modifications of the
system, particularly in terms of supply re-organization, with limited monetary
investments. In this case, models should use explicitly explanatory variables
simulating the supply characteristics (as frequency and/or departure times, fares,
land services, airport accessibility, available land mode to reach the airport and
so on) as well as user characteristics (as user type, age, trip purpose, family
income and so on). In other words, the level of details should be compatible with
the nature of the required demand estimate.
Models that better support the tactical and operational planning are discrete
choice models; they require a greater level of detail, but can provide more
information both about the expected characteristics of the air demand and the
potential share between competitive, alternative modes (as aircraft and fast
trains).

Time series Discrete choice

approach approach

Strategic level Tactical/operational level

(Long term forecast) (medium/short term forecast)

Figure 3: Approaches for air demand modelling and planning levels.

86 DEVELOPMENT OF REGIONAL AIRPORTS

However, time series models and discrete choice models can be used at both
planning levels (Figure 3), depending on the nature of the analysis and the data
availability as well as the required output detail levels.
For example, discrete choice models can be used to test hypotheses about the
development of the overall transport system, the air transport system being only
a part of it, in relation to long-term planning projects as the building of fast speed
railways or new road infrastructures. Similarly, time series models can be used to
verify airport developing policies; for example the introduction of new links or
variations in flight frequencies, for short term periods as 2–5 years.

Planning level:
• Strategic
• Tactical

Model Model
development application

Model Model choice

identification

Function Relevant
Function Relevant variables
identification variables
identification Model parameters

Relevant
Model calibration
variables data
base

Model check

Model
application

Figure 4: Model development and application.

 AIR DEMAND MODELLING 87

Forecast of the air transport demand can also be referred both to a single airport
inside a geographical (or administrative) region and to a set of airports inside a
common area where they can be considered ‘competitive’ to each other. The
choice to simulate the air demand only for one airport or to verify its distribution
among two or more of them depends on the effective competition among
airports, strongly linked to the identification of the airport catchment area.
Whatever be the planning level, the development and application of an air
demand model requires three main steps (see also Figure 4):
● identification of the most suitable mathematical model able to
simulate/forecast air transport demand with respect to the expected results
and/or the prefixed goals;
● availability of data to calibrate the model parameters and to apply the model;
● check of the obtained model.
In the application stage, an available model can be used to simulate the air
demand, the only care being the opportunity to use model parameters referred to
similar socio-economic contexts. In this case, the check stage concerns the
application of the model to a known situation in order to verify the congruence of
the parameters, while in the case of model development the check stage refers to
some statistical tests about the goodness of the estimated parameters and the
statistical reliability of the overall model.
One important aspect concerning both the development and application of an
air demand model is the data collection. Data referred to (air) transport demand
concerning both user socio-economic characteristics and travel behaviour are often
difficult to obtain. Generally, available data refer to official, aggregate statistics on
boarded/de-planed passengers, pro-capita income for geographical/administrative
regions and so on, but depending on the kind of model and detail required they can
be inadequate to develop a suitable demand model.
Moreover, travel times and costs are the most relevant level-of-service
variables introduced in a demand function. For air transport systems, travel times
refer to flight duration, possible waiting time for connecting flights, boarding/
disembarkation, baggage claim, and access/egress times. Costs mainly refer to
monetary costs and generally to airfare.
Some data concerning airline supply can be difficult to obtain without
specific surveys; particularly, airfare is the most difficult variable to quantify for
at least two main reasons: (1) useful data are not always available and (2) there is
a very large set of fares proposed by different air carriers and also inside the
same air carrier. Actually, airfares can change significantly depending on many
factors as the day on which the ticket is bought, the time period (week-end,
particular days or months of the year and so on), the number of booked people,
the age, the participation to flight programs (as frequent flier programs) and so
on. When international trips are considered, the problem is still more complex
because origins and/or destinations are in different countries with different
currencies, while the fare has to be expressed in one reference monetary value,
e.g. by using the exchange rate that, in turn, is variable during the year.
88 DEVELOPMENT OF REGIONAL AIRPORTS

To overcome the problem by considering the quality of the offered service

(and then, implicitly, the willingness to pay to use it), the hedonic pricing theory
can be used (Rosen [16]). Its basic foundations are that users evaluate the
characteristics of a good or the services it offers rather than the good itself.
Following this approach, the observed fare can be considered as a function of the
offered service and/or user characteristics; then, users are willing to pay
according to the satisfaction they receive.
The following sections provide a brief overview of the main characteristics
for both time series and discrete choice models.

3.1 Time series approach

Time series models to simulate air transport demand can have different levels of
complexity depending on the general aims and the data availability for both
model calibration and application. They have been largely used to predict air
demand levels, see among others [17, 18, 19, 20, 21, 22, 23].
To briefly summarize, a time series is a stochastic process where the time index
takes on a finite or countable infinite set of values. A stochastic process is an
ordered and infinite sequence of random variables: if the time index t assumes only
integer values, then it is a discrete stochastic process. To describe it, its mean and
its variance are used as well as two functions: the AutoCorrelation Function (ACF)
ρk, k being the lag, and the Partial AutoCorrelation Function (PACF) πk, k being
the lag. The ACF is a measure of the correlation between two variables composing
the stochastic process, which are k temporal lags far away; the PACF measures the
net correlation between two variables which are k temporal lags far away [24, 25].
AutoRegressive Moving Average (ARMA) models are a class of stochastic
processes expressed as follows:
p q
X t − ∑ φi X t − i = at − ∑ θ j at − j , (1)
i =1 j =1

where at is a White Noise process, φ and θ the model parameters, p and q the order
of the AutoRegressive (AR) and Moving Average (MA) processes, respectively
[24]. If the B operator such as Xt–1 = BXt is introduced, the general form of an
ARMA model can be written as follows:
φ ( B) ⋅ X t = θ ( B) ⋅ at .
To estimate these models, some conditions should be verified: the series must be
stationary and ACF and PACF must be time-independent. The non-stationarity in
variance can be removed if the series is transformed with the logarithmic
function. The non-stationarity in mean can be removed by using the operator ∇ =
(1–B) applied d times in order to make the series stationary. In this way, the
ARMA model becomes an ARIMA (AR Integrated MA) model:
∇ d φ ( B ) ⋅ X t = θ ( B ) ⋅ at . (2)
 AIR DEMAND MODELLING 89

This family of univariate models is largely used to obtain air demand prediction
at a first level of knowledge and when no more data other than the demand time
series is available. In this case, Xt represents the air demand at an airport i (or for
an origin/destination pair i, or a traffic zone i) at time t, dit:
∇ d φ ( B ) ⋅ dit = θ ( B) ⋅ at .
For a given set of data, the Box-Jenkins approach [24] is the most known method
to find an ARIMA model that effectively can reproduce the data generating the
process. The method requires three stages: identification, estimation and
diagnostic checking.
Preliminarily, data analyses should be carried out in order to verify the
presence of outliers. The identification stage provides an initial ARIMA model
specified on the basis of the estimated ACF and PACF, starting from the original
data:
● If the autocorrelations decrease slowly or do not vanish, there is non-
stationarity and the series should be differenced until stationarity is obtained.
Then, an ARIMA model can be identified for the differenced series.
● If the process underlying the collected series is a MA(q), then the ACF ρk is
zero for k > q and the PACF is decreasing.
● If the process underlying the collected series is an AR(p), then the PACF πk
is zero for k > p and the ACF is decreasing.
● If there is no evidence for a MA or an AR then a mixture ARMA model may
be adequate.
Several statistical tests have been developed in the literature to verify if a series is
stationary, among these, the most widely used is the Dickey-Fuller test
(Makridakis et al. [26]). After an initial model has been identified, the AR and MA
parameters have to be estimated, generally by using least squares (LS) or
maximum likelihood (ML) methods. The choice of the AR component order
derives from the analysis of the PACF correlogram; for large sample size, if the
order of the AR component is p, the estimate of the partial autocorrelations πk are
approximately normally distributed with mean zero and variance 1/N for k > p,
where N is the sample size. The significance of the residual autocorrelations is
often checked by verifying if the obtained values are within two standard error
bounds, ±2/√N, where N is the sample size (Judge et al. [25]). If the residual
autocorrelations at the first N/4 lags are close to the critical bounds, the reliability
of the model should be verified. Another test that can be used is the Ljung and Box
one [27]:
m
Q = N ⋅ ( N + 2) ⋅ ∑ ( N − k ) −1 ⋅ [ ρ aˆ (k )]2 ,
k =1

where ρ aˆ (k ) are the autocorrelations of the estimate residuals and k is a prefixed

number of lags. For an ARMA (p, q) process this statistic is approximately χ2
90 DEVELOPMENT OF REGIONAL AIRPORTS

distributed with (k–p–q) degrees of freedom if the orders p and q are specified
correctly.
To check the residuals normality, the Jarque-Bera (JB) test [28] can be used:

N − n p ⎛ 2 ( K − 3) ⎞
2

JB = ⋅⎜ S + ⎟,
6 ⎜ 4 ⎟
⎝ ⎠
where S is a measure of skewness, K is a measure of Kurtosis, np is the number
of parameters and N is the sample size. This test verifies if skewness and kurtosis
of the time series are different from those expected for a normal distribution.
Under the null hypothesis of normal distribution, the JB test is approximately χ2
distributed with two degrees of freedom.
Models (1) or (2) use the past values of the examined variable to predict its
future values. If some explanatory (or independent) variables are inserted in
order to verify cause-and-effect relationships, the dependent variable Xt generally
depends on lagged values of the independent variables and the model can be said
multivariate. The length of the lag may sometimes be known a priori, but usually
it is unknown and in some cases it is assumed to be infinite.
The simplest multivariate time series demand models are of the kind as
follows:
dit = β 0 + β T yit + uit ,
uit = ρ ui ,t −1 + ε it ,

where demand for an airport i (or for an origin/destination pair i, or a traffic zone
i) at time t, dit, is specified as function of n explanatory (and relevant) variables
yit. βT are the unknown model parameters, β0 the model constant, uit a random
term, εit a White Noise random residual and ρ the autocorrelation parameter
taking into account the time dependence among the variables. The basic
hypothesis is that the variable at year t is a function of the same variable at year
t–1, as specified by the random term.
More general models can be obtained by starting from univariate ARIMA
models and introducing more explanatory variables. Normally, if one dependent
variable and one explanatory variable are considered, then the model has the
form as follows:
dit = α + β 0 yit + β1 yi ,t −1 + … + β P yi ,t − P + eit , (3)

under the hypothesis that βk = 0 for k greater than a finite number P, called lag
length. Models (3) are called finite distributed lag models, because the lagged
effect of a change in the independent variable is distributed into a finite number
of time periods.
If e ~ (0, σ2I) and yt are fixed, then, based on the sample information, the LS
estimator is the best linear unbiased estimator for (α, β0, …, βP). If the true lag
length P is unknown but an upper bound M is known, then the LS estimator of
β’ = (α, β0, β1, …, βM)T is inefficient since it ignores the restrictions βP+1 = … =
 AIR DEMAND MODELLING 91

βM = 0. In order to compute P, these sequential hypotheses can be set up as

follows:
d o (k1 , k2 ,..., kn ) = no Π i p (ki ),
versus
m 1 2 m −1
H a
: P = M − m + 1, ⇒ β M − m +1 ≠ 0 H ,H 0 0
,...., H 0 .

The null hypotheses are tested sequentially beginning from the first one. The
testing sequence ends when one of the null hypotheses of the sequence is
rejected for the first time. The likelihood ratio statistic to test the m-th null
hypothesis can be written as follows:
SSE M − m − SSE M − m +1
λm = ,
σˆ M2 − m +1
where SSEP is the sum of the squared errors for a model with lag length P. This
statistic has an F-distribution with 1 and (T – M + m – 3) degrees of freedom if
H 01 , H 02 , H 0m are true.
When the lag has been computed, the explanatory variable can be inserted in
the univariate model, in order to derive a so-called multivariate ARIMAX model.
In the general case of more than one explanatory variables, the model has the
form as follows:
P1 P2
∇ d Φ ( B ) ⋅ dit = θ ( B )t ⋅ at + ∑ β t(1)
− l yi , t − l + ∑ β t − l yi , t − l +........
(1) (2) (2)
(4)
l =0 l =0

where: yt(−jl) is the j-th independent variable at time (t–l) and β t(−jl) is the
corresponding parameter.
Figure 5 shows the different kinds of applications of air demand time series
models, preferably for strategic planning levels.
Univariate models do not require explanatory variables but only the demand
past ‘history’; furthermore, they do not require the explicit identification of the
airport catchment area but, for example, only boarded/de-planed passengers at
the airport, time series data being available.
On the other hand, multivariate models present one or more explanatory
variables as frequencies, income, number of employment and so on; some of
them, as socio-economic variables, refer to the airport catchment area that has to
be explicitly identified.
As Figure 5 shows, univariate and multivariate models can be used at
aggregate and disaggregate levels; in the last case, the variables are defined for
each traffic zone, demand generated by each traffic zone at year t can be
estimated and then characterized as function of destination, departure time,
transport mode and so on, by using discrete choice models (Section 3.2).
Furthermore, mode-specific travel demand (as air demand) can be directly
92 DEVELOPMENT OF REGIONAL AIRPORTS

generated for each traffic zone, and again the other characteristics as destination
and departure time as well as airport, airlines, access mode are simulated.

Time series demand

Disaggregated level
Demand time series Dependent and independent
at airport or variables time series Traffic zones
for airport pairs

Aggregated level Univariate and

multivariate time series
models
Univariate time series Airport
models catchment area

Predicted demand level

Multivariate time series for each traffic zone
Predicted demand levels at time t
models
at the airport or
for airport pairs
Discrete choice models
Predicted demand levels
at the airport or
for airport pairs

Predicted demand level

for each traffic zone
and given characteristics
at time t

Figure 5: Application of time series models to estimate air travel demand.

3.2 Discrete choice models

Discrete choice models are a well-known class of models largely used in the
transportation field to obtain trip demand specified with some characteristics as
trip purpose, trip origin and destination, departure time, transport mode and so on
[14, 15]. The most general form of a discrete choice multistage demand model is:
d o (k1 , k2 ,..., kn ) = no Π i p (ki ), (5)

where do(k1, k2,…, kn) is the travel demand with trip origin o and characteristics
k1, k2, …, kn that can be specified from time to time depending on the exigencies;
no is the number of potential users in the origin o and p(ki) is the choice
percentage referred to the characteristic (or choice dimension) ki. They can be
estimated by using simple statistical approaches or Random Utility Models
(RUM).
 AIR DEMAND MODELLING 93

To estimate the air demand by starting from model (5), suitable choice
dimensions ki and the corresponding p(ki)s should be identified, as in the
following simple sequential specification:
d odh ( s, m) = d o ( sh) [SE, TS] ⋅ p (d / osh) [SE, TS] ⋅ p(m / odsh) [SE, TS] (6)

where SE and TS represent the vector of the socio-economic and territorial

system characteristics, because the choice percentages depend on both user
(socio-economic) and level-of-service/activity (territorial system) characteristics.
Indexes o, d, h, s, m represent the trip dimensions, respectively, trip origin, trip
destination, time period, trip purpose, trip travel mode, while p(./..)s represent the
choice probability (or choice percentage) for each choice dimension.
As models (6) shows, the order in the sequence also defines the dependence
of each choice dimension on the previous one by means of suitable variables; the
identification of the more suitable sequence is not a trivial task; particularly, for
mode-related choices it is not easy to identify the best and more reasonable
sequence to model the complex user behaviour concerning travel planning. For
example, the choice to travel by aircraft implies also the choices of departure and
arriving airports, the airport access/egress mode, the airline (e.g. traditional vs.
low-cost); the latter can be simulated by means of suitable variables inside the
mode choice dimension or within a decision process where the choice
dimensions, their hierarchical order, if any, and their reciprocal effects should be
simulated (more on airport choice is in Chapter 5).
In any case, the sequence (6) might be completely changed if the user choice
process happens in a different way, e.g. users travelling for leisure, first of all,
can decide to use an aircraft to start their trip and then make all the other mode-
related choices, included the trip destination. On the other hand, if destination is
compulsory (e.g. business travel), sequence (6) matches the decision process. In
any case, the identification of the best sequence and then the best model is the
result of a trial-and-error procedure.
Sequential discrete choice models as in eq. (5) have been used at national
level to simulate the trip demand on many available transport modes, included
aircraft, so as to define the best developing policies for the overall transport
system by taking into account also its impact. This approach could be
particularly useful to verify how much the overall transport system and each of
its components are responsible for the greenhouse effects and which actions
could be undertaken in order to satisfy the Kyoto protocol.
The hypotheses underlying a RUM approach to estimate the p(./..)s suppose
users are rational decision makers and they choose the best option among a set of
available alternatives; such a choice is based on the random utility value
associated to each alternative belonging to the choice set and depending on the
characteristics (attributes) of the alternative itself and the other available
alternatives. Then, users choose the option with the highest value of utility, and
since utility is a random variable only the probability that users choose a specific
alternative can be computed.
Starting from these hypotheses, many multistage RU discrete choice demand
models can be identified by specifying choice dimensions, choice sequences and
94 DEVELOPMENT OF REGIONAL AIRPORTS

discrete choice models (Figure 6); whatever be the discrete choice model, its use
requires: (1) the identification of the choice set; (2) the identification of the
relevant attributes characterizing each alternative; (3) the identification of the
mathematical form for the random utility variable [8, 29, 30, 31].

Figure 6: Steps in the discrete choice approach.

RUMs are a family of behavioural models trying to simulate user behaviour

starting from some mathematical hypotheses. Recently, other paradigms have
been proposed to understand user preferences, by using Neural Networks (NN)
and fuzzy-NN approaches [32, 33, 34]. However, NN do not allow the explicit
values of the parameters to be computed, so the interpretation of the model in
terms of elasticity values, parameter ratios and so on cannot be obtained.
Estimation of the air demand requires more than the mode dimension because
subsequent, relevant choices are also important, as the airport choice that allows
obtaining the number of (potential) passengers at the airport, or the airport access
mode in order to identify the needs of users and then identify solutions to offer
suitable landside facilities.

4 An overview of the Italian airport system

Currently, the number of airports opened to civil and military aviation on the
Italian territory is 115; among these, about 15 are military airports and the
remaining can be classified half as commercial and half as general aviation.
 AIR DEMAND MODELLING 95

Commercial airports refer to scheduled and charter flights, general and

military aviation. General aviation airports usually have short runways and
services are provided for medical and Civil Protection.
Following the EU classification, as shown in Table 2, in Italy there are 2
community airports (yearly passengers greater than 10 million), 5 national
airports (yearly passengers from 5 to 10 million), 14 large regional airports
(yearly passengers from 1 to 5 million) and 16 small regional airports (yearly
passengers less than 1 million).
In recent years, the air demand in the Italian market has increased, following
the general world tendency. According to forecasts provided by IATA and the
European agency for the air transport safety (Eurocontrol), passenger volumes
are expected to increase further in the future years; particularly, IATA [35, 36]
predicts an average increase at a rate of about 3–4% starting from 2005, while
Eurocontrol [37] predicts a yearly average increase in the European market at a
rate of about 3% during the period 2005–2025. Anyway, short-medium trends
are continuously revised due to contingent situations (as political stability, oil
price and so on).
As for the Italian market, Eurocontrol forecasts a yearly average increase of
about 2–3% for the domestic market, while international markets are expected to
increase at a rate of about 4% for the next 8 years.
In the period 1997–2006, the annual passenger growth rate in the Italian
market was about 6%; the Rome/Fiumicino-Milan/Linate pair has been the
second busiest route within the EU market, with more than two million
passengers carried (source: Eurostat, http://www.ec.europa.eu/eurostat).
The analysis of the current situation in the Italian air market shows a rather
scattered demand, probably due to the relatively high number of commercial
airports per square mile; at the same time, the absence of an effective and
widespread land network that can guarantee a suitable accessibility to/from the
airport (mainly regional airports but also many national airports) reduces the
potential air demand. Furthermore, in many cases the average distances among
airports are about 130–160 kilometres, supplied air services are often similar and
small airports are in competition to capture demand from overlapping catchment
areas.
In this context, an important role is also played by airlines and their
relationships with airports. Starting from the liberalization of the air transport
system in Europe, more than 30 new commercial airlines have risen in Italy, but
today less than half is still in the market and most of them have a marginal
market share. Routes between many Southern Italy regional airports and the
main airports (Rome, Milan above all) are operated by a few airlines without a
significant competitiveness among them, while for some others airports,
specially located in Northern Italy, supply exceeds demand.
Many regional airports offer point-to-point international links, often operated
by low-cost companies; many of them serve tourist destinations in Northern Italy
(as Venice, Florence) but seasonal flights at lower costs are also starting for more
decentralized regions (as Sardinia, Calabria), thus improving the tourist flows
96 DEVELOPMENT OF REGIONAL AIRPORTS

towards them. Low-cost airlines in Italy have a share of about 13% on the
domestic market and about 30% on the international market.
The Italian international traffic is greater than domestic traffic: in 2007, the
percentage of international passengers takes about 59% of the overall Italian
market, while in the period 2004–2007 the number of passengers carried on
international routes had an increase of more than 10%.
Table 2: Airport classification based on passengers traffic.

Source: ENAC – National Agency for Civil Aviation, Italy.

Critical aspects of the Italian air transport system are mainly the poor
accessibility, the under-utilization of the potential capacity for many airports and
a flight supply that sometimes does not match the transport demand needs.
To summarize, the Italian airport system situation is very heterogeneous, both
from an operational and an economic point of view. The two larger airports
(Rome Fiumicino and Milan Malpensa) gather the most part of the international
traffic (Table 2), but some others have succeeded in creating a good international
network that has increased the overall traffic volumes (as, e.g. the airports of
Pisa, Rome Ciampino, Bergamo, Bologna). Finally, airports where low-cost
carriers operate are also the most efficient, as it is expected, given the operational
characteristics of low-cost carriers.

5 Application to a test case

Demand simulation at an airport by using time series models requires the
knowledge of passengers and, possibly, explanatory variable time series data
referred to the airport catchment area. RUM approaches can be preferred if an in-
depth analysis has to be realized, particularly to understand the user behaviour
with respect to airlines, airfares, airport characteristics and so on. However, some
data are often difficult to obtain, particularly airfares. It is almost impossible to
have authorized airfare data, they can be obtained indirectly by using IATA data,
 AIR DEMAND MODELLING 97

but information on discounted airfare is very limited. Finally, the application of

RUMs often requires passenger data collected by means of suitable surveys,
while time series models can use general data easier to obtain (as population,
GDP and so on).
In any case, to have information on the air demand at an airport, as a starting
point to understand the airport general trend and without specific analyses on the
user choice behaviour, time series models can represent a useful tool.
The application refers to the airport of Reggio Calabria, in the South of Italy
(Figure 7), located very near the city centre (about 5 kilometres), well connected
to the main road network but weakly served by public systems (both buses and
trains). Thanks to its position in front of the island of Sicily and near the Aeolian
isles, it might become an important node of the Mediterranean transport system,
particularly with reference to leisure traffic flows.

Figure 7: Location of Reggio Calabria airport (Southern Italy) and its main competitive
airport within the same administrative region.

The main competitive airport located in the same administrative region (Lamezia
Terme) is about 140 kilometres far away. The second one, Catania Fontanarossa,
is located in Sicily and the overall land distance is about 135 kilometres, but the
access/egress time, included time spent to cross the Strait of Messina (between
Sicily and Calabria), makes it less attractive to potential users.
As reported in Table 2, Lamezia Terme airport can be classified as a large
regional airport and Catania as a national one, while Reggio Calabria is a small
regional airport.
Starting from 2005, Reggio Calabria airport management has begun some
developing policies by increasing the flight frequencies and the number of
reached destinations. The new flights have been operated by some low-cost-like
companies thus allowing lower fares for passengers. Before starting these
98 DEVELOPMENT OF REGIONAL AIRPORTS

developing policies, only hub-and-spoke flights were operated towards Rome

and Milan by only two companies, one of them under a significant monopoly
system. The new situation (more companies, more destinations both national and
international, competition on some links) has had as consequence:
● the end of the monopoly system and then the opportunity to have more
advantageous airfares;
● the opportunity to reach some destinations without transfer at hub(s);
● the increase of frequencies and the opportunity to choose among more
destinations and for the same destination among more departure times and
airfares.
The current situation is still in a developing but uncertain stage. In fact, the
presence of the competitive Lamezia Terme airport, that is continuously
expanding its supply and the served demand, makes the growth of the airport
difficult, given also that its role in terms of both kind of services and reached
destinations (and then market share) is not well defined if compared with the
competing airport.
Furthermore, the airport catchment area, obtained by means of some RP
surveys at the airport, is rather limited from a geographical point of view
(Figure 8), the most part being concentrated around the city of Reggio Calabria
(about 52% of demand is resident in the municipality area) and its province
(about 30%), while a little part comes from the city of Messina (Sicily). Finally,
a negligible percentage comes from the nearest provinces of the Calabria
administrative region.
During the years 2005 and 2006, three surveys were conducted at Reggio
Calabria airport, within the research project ‘Methods and models to forecast the
air passenger transport demand’, part of a more general national project entitled
‘Guidelines to plan the development of the Italian regional airports’. The goal of
the surveys was to understand the main characteristics of users at the airports, to
identify the catchment area and to have information on the airfare paid by users.
The first survey was realized immediately after the introduction of new links
by new air companies, and the second one after some months during which more
frequencies and more destinations were added. As Table 3 shows, the percentage
of users in the price class 50–100 increases notably from the first to the second
survey, while the percentage in the price class 0–50 is drastically reduced (really,
this price class is linked to the initial launch bargain of new destinations with
new companies). The increase of the percentage of users in the price class 50–
100 is probably due to the presence of low-cost-like companies, that has had as
an expected consequence the increase of competition on some routes and then a
reduction of the average airfare.
The available, aggregate data refer to passengers and the main supply
characteristics at the airports (as number of movements, frequencies, number of
operating airlines and so on) for a given period (sources: Italian Official Statistic
Institute ISTAT; Ministry of Infrastructure and Transport; Association of Italian
Airports: www.assaeroporti.it).
 AIR DEMAND MODELLING 99

Figure 8: Catchment area of Reggio Calabria airport.

Table 3: Comparison between the 1st and 2nd survey: Price class.
Price class (Euro) Users [%] 1st survey Users [%] 2nd survey
0–50 14.9 1.5
50–100 34.1 57.5
100–150 18.3 20.9
150–200 12.6 11.5
200–250 10.3 4.2
250–300 4.3 0.8
300–350 0.2 1.5
350–400 3.7 0.7
400–450 0.5 0.8
>450 1.1 0.5
Table 4 (and Figure 9) reports passengers data at the airport; note the outlier
referred to 2004 when the airport was closed during the months of March, April
and May for some adjustment work on the runway.
100 DEVELOPMENT OF REGIONAL AIRPORTS

Table 4: Boarded/de-planed passengers at the airport of Reggio Calabria (period

1989–2007)*.
Year Pax Year Pax Year Pax Year Pax
1989 157,225 1994 260,539 1999 543,041 2004 272,470
1990 245,711 1995 252,294 2000 538,048 2005 398,089
1991 222,571 1996 364,036 2001 481,857 2006 578,250
1992 246,306 1997 464,161 2002 463,662 2007 547,814
1993 266,782 1998 461,091 2003 441,795
* Data have been collected by using more sources, as a unique data base does not exist.

Figure 9: Passenger demand trend at the airport of Reggio Calabria (period 1989–2007).

After a positive trend from 1989 to 1999, the passenger demand has started to
decrease systematically in successive years till 2004. The main reasons for this
decrease are the progressive reduction of the supply and also the more and more
expensive airfares. After 2004, the demand trend seems essentially positive, but
the potential demand is probably higher even if a poor accessibility and the still
uncertain developing policy at the airport stop its expansion.
Starting from the same boarded/de-planned passenger data base, both univariate
and multivariate time series models have been calibrated.
Following the Box-Jenkins approach, some preliminary analyses have been
carried out; estimated ACF and PACF for the boarded/de-planned passenger time
series (Figure 10) show that ACF decreases linearly and the value of PACF at lag
1 is close to 1, i.e. there is mean non-stationarity that has been removed by
differencing the series once. After that transformation, the Dickey-Fuller test
applied to the differenced series confirms its stationarity. To remove the variance
non-stationarity, the series has been transformed by using the logarithmic
function. The estimate of the partial autocorrelation coefficients shows that only
π1 does not fall within the two standard error bounds ±2/√N (Figure 10), so the
order 1 can be established for the AR component. The same procedure is applied
to choose the MA component order by using the correlogram of ACF, that
 AIR DEMAND MODELLING 101

suggests a MA(2) component. Then, from data analysis the identified general
model is ARIMA(1,1,2):
(1 − φ B)∇ ln dit = c + (1 − θ1 B − θ 2 B 2 )at ,
where c is the model constant.

Figure 10: ACF and PACF correlograms.

Demand at year 2004 can be considered an outlier because the airport was closed
during three months; then, to calibrate the model, the outlier has been suitably
estimated in order to better follow the natural trend of the series. The model
calibration has been carried out by using data till year 2005 while the remaining
102 DEVELOPMENT OF REGIONAL AIRPORTS

2 years have been used as hold-out-sample to verify the model prediction

capability (Figure 11).
A multivariate ARIMAX model has been calibrated too, by introducing some
explanatory variables. As it is well known, the most powerful explanatory level-
of-service variables to estimate transport demand are times and monetary costs,
mainly airfares in this case. As said in Section 3, it is very difficult to obtain
airfares, and especially for long time periods, but airfares are also one of the
most interesting variables that can explain the air demand trend and its variations
when the market conditions change (e.g. low-cost companies, competition on
routes, within mode and between mode competitions and so on).

Figure 11: True and predicted air demand at the airport of Reggio Calabria – ARIMA
and ARIMAX models.

As briefly described in Section 3, airfare can be estimated indirectly by using the

hedonic pricing theory. Trip fare can then be expressed as follows:
F = f (a, b,… , d ),
where:
● F is the trip fare to move between two airports;
● f is a function to be specified;
● a, b, …, d are user and/or trip characteristics as comfort, timetables,
accessibility, delays, frequency and so on, which can be obtained by suitable
surveys at airports.
In this case, function f has been specified in linear form, and the relevant
variables are user average income, flight duration time, kind of air carrier and
transfer waiting time (Table 5). All the data have been collected during the RP
interview introduced before. The variable ‘kind of air carrier’ assumes value 1 if
users choose the flag (or traditional) carrier and 0 otherwise. Time variables
depend on the scheduled flights for the various legs and are expressed in
 AIR DEMAND MODELLING 103

hours. Finally, fares, transformed by using the logarithmic function, refer to one-
way trips.
Table 5: Results of the fare model.
Variable Coefficients t-student (1%)
Income 0.88 23.93
Flight duration 1.45 26.88
Kind of airline 0.47 6.43
Waiting time –0.82 –13.39
Statistical tests
R2 Adjusted R2
0.958 0.9579
As Table 5 shows, all the parameters have correct signs and are statistically
significant as well as the overall model (see R2 and adjusted R2). Users are
willing to pay more for longer trips, but prefer direct flights or good connections,
as the negative value of the waiting time variable suggests. Furthermore, despite
a greater monetary cost they prefer flag carriers, probably due to the image of
reliability and safety they inspire.
Generally, after the calibration of a fare model, its results can be used into a
demand model as ARIMAX. In any case, even if this fare model specification
has given good results in terms of descriptive power, for this application not all
the time series data of the explanatory variables are available and then the fare
model results cannot be used into an ARIMAX model. Then the explanatory
variables considered here are the number of movements at the airport at year t,
mt, and the average per capita income at year t, It, that can be considered a proxy
of the willingness to pay, and in some ways linked to the airfares at the airport.
The number of movements is a level-of-service variable, representing the
capability of the airport to offer flights, while income is a socio-economic
variable depending on the activity system in the catchment area.
The sequential testing procedure described in Section 3.1 allows the P values
for both variables to be identified; particularly, demand at year t depends on
movements in the same year t and income from year t to year t–6. The resulting
multivariate ARIMAX model is:

(1 − ΦB)∇ ln dt = (1 − θ1 B − θ 2 B 2 ) ⋅ at + δ ⋅ ln mt + α1 ln I t + α 2 ln I t −1 +
+α 3 ln I t − 2 + α 4 ln I t − 3 + α 5 ln I t − 4 + α 6 ln I t −5 + α 7 ln I t − 6 + κ .

The model has been calibrated by using data from 1989 to 2005, while the
remaining data (2006–2007) were used as hold-out sample (Figure 11).
As Figure 11 shows, both ARIMA and ARIMAX models can well predict the air
demand at the airport; years from 2005 to 2007, considered as hold-out sample,
are better simulated by the ARIMAX model. Anyway, it is interesting to note
that both models show good performances, although the theoretically most
appealing multivariate model needs more explanatory variables to capture the
104 DEVELOPMENT OF REGIONAL AIRPORTS

demand trend. In some cases, the ARIMA model explains the demand trend
better than the ARIMAX model. However, apart from the similar simulation
capabilities, multivariate models can help to test possible developing policies by
means of suitable hypotheses about the values of the explanatory variables. In
this case, the level-of-service explanatory variable (movements) depends on the
airport management and airline policies, while income depends on socio-
economic developing policies, more complex and more difficult to estimate and
control. Without specific developing policies on the territory and then all things
being equal, income follows its trend, while hypotheses can be made on the
number of movements in order to verify if and how demand can further increase.

number of movements

10000

8000

6000

4000

2000

0
1989

1990

1992

1994

1996

1998

1999

2000

2001

2003

2005

2007
1991

1993

1995

1997

2002

2004

2006
Figure 12: Number of movements trend at the airport of Reggio Calabria (period
1989–2007).
It is interesting to compare the trend of the number of movements (Figure 12)
and passenger demand percentage variations at the airport (Figure 13) as well as
the percentage variations with respect to 1999 (corresponding to the greatest
demand value before 2005, when new companies began to operate at the airport).
From Figure 13, it can be seen that the percentage variations of movements and
passengers are rather similar, apart from the transition year 2005, when many
developing policies started at the airport. More interestingly, Figure 14 shows the
percentage variations with respect to the reference year 1999. In this case, till
2004 demand is decreasing quicker than the number of movements with respect
to the reference year, but at a rather similar rate. After 2005, while the number of
movements increases strongly, demand increases weakly and simply reaches the
values already achieved at 1999.
The analyses of data thus suggest that the policies started at the airport do not
capture the actual needs of passengers, as demand at years 2006–2007,
practically equal to demand at year 1999, is satisfied by a supply really larger
than that at the reference year. Then, even if the number of movements increases
– suggesting more possible destinations, greater frequencies for the same
destination, more flights at different times of the day – demand does not increase
significantly with respect to the reference year 1999, when the number of
movements was largely lower, but probably more suitable for the passengers
needs.
 AIR DEMAND MODELLING 105

% pax variation % mov variation

80,0
60,0
40,0
20,0
0,0
-20,0
-40,0
-60,0
1999 2000 2001 2002 2003 2004 2005 2006 2007

% pax variation 17,8 -0,9 -10,4 -3,8 -4,7 -38,3 46,1 45,3 -5,3
% mov variation 13,8 -0,1 -2,1 -5,3 -4,5 -39,2 69,5 58,4 -5,1

Figure 13: Number of movements and passenger demand percentage variations at the
airport of Reggio Calabria.

Figure 14: Number of movements and passenger demand percentage variations at the
airport of Reggio Calabria with respect to year 1999.

To simulate the relationship between demand needs and air supply, probably
the time series model should use more level-of-service explanatory variables
taking into account not only the amount of supply but its distribution and its
specific characteristics. However, when the required level-of-detail increases,
data are more difficult to obtain, as official agencies at national and international
level (as Eurostat) generally provide considerably aggregate data. Then, specific
surveys have to be carried out that also allow combined time series and RUM to
be used.
As the application at the regional airport of Reggio Calabria has showed, the
regular collection of data at a given airport can be of great importance for the
airport management, helping them to identify the best developing strategies,
particularly when competition between airports exists. In this case, the decrease
in demand despite the increase in the flight number can be due to the superior
106 DEVELOPMENT OF REGIONAL AIRPORTS

supply at the nearest competing airport, that in fact has strongly increased its
demand, probably becoming attractive for people initially being in the Reggio
Calabria airport catchment area.
From a modelling point of view, the results obtained with both univariate and
multivariate time series models do not allow asserting that univariate models are
better than multivariate models and vice versa. As obtained in this study, the
better forecasting power of the univariate model is offset by its limits of validity,
which depends on the stability of the boundary conditions. Multivariate models
solve this problem by using explanatory variables, whose time series, however,
are often difficult to find. Thus, the potential explanatory power of multivariate
time series models is limited by the lack of suitable data.

References
[1] Graham, B. & Guyer, C., Environmental sustainability, airport capacity and
European air transport liberalization: Irreconcilable goals? Journal of
Transport Geography, 7(3), pp. 165–180, 1999.
[2] Cohas, F., Belobaba, P. & Simpson, R., Competitive fare and frequency
effects in airport market share modelling. Journal of Air Transport
Management, 2(1), pp. 33–45, 1995.
[3] Air Transport Action Group (ATAG), The economic & social benefits of air
transport, September 2005, www.atag.org.
[4] Italian Ministry of Transport, National Transport Analysis, Rome: Istituto
Poligrafico di Stato, 1998.
[5] Transport Canada, Regional and Small Airports Study (TP 14283B), 2004.
[6] Harvey, G., Airport Choice in a Multiple Airport Region. Transportation
Research, 21A(6), pp. 439–449, 1987.
[7] Ashford, N. & Benchemam, M., Passengers’ choice of airport: An
application of the multinomial logit model. Transportation Research
Record, 1147, pp. 1–5, 1987.
[8] Hess, S. & Polak, J.W., Mixed logit modelling of airport choice in multi-
airport regions. Journal of Air Transport Management, 11(2), pp. 59–68,
2005.
[9] Fuellhart, K., Airport catchment and leakage in a multi-airport region: The
case of Harrisburg International. Journal of Transport Geography, 15(4),
pp. 231–244, 2007.
[10] Van Reeven, P., de Vlieger, J. & Karamychev, V., BOB airport accessibility
pilot, final report within the Fifth Framework Program for research,
technological development, Erasmus University Rotterdam, Transport
Economics Dept, 2003.
[11] Milone, R., Humeida, H., Moran, M., Seifu, M. & Hogan, J., FY-2003
Models development program for COG/TPB travel models, Metropolitan
Washington Council of Governments, National Capital Region
Transportation Planning Board, 2003.
 AIR DEMAND MODELLING 107

[12] Suzuki, Y., Modelling and testing the ‘two-step’ decision process of
travellers in airport and airline choices. Transportation Research, 43E(1),
pp. 1–20, 2007.
[13] Button, K., The European market for airlines transportation and
multimodalism, Airports as multimodal interchange nodes, Economic
Research Centre, European Conference of Ministers of Transport 2005,
http://internationaltransportforum.org/europe/ecmt/pubpdf/05RT126.pdf.
[14] Cascetta, E., Transportation Systems Engineering: Theory and Methods,
Kluwer: The Netherlands, 2001.
[15] Ben Akiva, M. & Lerman, S., Discrete Choice Analysis, Cambridge, MA:
MIT Press, 1985.
[16] Rosen, S., Hedonic prices and implicit market: Product differentiation in
pure competition. Journal of Political Economy, 82, pp. 34–55, 1974.
[17] Hensher, D.A., Determining passenger potential for a regional airline hub at
Canberra International Airport. Journal of Air Transport Management, 8(5),
pp. 301–311, 2002.
[18] Andreoni, A. & Postorino, M.N., A multivariate ARIMA model to forecast
air transport demand. Proc. of the European Transport Conference 2006,
www.aetransport.org.
[19] Inglada, V. & Rey, B., Spanish air travel and September 11 terrorist attacks:
A note. Journal of Air Transport Management, 10(6), pp. 441–443, 2004.
[20] Karlaftis, M.G. & Papastavrou, J.D., Demand characteristics for charter air-
travel. International Journal of Transport Economics, XXV(3), pp. 19–35,
1998.
[21] Lim, C. & McAleer, M., Time series forecasts of international travel
demand for Australia. Tourism Management, 23(4), pp. 389–396, 2002.
[22] Lai, L. & Lu, L., Impact analysis of September 11 on air travel demand in
the USA. Journal of Air Transport Management, 11(6), pp. 455–458, 2005.
[23] Melville, J.A., An empirical model of the demand for international air travel
for the Caribbean region. International Journal of Transport Economics,
XXV(3), pp. 313–336, 1998.
[24] Box, G.E.P. & Jenkins, G.M., Time-Series Analysis, Forecasting and
Control, San Francisco: Holden-Day, 1970.
[25] Judge, G.G., Introduction to the Theory and Practice of Econometrics, 2nd
edition, New York: Wiley & Sons, 1988.
[26] Makridakis, S., Wheelwright, S.C. & Hyndman R.J., Forecasting. Methods
and Applications, 3rd edition, New York: Wiley & Sons, 1998.
[27] Ljung, M.G. & Box, G.E.P., On a measure of lack of fit in time series
model. Biometrika, 65, pp. 297–303, 1978.
[28] Jarque, C.M. & Bera, A.K., Model specification tests: A simultaneous
approach. Journal of Econometrics, 20(1, A), pp. 59–82, 1982.
[29] Bhat, C.R., A heteroscedastic extreme value model of intercity travel mode
choice. Transportation Research, 29B(6), pp. 471–483, 1997.
[30] Bhat, C.R., Covariance heterogeneity in nested logit models: Econometric
structure and application to intercity travel. Transportation Research,
31B(1), pp. 11–21, 1997.
108 DEVELOPMENT OF REGIONAL AIRPORTS

[31] Kroes, E., Lierens, A. & Kouwenhoven, M., The airport network and
catchment area competition model: A comprehensive airport demand
forecasting system using a partially observed database. Proceedings of
ERSA 2005.
[32] Postorino, M.N. & Versaci, M., A neuro-fuzzy approach to simulate the
user mode choice behaviour in a travel decision framework. International
Journal of Modelling and Simulation, 27(4), 2007.
[33] Sadek, A.W., Spring, G. & Smith, B.L., Towards more effective
transportation applications of computational intelligence paradigms.
Transportation Research Record, 1836, pp. 57–63, 2003.
[34] Pribyl, O. & Goulias K.G., Application of adaptive neuro-fuzzy inference
system to analysis of travel behaviour. Transportation Research Record,
1854, pp. 180–188, 2003.
[35] IATA, Air/rail intermodality study, Brussels, 2003.
[36] IATA, Passenger and freight forecast publications 2005–2009,
www.iata.org.
[37] Eurocontrol, STATFOR, Long-term forecast of air traffic (1006–2025)
report, www.eurocontrol.int/statfor, 2003.
5

Theory and practice in modelling air travel

choice behaviour
S. Hess
University of Leeds, UK

Abstract
There is growing interest in modelling the choices made by air travellers.
However, these choices are complex and many studies do not do that complexity
justice. In this chapter, we take another look at the different choices made by an
air passenger for a single journey and discuss ways of modelling these processes.
We make a number of recommendations for good practice and also highlight a
number of issues that need to be dealt with by the analysts. Finally, we present
an application making use of state-of-the-art modelling techniques in an air
travel behaviour context.

1 Introduction
An increasing number of studies are looking at modelling air travel choice
behaviour, mainly with the help of discrete choice models. Existing applications
range from the choice of air as a mode of travel [1] to the choice of airport in
multi-airport regions [2, 3, 4, 5, 6, 7], the choice of airline or fare classes [8, 9],
and the choice of access mode [10, 11]. Some applications look jointly at
multiple travel dimensions, for example the joint choice of airport and airline
[12] or even the choice of an airport, airline and access mode triplet [13]. Finally,
there is also an increasing reliance on advanced model structures available, used
for example in the representation of random taste heterogeneity across travellers
[6, 7, 14], or the multi-dimensional correlation between alternatives sharing sub-
choices along some of the travel dimensions [15].
Despite the recent progress in the area, a lot of work remains to be done.
Indeed, the choice processes undertaken by air travellers are arguably more
complex than those taken with other modes. While authors are gradually
110 DEVELOPMENT OF REGIONAL AIRPORTS

acknowledging this in their work, and while the use of advanced model
structures has allowed for more realism, many studies still make strong
assumptions that unduly simplify the choice processes. Crucially, there is also
still a general lack of understanding of the actual choice processes undertaken by
air travellers, a fact that is not helped by the dynamic nature of the problem at
hand, as witnessed for example with the advance of low-cost carriers.
This chapter aims to take an overview look at the modelling issues and makes
some suggestions for good practice. We first give an overview of the choice
processes undertaken by air travellers (Section 2). This is followed in Section 3
by a discussion setting targets for practical research in this area while also
acknowledging certain simplifications that are necessary in such analyses.
Section 4 is concerned with data issues, while Section 5 looks at the question of
model structure. Finally, Section 6 presents an empirical example, and Section 7
provides a brief summary of the chapter.

2 Air travel choice behaviour

The choices made by an air passenger for a single journey are complex and
involve decisions along a number of different dimensions, some of which are
strongly interrelated. In simple terms, the choices made by an air traveller can be
divided into three main subcategories, namely those at the origin side, those at
the destination side and those concerning the actual air journey. It is the latter
ones and to some extent the origin side decisions that have received the most
attention in the existing literature.
At the origin side of an air journey, a passenger chooses a departure airport
and also makes a number of choices relating to the ground level journey to this
departure airport. In many ways, the destination side choices are the mirror
image of those made at the origin side.
This leaves us with choices relating to the actual air journey. Here, the first
choice is that of an airline operating a route to the chosen destination. While
most passengers will travel on a single airline for the duration of their journey,
there are some routings on which passengers will have to rely on a combination
of airlines, a situation that has in recent years increased in complexity given that
a large number of routes are now operated under code share agreements. The
next level of choice is that of a routing, which looks first at the choice between
direct and connecting flights, before, when choosing a connecting flight, looking
at the choice between different routes in terms of the number of connections and
the choice of connecting airports. The final dimension of choice for the actual air
journey is that of timing, that is the choice of a departure time and a departure
date.
The above discussion has already shown that air journeys involve decisions
along a multitude of dimensions. What makes the analysis of these choices even
more complicated are the complex interdependencies between the various
dimensions of choice, both in terms of interactions as well as ordering of
priorities. To illustrate the latter point, it should be obvious that, for most
passengers, the choice of destination will influence the choice set in terms of
 THEORY AND PRACTICE IN MODELLING AIR TRAVEL CHOICE BEHAVIOUR 111

airlines or departure airports. However, there may also be passengers where the
decision to travel on a specific airline (e.g. low-cost airline) will determine the
choice set of possible destinations. From this perspective, a simultaneous
modelling approach is clearly preferable to a potentially misguided sequential
approach.
Finally, it should be acknowledged that passengers actually also make a
decision to use air in the first place, as opposed to travelling on a different mode.
For most destinations, this choice is a direct result of distance, making all other
modes either impractical or impossible. However, for a number of short-haul
destinations, there is increasing competition with high speed rail, and the
analysis of the choices in this context are an interesting area for further work.
However, mainly due to data requirements, most applications looking at detailed
air travel choice dimensions have to rely on the assumption that the choice of air
as a mode of travel has been made at a higher level, prior to making choices
relating to the actual air journey.

3 Guidance for good modelling practice and contrast with

status quo
The above discussion has highlighted the complexity of the choice processes
undertaken by air travellers. This section sets out some guidelines for good
practice and contrasts them with the status quo in modelling work.

3.1 Guidelines

3.1.1 Recognize the multi-dimensional nature of the choice process

Air travellers make choices among a multitude of travel dimensions, and not
recognizing this in practical work can potentially lead to biased results. As an
example, let us look at the case of a study that looks solely at the choice of airport but
not the choice of airline. If a traveller in this study has a strong allegiance to an airline
flying from a smaller airport with low overall frequency, then a simple airport choice
model will be confused by the fact that this passenger actually flies from an airport
with lower frequency to the desired destination. From this point of view, it is
important to work on the basis of frequencies specific to airport-airline pairings, and
not airport-specific frequency, as has commonly been the case in existing work. The
same reasoning applies to other combinations of choice dimensions. In practice, this
issue can be addressed by dividing the multi-dimensional alternative into a
combination of elementary alternatives, with a passenger simultaneously choosing
one alternative along each dimension of choice.

3.1.2 Account for correlation along different choice dimensions

By understanding the multi-dimensional nature of the choice process, it becomes
evident that some of the combined alternatives share the attributes of other
alternatives along one or more of the choice dimensions (e.g. same airline, same
access mode). This clearly creates correlation between alternatives, and in the
likely case where not all characteristics are observed by the analyst, this will
112 DEVELOPMENT OF REGIONAL AIRPORTS

result in correlated error terms. Here, it is important to account for this

correlation, as described in Section 5.2.

3.1.3 Use highly disaggregate level-of-service data

Many existing studies have produced poor results partly as a result of using an
insufficient level of disaggregation in the level-of-service data, often as an effect
of using simplifications along a number of choice dimensions. While some
aggregation is almost inevitable, the use of an excessive amount of aggregation
can lead to biased results. This can for example arise in the case of studies
making use of weekly (or even monthly) data instead of daily data, hence
ignoring the often significant variations in level-of-service attributes between
different days of the week, especially in terms of flight frequencies.

3.1.4 Account for differences in behaviour across travellers

As in most contexts, there are significant differences between individual air
travellers in their sensitivities to attributes affecting their journey, for example air
fares and departure times. As a result, it is crucial to take such taste heterogeneity
into account at the modelling stage, an issue that is addressed in more detail in
Section 5.3.

3.2 Limitations of practical research

Without exemption, existing studies of air travel choice behaviour use a number
of simplifications of the choice processes, often due to data issues. These
simplifications can be looked at in turn for each of the various dimensions of
choice:
● The choice of destination and the actual decision to travel are not generally
modelled, where a reservation applies for the latter in the case of SC surveys
giving respondents the option not to travel. It is thus normally assumed that
these decisions are taken at an upper level, prior to the air-journey specific
choices.
● As already mentioned above, the same reasoning applies to the decision to
travel by air. Here, it is thus important to acknowledge that the estimates
obtained from such models relate to the part of the population that has
decided to travel by air, and are not representative of the overall population.
● The majority of studies of air travel choice behaviour look solely at the
choice of departure airport and ignore the choice of arrival airport, a
simplification primarily resulting from data issue.
● Just as for the choice of airport, the analysis of ground level decisions is
generally limited to the origin end, again primarily due to data reasons.
Additionally, most studies are only able to look at the choice of main mode,
ignoring the possibility of trip-chaining, as well as the choice of different
routes. The effects of these restrictions are dependent on the geographical
context.
 THEORY AND PRACTICE IN MODELLING AIR TRAVEL CHOICE BEHAVIOUR 113

● In RP models, trip timing cannot generally be modelled due to a relative lack

of information on preferred departure time and flight availabilities. These
issues do not apply in the case of SC models.
● The issue of flight routing is generally left untreated in RP studies, while in
SC studies, this is often limited to a choice between direct and connecting
flights.
● Given the major role that airline allegiance plays for some travellers,
advanced studies increasingly model the choice of airline in addition to the
choice of airport. However, combinations of airlines are generally not
allowed in such models, and additional issues can arise in the case of code
share flights.

4 Data issues
One of the main problems that need to be faced in the modelling of air travel
choice behaviour is that of data quality. This is the topic of the present section,
where the discussion is divided into two parts, looking first at RP data before
turning our attention to SC data.

4.1 Revealed preference data

4.1.1 Availability and attributes of unchosen alternatives

With RP data, the main issue that needs to be faced is the relative lack of
information on the choice set that the traveller was faced with. While detailed
information is generally available on the chosen flight, this does not extend to
unchosen alternatives. In fact, in many cases, it is not even just the attributes of
these alternatives that are unknown, but also their availability to a given traveller
at a given time. Generally, an assumption of availability is required, alongside
significant aggregation of level of service attributes for the unchosen flight
options. Similar problems also arise along other dimensions, such as the access
journey related choices, where again information on the availability and
attributes of unchosen modes is often not available.

4.1.2 Fare data

Air fares should naturally be expected to play a major role in air travel choice
behaviour. Despite this, the majority of RP studies of air travel choice behaviour
have struggled or been unable to retrieve meaningful marginal fare effects. This is
almost certainly a direct effect of the poor quality of the fare data, characterized by
a high level of aggregation. Indeed, it is in most studies only possible to obtain
information on the average fare charged by a given airline on a specific route. This
clearly involves a great deal of aggregation, as no distinction is made between the
fares paid across different travellers (i.e. in terms of travel classes as well as
booking classes). With the dynamic nature of air fares, the levels used in modelling
thus often bear little or no resemblance to those actually faced in the choice set,
with the unavoidable implication of poor modelling results. Even though some
progress can be made with the help of bookings data, issues of aggregation do
114 DEVELOPMENT OF REGIONAL AIRPORTS

remain. In fact, it can be seen that, in RP studies, disaggregate choice data is used
in conjunction with aggregate level-of-service data, for at least some of the
attributes. While, for some characteristics, this may be acceptable, it does, as
described above, create significant problems in the treatment of air fares, and flight
availability by extension.

4.1.3 Frequent flier information

While it is well known that airline allegiance as a result of membership in
frequent flier programmes plays an important role in air travel choice behaviour,
such information is often not collected in passenger surveys. As a result, this
potentially crucial influence on choice behaviour cannot usually be taken into
account in RP case studies.

4.1.4 Influence of other attributes

While most applications generally only look at a limited set of attributes, such as
access time, flight time, frequency and fare, various other factors, such as on-
time performance and in-flight entertainment, conceivably also have an influence
on travellers’ choices. Often, information on such attributes is however not
available, potentially significantly increasing the error terms in the models,
although some effects may be captured in airline specific constants.

4.1.5 Survey design issues and inter-dataset compatibility

The main input into RP air travel behaviour studies comes in the form of data
collected at departure airports, generally by airport operators or civil aviation
authorities. The factors of interest in these questionnaires often differ from those
relevant to a modelling analysis (e.g. the lack of information on frequent flier
programme membership), with the obvious impacts this has on any analysis.
Additionally, compatible sources of level-of-service data need to be found.

4.2 Stated choice data

Some of the problems discussed above in the context of RP data can be

alleviated by making use of SC data collected on the basis of tailor-made
surveys, designed specifically for the use in advanced modelling analyses. Here,
the main advantage is the fact that complete and accurate information is available
on all alternatives faced by the respondent. From this point of view, it should
come as no surprise that studies making use of SC data have been much more
successful in retrieving significant effects for factors such as air fares and airline
allegiance [cf. 16, 17]. However, it is important to remember that the use of SC
data does pose some philosophical problems, in terms of how the behaviour
differs from that observed in RP data [see, 18]. Additionally, issues of survey
complexity need to be addressed. To allow respondents to better relate to the
presented choices, surveys now often include a real-world reference alternative
in the choice set; this however poses some additional issues, as discussed
recently by Hess [19]. Both RP and SC data have advantages in their own right,
and to a large extent, the choice of the optimal approach depends on the issues to
be investigated, as well as the quality of the level-of-service data in the RP
 THEORY AND PRACTICE IN MODELLING AIR TRAVEL CHOICE BEHAVIOUR 115

context. An interesting approach in this context is to combine RP and SC data, as

done by Algers and Beser [20], hence correcting for the bias inherent to models
estimated on SC data. The problem in this case however is one of obtaining
compatible RP and SC datasets.

5 Model structure
Probably the most important question to address at the modelling stage is the
choice of a mathematical structure. Given the important differences across
travellers both in terms of behaviour as well as choice context, the use of a
disaggregate modelling approach is clearly preferable to an aggregate one, and
here, discrete choice structures belonging to the class of random utility models
(RUM) have established themselves as the preferred approach. For an in-depth
discussion of these modelling structures, see Train [21]. Here, we look solely at
two main issues of great relevance in the analysis of air travel behaviour, namely
that of inter-alternative correlation in the error terms (Section 5.2) and that of taste
heterogeneity (Section 5.3) across travellers. This is preceded by a brief
introduction of some common notation.

5.1 Basic concepts

A discrete choice model looks at the choices made by a decision maker n

amongst a set of mutually exclusive alternatives contained in a choice set Cn.
Each alternative i = 1, …, I in the choice set has an associated utility Ui,n, which
is specific to decision-maker n, due to variations in attributes of the individuals,
as well as in the attributes of the alternative, as faced by different decision-
makers. Under the assumption of utility maximising behaviour, respondent n will
choose alternative i if and only if Ui,n > Uj,n ∀j ≠ i, with i, j ∈ Cn.
The utility of an alternative is a function of its attributes and the tastes of a
decision maker. Given inherent randomness in behaviour as well as data
limitations, only part of the utility can be observed, such that we rewrite:
U i ,n = Vi ,n +ει ,n , (1)

with Vi,n and εi,n giving the observed and unobserved parts of utility, respectively.
Here, Vi,n is defined as f(βn, xi,n), where xi,n represents a vector of measurable (to
the researcher) attributes of alternative i as faced by decision-maker n,1 and βn is
a vector of parameters representing the tastes of decision-maker n, which is to be
estimated from the data.
Due to the presence of the unobserved utility term εi,n, the deterministic
choice process now becomes probabilistic, leading to a RUM, with the
alternative with the highest observed utility having the highest probability of
being chosen, where the individual probabilities are given by:
Pn (i ) = P(ε j , n − ε i , n < Vi , n − V j , n ∀ j ≠ i ). (2)

After noting that the mean of the unobserved utility terms can be added to the
observed part of utility in the form of an alternative-specific constant (ASC), the
116 DEVELOPMENT OF REGIONAL AIRPORTS

vector εn = {ε1,n, …, εI,n} is now defined to be a random vector with joint density
f (εn), zero mean and covariance matrix Σ. From this, we can rewrite (2) as
follows:
Pn (i ) = ∫ I (ε
εn
j ,n − ε i ,n < Vi,n − V j,n ∀ j ≠ i ) f (ε n )dε n , (3)

where I (⋅) is the indicator function which equals 1 if the term inside brackets is
true and 0 otherwise. Different assumptions on the distribution of the error terms
lead to different forms for the choice probabilities, and the multi-dimensional
integral in Equation (3) will only take a closed form for certain choices of
distribution for εn. In the most basic model, the multinomial logit (MNL) model,
the error terms are assumed to be distributed identically and independently (iid),
with more advanced models allowing for complex interdependencies.

5.2 Correlation between alternatives

As mentioned in Section 3.1.2, it is clearly a major and probably unwarranted

assumption to rule out the presence of heightened correlation in the unobserved
utility terms along any of the choice dimensions. Indeed, it cannot generally be
expected that any commonalities between two alternatives sharing a common
component along one or more of the choice dimensions would need to be
explained in the observed part of utility. This makes basic models that assume
independence of error terms, such as MNL, inappropriate for such modelling
purposes.
The typical departure from the MNL model in the context of air travel choice
behaviour has been to make use nested logit (NL) models which form the most
basic nesting structure within the generalized extreme value (GEV) family of
models, introduced by McFadden [22]. This set of models are all based on the
use of the extreme-value distribution and allow for various levels of correlation
among the unobserved part of utility across alternatives. This is accomplished by
dividing the choice set into nests of alternatives, with increased correlation, and
thus higher cross-elasticities, between alternatives sharing a nest. Alternatives
sharing a nest are more likely substitutes for each other. These structures are
generally represented by an upside-down tree, with the root at the top,
elementary alternatives at the bottom and composite alternatives, or nests, in
between.
In a two level NL model, alternatives are grouped into mutually exclusive
nests, where, for each nest, a structural (nesting) parameter is estimated that
relates to the level of correlation in the error terms of alternatives sharing that
nest. One example of such a structure is to nest the alternatives by airport, that is
allowing for heightened correlation between alternatives sharing the same
departure airport. This structure is illustrated in Figure 1, with K mutually
exclusive nests, one for each airport, and where each nest has its own nesting
parameter, λk, thus allowing for different correlation levels with different
airports.
 THEORY AND PRACTICE IN MODELLING AIR TRAVEL CHOICE BEHAVIOUR 117

Figure 1: Structure of two-level NL model, using nesting along airport-dimension.

The example in Figure 1 can be replicated for the case of correlation along the
airline dimension as well as correlation along the access mode dimension (or
other dimensions of choice). However, the structure discussed so far only allow
for correlation along one dimension of choice at a time. While it is possible to
extend the NL structure to multiple dimensions of nesting, as shown in Figure 2
for the case of airport and airline choice,2 where πl is used as the nesting
parameter for airline nests, this is not optimal. Indeed, the ordering of choice
dimensions plays a role and the full level of correlation is only allowed for along
the highest level of nesting. Indeed, by nesting the alternatives first by airport,
and then by airline, the nest for airline l inside the nest for airport k will only
group together the options on airline l for that airport k. The model is thus not
able to capture correlation between alternatives using airline l at airport k1 and
alternatives using airline l at airport k2, which is clearly a restriction. Finally, it
can also be noted that this structure can only accommodate correlation along all
but one of the dimensions of choice. Indeed, using the example shown in Figure
2, it can be seen that, by adding in an additional level of nesting by access mode
below the airline level, each access mode nest would contain a single alternative,
as the airline nest preceding the access mode nest would contain exactly one
alternative for each access mode. As such, the lower level of nesting becomes
obsolete.
118 DEVELOPMENT OF REGIONAL AIRPORTS

Figure 2: Structure of three-level NL model, using nesting along airport-dimension and

airline-dimension.

Figure 3: Structure of CNL model for the joint analysis of correlation along the airport,
airline and access mode dimensions.
 THEORY AND PRACTICE IN MODELLING AIR TRAVEL CHOICE BEHAVIOUR 119

Both of these problems can be addressed by using a cross-nested logit (CNL)

model, as discussed in the application in Section 6. This structure would be
specified by defining three groups of nests, namely K airport nests, L airline
nests and M access mode nests, and by allowing each alternative to belong to
exactly one nest in each of these groups. The resulting structure allows for
correlation along all three dimensions of choice and does so in a simultaneous
rather than sequential fashion, that is not giving priority to one dimension of
choice.
An example of such a model is shown in Figure 3, where, in addition to the
previously defined λk and πl, ψm is used as the structural parameter for access
mode nest m. Again, only a subset of the composite nests and of the triplets of
alternatives is shown. Additionally, the allocation parameters [cf. 21, Chapter 4],
governing the proportion by which an alternative belongs to each of the three
nests, are not shown in Figure 3.

5.3 Variation in behaviour across respondents

Another important issue that needs to be addressed during model specification

are potential differences between travellers in their sensitivities to changes in
attributes defining the alternatives. Such differences can clearly be seen to be
likely to apply to air fare and travel sensitivities, but may also extend to other
factors such as the willingness to pay for flying on a certain airline or on a
certain type of aircraft. From this point of view, it is clearly a major assumption
to make an assumption of taste homogeneity.
Various possible approaches exist to allow for taste heterogeneity. The most
basic approach is to use discrete segmentations of the sample population,
estimating separate models for different subsets of the sample, or in a less
extreme case, separate coefficients within the same model. There are also many
situations in which tastes can be expected to be related to socio-demographic
attributes, such as lower cost sensitivity with rising income. While it would be
possible to segment the sample population into different income classes, it may,
in many cases, be preferable to allow for continuous interactions between tastes
and sensitivities. Here, the modeller needs to additionally make a choice between
linear or non-linear interactions.
Due to data limitations as well as inherent randomness in sensitivities, not all
taste heterogeneity can be explained in a deterministic fashion, and analysts
are increasingly turning to models such as mixed multinomial logit (MMNL)
that allow for a representation of random (as opposed to deterministic) taste
heterogeneity. In a MMNL model, the choice probabilities are given by integrals
of MNL probabilities over the assumed distribution of taste coefficients.
Although incredibly flexible, the MMNL model comes with a high
computational cost due to the reliance on simulation in estimation and
application.
The various methods for representing taste heterogeneity are characterized by
differences in terms of flexibility, ease of implementation/estimation and ease of
interpretation. These factors are strongly correlated, with more advanced
120 DEVELOPMENT OF REGIONAL AIRPORTS

specifications offering gains in flexibility at the expense of higher computational

cost as well as issues in interpretation. Nevertheless, whichever approach is used,
it is important to at least challenge the assumption of taste homogeneity.

5.4 Discussion

The modelling approaches described in Section 5.2 and Section 5.3 have quite
separate aims; the analysis of inter-alternative correlation, along multiple
dimensions, and the representation of deterministic and random variations in
choice behaviour. When both phenomena play a role simultaneously, analysts
may need to rely on even more advanced model structures that allow for the joint
representation of inter-alternative correlation and random taste heterogeneity
[cf. 23, 24].

6 Empirical example
For the empirical example used in this section, we summarize part of the results
described in the study by Hess and Polak [15]. Particularly, we look at the
combined choice of airport, airline and access mode for air passengers departing
from Greater London.

6.1 Description of data

The application described here made use of RP data, with information on actual
trips taken from the 1996 passenger survey conducted by the Civil Aviation
Authority [25]. The final sample used for the analysis reported here contained
data from 8704 respondents travelling for business reasons with the London
departure being the outbound leg of the journey. A total of 31 destinations used
in the analysis, all served by a single airport, and spread across Great Britain,
Europe, the Middle East and North America. For short haul destinations an
assumption was made that respondents made an a priori assumption to travel by
air. Air-side level-of-service data were obtained from BACK aviation, with
information on fares compiled from the International Passenger Survey [26] and
the fare supplement of the Official Airways Guide for 1996 [27]. As is the case
with most RP studies, the resulting dataset is of highly aggregate nature, leading
to the previously discussed problems in the estimation of the marginal utility of
air fares (cf. Section 4.1.2). Additionally, no information was available on
frequent flier programmes. Finally, for the analysis of the ground-level choice
dimension, data from the national airport access model (NAAM) were obtained
for the base year 1999 [28], and corresponding cost information for 1996 was
produced with the help of the retail price index, while assuming that relative
travel times have on average stayed constant.

6.2 Model structure and specification

The final choice contained five departure airports, namely Heathrow (LHR),
Gatwick (LGW), Stansted (STN), Luton (LTN) and London City (LCY). This
 THEORY AND PRACTICE IN MODELLING AIR TRAVEL CHOICE BEHAVIOUR 121

mix of major airports, outlying airports and a small city centre airport means the
area if of special interest in the context of the present book. Along the airline
dimension, there were 37 options, airlines, with six modes available along the
access journey dimension.3 This leads to a total of 1110 combinations of airports,
airlines and access modes arise. However, with not all airlines operating from all
airports, the total number of airport–airline pairs is actually 54, which reduces
the number of alternatives (airport, airline, access mode triplets) to 324.4
A large number of different attributes were used in the initial modelling
analysis, including attributes relating to the air journey, such as frequency, fare,
flight time, aircraft type and seat capacity, and attributes relating to the access
journey, such as access cost, in-vehicle access time (IVT), out-of-vehicle access
time, wait time, number of interchanges and parking cost. Different model
structures were investigated, namely MNL, two-level NL and CNL, where in all
models, weights were used in the specification of the log-likelihood function to
account for the quota used in data collection, which are not representative of the
population level.

6.3 Model results

Despite an extensive specification search, only a limited number of attributes

were observed to have a significant impact on behaviour, namely access cost,
IVT, flight frequency and flight time, where a log-transform was used for all four
attributes. Crucially, no effect could be identified for air fare, where this is at
least partly due to the poor quality of the data as discussed in Section 4.1.2. The
list of significant attributes was observed to stay identical across model
structures, i.e. MNL, NL and CNL.
Table 1 presents a summary of the mathematical performance of the different
models. Here we can see that all three NL models outperform the MNL model,
with the best performance offered by the model using nesting by access mode. In
turn, the CNL model outperforms not only MNL but also all three NL structures.
Finally, the total improvement of the CNL model over the MNL model is bigger
than the combined improvements in the adjusted ρ2 measure for the three NL
models. However, while the results show that the CNL model offers significant
improvements over the more simple NL models, these come at the expense of a
very significant increase in estimation cost.
Table 1: Model performance on London data.
Model Final LL Parameters Adjusted ρ2
MNL –14,945.3 55 0.3445
NL by airport –14,896.1 59 0.3465
NL by airline –14,870.7 74 0.3469
NL by access mode –14,816.7 60 0.3499
CNL –14,603.9 91 0.3578
122 DEVELOPMENT OF REGIONAL AIRPORTS

Table 2: Model results for London data.

IVT vs. access Freq. vs. Freq. vs. Flight

cost access cost IVT time vs.
IVT
(£/hour) (£/flight) (hours/flight)
MNL
Minimum 1.18 0.02 0.01 0.04
Mean 16.24 1.56 0.11 1.07
Maximum 143.38 231.05 4.06 7.43
Standard 25.44 4.85 0.18 0.7
deviation
NL by airport
Minimum 1.3 0.02 0.01 0.04
Mean 17.85 1.63 0.11 0.97
Maximum 157.65 242.38 3.87 6.72
Standard 27.98 5.09 0.17 0.63
deviation
NL by airline
Minimum 1.29 0.03 0.01 0.04
Mean 17.76 1.79 0.12 1.13
Maximum 156.8 265.11 4.26 7.85
Standard 27.83 5.57 0.19 0.74
deviation
NL by access mode
Minimum 0.99 0.02 0.01 0.04
Mean 13.52 1.11 0.1 1.05
Maximum 119.35 164.74 3.47 7.31
Standard 21.18 3.46 0.15 0.69
deviation
CNL
Minimum 1.16 0.01 0 0.04
Mean 15.96 0.9 0.07 0.95
Maximum 140.89 132.96 2.38 6.59
Standard 25 2.79 0.1 0.62
deviation
We next turn our attention to substantive results, which are summarized in Table
2. Given the use of the log-transform in nominators as well as denominators, the
various trade-offs were calculated separately for each individual,5 and summary
statistics were then calculated across respondents. We can see that the first three
models produce roughly similar results, while those produced by the CNL model
and the NL model using nesting access mode are more extreme (when compared
to the three first models), especially when looking at the value of travel time
savings (VTTS) measures for the model using nesting by access mode.
Another observation that can be made for the trade-offs is that the VTTS
measures are markedly lower than those reported, for example, by Pels et al. [3],
 THEORY AND PRACTICE IN MODELLING AIR TRAVEL CHOICE BEHAVIOUR 123

although they are still higher than in other contexts, which can be explained
partly by concepts of risk-averseness, as discussed, for example, by Hess and
Polak [7]. Travellers are willing to pay for a reduction in the risk of missing their
flight, where this risk clearly increases with access time. The still high values
should also be put into context by noting that the average access journey in this
population segment was measured as 57 minutes.
Finally, a detailed analysis of the correlation structure in the different models
[cf. 15] highlights the presence of high levels of correlation between the errors
for alternatives sharing the same airport, airline or access mode. Here, the CNL
model was more successful at retrieving these correlation patterns than was the
case for the different NL structures.

7 Summary and conclusions

The discussion in this chapter has highlighted the complexity of the choice
processes undertaken by air travellers, with decisions being taken along a
multitude of dimensions. In practice however, it is almost inevitable to use some
simplifications of the choice process, partly because of modelling complexity,
but mainly because of data issues. In this context, the advantages of SC data
make the use of such datasets an important avenue for further research,
potentially in conjunction with compatible RP data.
The problems that need to be faced when making use of RP data were also
highlighted in the empirical example in Section 6 where it was not possible to
retrieve significant effects for a range of important variables that included air
fares, schedule delay, and airline and airport allegiance. From a model structure
point of view, the application has shown that the use of more advanced model
structures can lead to improvements in model fit. However, although the
improvements are statistically significant, they are too small to lead to any major
differences in model performance. Nevertheless, the advanced model structures
provide further insights into choice behaviour, and there are also differences in
the substantive results between the various models.
The one common observation from this application and that of other studies
is that the results do suggest that access time plays a major role in the choice
process, with passengers having a strong preference for their local airport. As
such, the attractiveness of outlying airports depends heavily on good access
connections, unless there are other incentives, such as low air fares. This is
reflected in the fact that only low-cost carriers find it relatively easy to attract
passengers to outlying airports that are not served by convenient and fast ground-
level services. It is conceivable that the sensitivity to access time decreases with
flight time [cf. 17], such that moving long haul services to outlying airports
would seem wise; this however causes problems as the associated (and
necessary) short haul feeder flights will also carry point-to-point passengers, who
will again have a preference for more centrally-located airports.
124 DEVELOPMENT OF REGIONAL AIRPORTS

Acknowledgements
The author would like to acknowledge the input of John Polak in earlier stages of
this work and would similarly like to thank the Civil Aviation Authority and the
Department for Transport for data support.

Notes
1. The vector xi,n potentially also includes interactions with socio-demographic attributes of
respondent n.
2. It is important to stress that this should not be seen as representing a sequential choice
process. Rather, it means that there is correlation between two alternatives that share
the same airport, but that the correlation is larger if they additionally share the same
airline.
3. The options were private car, rental car, public transport (rail, bus, local transport),
long distance coach, taxi and minicab (MC), where, for data reasons, no combinations
of modes were considered in the present analysis.
4. The number of available alternatives for specific individuals in the estimation sample
ranges from 6 to 58, with a mean of 31.
5. With U = … + β1 ln (x1) + β2 ln (x2) + …, the ratio of the partial derivatives of U with
respect to x1 and x2 is given by β1/β2 x2/x1, as opposed to the simple β1/β2 ratio used in
the case of a linear parameterization.

References
[1] González-Savignat, M., Competition in air transport: The case of high speed
train. Journal of Transport Economics and Policy, 38(1), 77–108, 2004.
[2] Pels, E., Nijkamp, P. & Rietveld, P., Airport and airline choice in a multi-
airport region: An empirical analysis for the San Francisco bay area.
Regional Studies, 35(1), pp. 1–9, 2001.
[3] Pels, E., Nijkamp, P. & Rietveld, P., Access to and competition between
airports: A case study for the San Francisco Bay area. Transportation
Research Part A: Policy and Practice, 37(1), pp. 71–83, 2003.
[4] Pathomsiri, S., Mahmassani, H. & Haghani, A., Airport Choice Model for
Baltimore-Washington Region. Paper presented at the 83rd annual meeting
of the Transportation Research Board, Washington, DC, 2004.
[5] Basar G. & Bhat, C.R., A parameterized consideration set model for airport
choice: An application to the San Francisco Bay area. Transportation
Research Part B: Methodological, 38(10), pp. 889–904, 2004.
[6] Hess S. & Polak, J.W., Accounting for random taste heterogeneity
in airport-choice modelling. Transportation Research Record, 1915,
pp. 36–43, 2005.
[7] Hess, S. & Polak, J.W., Mixed logit modelling of airport choice in multi-
airport regions. Journal of Air Transport Management, 11(2), pp. 59–68,
2005.
[8] Proussaloglou, K. & Koppelman, F.S., Air carrier demand: An analysis of
market share determinants. Transportation, 22(4), pp. 371–388, 1995.
 THEORY AND PRACTICE IN MODELLING AIR TRAVEL CHOICE BEHAVIOUR 125

[9] Chin, A.T.H. Impact of frequent flyer programs on the demand for air
travel. Journal of Air Transportation, 7(2), pp. 53–86, 2002.
[10] Monteiro, A.B.F. & Hansen, M., Improvements to airport ground access
and behavior of multiple airport system: BART extension to San Francisco
International Airport. Transportation Research Record, 1562, pp. 38–47,
1997.
[11] Psaraki, V. & Abacoumkin, C., Access mode choice for relocated airports:
The new Athens International Airport. Journal of Air Transport
Management, 8(2), pp. 89–98, 2002.
[12] Bondzio, L., Modelle für den Zugang von Passagieren zu Flughäfen
(Models for the Passengers’ Access to Airports), PhD thesis, Ruhr-
University, Bochum, 1996.
[13] Hess, S. & Polak, J.W., Airport, airline and access mode choice in the san
francisco bay area. Papers in Regional Science, 85(4), pp. 543–567, 2006.
[14] Hess, S., Posterior analysis of random taste coefficients in air travel choice
behaviour modelling. Journal of Air Transport Management, 13(4),
pp. 203–212, 2007.
[15] Hess, S. & Polak, J.W., Exploring the potential for cross-nesting structures
in airport-choice analysis: A case-study of the Greater London area.
Transportation Research Part E: Logistics and Transportation Review,
42(2), pp. 63–81, 2006.
[16] Adler, T., Falzarano, C.S. & Spitz, G., Modeling Service Trade-offs in Air
Itinerary Choices. Paper presented at the 84th annual meeting of the
Transportation Research Board, Washington, DC, 2005.
[17] Hess, S., Adler, T. & Polak, J.W., Modelling airport and airline choice
behavior with stated-preference survey data. Transportation Research Part
E: Logistics and Transportation Review, 43, pp. 221–233, 2007.
[18] Louviere, J.J., Hensher, D.A. & Swait, J., Stated Choice Models: Analysis
and Application, Cambridge: Cambridge University Press, 2000.
[19] Hess, S., Treatment of reference alternatives in stated choice surveys for air
travel choice behaviour. Journal of Air Transport Management, 14(5), pp.
275–279, 2008.
[20] Algers, S. & Beser, M., Modelling choice of flight and booking class: A
study using stated preference and revealed preference data. International
Journal of Services Technology and Management, 2(1–2), pp. 28–45, 2001.
[21] Train, K., Discrete Choice Methods with Simulation, Cambridge, MA:
Cambridge University Press, 2003.
[22] McFadden, D., Modelling the choice of residential location (Chapter 25).
Spatial Interaction Theory and Planning Models, eds. A. Karlqvist, L.
Lundqvist, F. Snickars & J. W. Weibull, Amsterdam: North Holland, pp.
75–96, 1978.
[23] Walker, J., Extended Discrete Choice Models: Integrated Framework,
Flexible Error Structures, and Latent Variables, PhD thesis, MIT, Cambridge,
MA, 2001.
[24] Hess, S., Bierlaire, M. & Polak, J.W., Capturing taste heterogeneity and
correlation structure with Mixed GEV models (Chapter 4). Applications
126 DEVELOPMENT OF REGIONAL AIRPORTS

of Simulation Methods in Environmental and Resource Economics, eds.

R. Scarpa & A. Alberini, Dordrecht, The Netherlands: Springer,
pp. 55–76, 2005.
[25] Civil Aviation Authority (CAA), UK airports Passenger survey, CAP 677:
Passengers at Birmingham, Gatwick, Heathrow, London City, Luton,
Manchester and Stansted in 1996, The Civil Aviation Authority, London,
1996.
[26] ONS, International Passenger Survey, London: The Office for National
Statistics, 1996.
[27] OAG, World Airways Guide Fares Supplement 1996, Dunstable, UK: Reed
Elsevier, 1996.
[28] Scott Wilson Kirkpatrick, National Airport Accessibility Model: Model
Status Report (Prepared for the Department for Transport), Basingstoke:
Scott Wilson Kirkpatrick & Co Ltd., 1999.
6

Practical airport demand forecasting with

capacity constraint: methodology and
application
E. P. Kroes
Den Haag and VU University of Amsterdam, The Netherlands

Abstract
Airports require demand forecasts for operational, tactical and strategic purposes.
This chapter provides a description of a practical, operational methodology that has
been developed to provide such demand forecasts, for horizons ranging from short
term (next year) to long term (20 years ahead or more); and for regional airports as
well as major international airports. The model system consists of three elements:
a demand model, a supply model and an iterative procedure to take account of
capacity constraints. The demand model uses three main components: a detailed
data base describing existing passenger flows between air zones worldwide, a
growth model describing expected relative increases in passenger flows and a
discrete choice based competition model allocating passenger flows to different
airports, airlines, air routes and alternative modes of transport. The supply model is
essentially a three-dimensional cross-table of aircraft movements, distinguishing
movements by size, technology class and time-of-day of departure. Starting from a
detailed data base of existing aircraft movements, a simple aging method combined
with incremental log-linear modelling provides estimates of future aircraft
movements. A shadow-price mechanism is used to adjust demand and supply
levels to fit within the limits of available runway and environmental capacity, if
necessary. We have also described two applications of the methodology to
Amsterdam Airport, one addressing long-term airport capacity issues and the other
exploring the expected impact of an air flight tax for the Netherlands.

Keywords: air demand forecasting; airport competition; airport capacity constraint;

time-series models; random utility models.
128 DEVELOPMENT OF REGIONAL AIRPORTS

1 Introduction
A few decades ago air transport analysts have been using trend extrapolation (a
simple form of time-series analysis) to obtain estimates of demand in the near
future. Since then more advanced methods of time-series analysis have been
applied, using causal factors such as economic development, as the key drivers
of growth. The implicit assumption was still that the market shares of the airports
under consideration would remain constant, even in the longer term. Such an
assumption might be justified in a largely regulated environment.
However, in more recent years new dynamics and constraints have entered
the system. Liberalization has led to more competition between airlines and
airports. This dramatically increased competition between airports, airlines and
alliances on the one hand, and led to serious airport capacity issues on the other,
which made extrapolations of historic demand no longer adequate. Airport
demand forecasts now need to take account of the many competitive elements
and the physical and environmental constraints in addition to standard growth
scenarios.
As a consequence discrete choice models have found there way to explain
how choices of (potential) air passengers, reacting to a multitude of competing
offers, affect airport traffic flows. Chapter 4 has given an overview of the various
demand forecasting methods, while Chapter 5 has given an extensive description
of the most advanced methods of airport choice modelling.
In this chapter, we have a somewhat more practical focus: here we describe
how we use a combination of the different available methods to arrive at
estimates of expected future passenger volumes for airports, whether regional or
main airport. In order to do that we represent decisions by potential air
passengers (the demand side: passenger numbers) and airlines (the supply side:
aircraft movements) and their interaction, subject to external capacity constraints
(physical, environmental); in other words an equilibrium model approach.
In this paper, we provide a brief description of the AEOLUS model and its
main components. Then we report two applications of the model for Amsterdam
airport: first to assess capacity issues and a range of policy measures for the
planning horizons 2020 and 2040, and second to estimate the short and medium
term impact of alternative flight tax measures in the Netherlands.

2 Approach
The formal objective of the air demand forecasting model described here was to
develop a practical, operational tool capable of providing air demand forecasts at
airport level, for planning and policy evaluation purposes. In addition to this
fairly general objective a large number of requirements were specified, including
the following (in random order):
● the model had to be strategic in nature and suitable for quick policy
evaluation;
● the model had to be reasonably quick in terms of computing time and
pragmatic in application;
 PRACTICAL AIRPORT DEMAND FORECASTING WITH CAPACITY CONSTRAINT 129

● the methodology had to be intuitive and transparent in its operation;

● the model had to take into account the competition between airports and
airlines (including low-cost airlines and alliances) in North-West Europe;
● it had to take into account the landside accessibility of the airports under
consideration;
● the model had to include competition with surface transport modes including
high-speed rail where relevant;
● it had to take into account the effects of both airport capacity and noise
constraints;
● the method should be suitable for assessing the implications of a range of
policy measures (such as levies for specific market segments);
● the methodology should be capable of assessing the welfare effects of the
capacity constraints and policy measures.

The combination of the objective and the many requirements has lead to a fairly
simple modelling approach; it would have been impossible to include for
instance state-of-the-art demand modelling techniques as described in Chapter 4
in the system, as the run-times would have become excessive. On the other hand
the model system is unusually comprehensive, in that it includes also explicit
supply-side modelling, and pragmatic, in that it uses whatever data is available
and generates the missing detail where necessary.
The model considers worldwide traffic flows to, from and through the
airports under consideration. The architecture of the simulation system consists
of two modules: a module to forecast traveller choices (e.g. which airports to
depart from? Which air route?) and a module to forecast airline choices (which
mix of aircraft to use?). The traveller choice module requires current passenger
counts and level-of-service data for calculating travellers’ preferences for the
available alternatives in the current base year (in this example 2003). The airline
choice module computes numbers of yearly flights per type of aircraft and per
period of the day, also for the base year (in this example 2003; see Figure 1).
The observed numbers of passengers in the base year are extrapolated
towards the forecast year (for instance 2020, in the example) using origin-
destination (OD) specific growth factors that depend on expected economic and
price developments between each OD pair. This is essentially a simple time-
series model.
The distribution over departure airports, air routes etc. is calculated again in
the travellers’ choice module, this time using the level-of-service anticipated for
the forecast year. And the airline choice module is also calculated for the forecast
year. The traveller choice module is connected to the airline choice module by an
iterative loop, to meet any capacity constraints that may arise in the forecast
year.
130 DEVELOPMENT OF REGIONAL AIRPORTS

Figure 1: Basic structure of the AEOLUS model.

3 Demand forecasts
The demand forecasting method within the AEOLUS model contains three
elements:
1. An observed base year passenger flow data base.
2. A growth factor model.
3. A traveller choice module.
The logic is as follows:
1. Starting point is the existing observed base year pattern of passenger flows:
the numbers of passengers travelling from each origin to each destination
through all airports and modes. No attempt is made to model this base year
situation, to explain the frequency of air travel or the OD pattern
(distribution). By working this way the model is by definition consistent with
the base year statistics, and the complex interactions that exist within the base
year demand profile are retained without explicitly trying to capture them.
2. Growth factors are then used to extrapolate the base year OD pattern to new
year t volumes. This is done by using demand elasticities for key drivers of air
travel volume, such as GDP. The elasticities are applied using OD specific
developments in the drivers, and separately for different journey purposes. By
combining 1 and 2 a new modified OD matrix is obtained for the target year.
3. The traveller choice module reproduces the decisions of the air passengers,
and in particular how those affect the market shares of different airports under
consideration. The module is applied both for the base year and for the
forecast horizon, to obtain a factor that indicates the increase or decrease in
market share. This module simulates the expected change in competitive
 PRACTICAL AIRPORT DEMAND FORECASTING WITH CAPACITY CONSTRAINT 131

position of the airports under consideration, as a function of changes in the air

network (destinations, frequencies, alliances, hub-structure), air fares and also
surface accessibility.
All three elements are used in combination to obtain the ultimate demand
forecasts:

Vijpt = Vijpb Factijp ( Pijpta / Pijpba ) ,

where: Vijpt = volume of air passengers in year t;

i = origin zone;
j = destination zone;
p = purpose;
Vijpb = volume of air passengers in base year b;
Factijp = growth factor of passenger volume from year b to year t;
Pijpta = proportion of air passengers using airport a in year t;
Pijpba = proportion of air passengers using airport a in base year b.

Each of these elements is described in some more detail in the subsequent

sections.

3.1 Traveller choice module

The traveller choice module simulates the number of long-distance trips that
travellers make between all airports worldwide, which are represented in origin
and destination zones. And in particular, it calculates how these trips are
distributed over the wide range of available choice alternatives.
The total number of zones worldwide is typically some 100, with the
geographic coverage of the origin and destination zones being relatively small
within the catchment area of airport(s) under investigation. The zones become
larger for destinations within the same continent when distances to the airport
increase, and the zones are very large for intercontinental trips.
For trips with an origin (or destination) within the catchment area of the
airport(s) under consideration, the model forecasts the market shares for each of
the possible departure (or arrival) airports in this region (see Figure 2 for an
example of the airports in the catchment area of Amsterdam airport) and the
market shares of the modes used to access (or egress from) the airport. For trips
with an origin (or destination) inside the catchment area of the airport and with
a destination (or origin) somewhere else on the same continent, the model
forecasts the distribution over the available main modes as well: car, train (high-
speed) and aircraft. This specific structure reflects air passenger choices among
competing departure and hub airports in north-west Europe in a straightforward
way.
132 DEVELOPMENT OF REGIONAL AIRPORTS

Figure 2: Assumed catchment area of Amsterdam airport.

The market shares of the available combinations of travel alternatives are

determined by simulating traveller choices at up to three levels (see Figure 3):

Figure 3: Traveller choices.

 PRACTICAL AIRPORT DEMAND FORECASTING WITH CAPACITY CONSTRAINT 133

1. The choice between main modes of transport from O to D: car, train (high-
speed) or aircraft.
2. The choice between available air routes, specified by departure airport, airline
or alliance, direct flight or indirect flight via a hub.
3. The choice between access modes to the airport: normally car or train.
Not all choices are modelled for each OD combination, see Table 1.
Table 1: Choices that are modelled for each origin–destination combination.
Origin Destination
Catchment area Rest of Europe Rest of world
Main mode choice;
Route choice;
Catchment Route choice;
(Out of scope) Access mode
area Access mode
choice
choice
Main mode
choice;
Rest of
Route choice; Route choice Route choice
Europe
Egress mode
choice
Route choice;
Rest of
Egress mode Route choice Route choice
world
choice

The AEOLUS model uses random utility models, in this case standard nested
logit [1], to simulate the traveller choices. Travel times, waiting and transfer
times, travel costs, and service frequencies are the main determinants of choice
included in the utility functions. The coefficients have been inferred from a
number of previous studies, and a large number of alternative-specific constants
are calibrated from base year air passenger statistics for airports in the catchment
area.

3.1.1 Access mode choice

Two alternatives are typically included here: car and train (if available, or
coach). Generalized costs for the car mode are determined by fuel cost, parking
cost and travel time. Travel times are converted into generalized cost by means
of multiplication by a value-of-time depending on the travel purpose (business or
non-business). Generalized costs for the train mode are determined by the train
fare and generalized train travel time. Travel fares and times are taken from an
input file with surface access level-of-service information for all departure
airports in the area under consideration.
The same model is used to model the egress mode in case the destination of
the trip is within the catchment area (backward journey).

3.1.2 Route choice

The choice alternatives are defined by airline (e.g. Skyteam, Star Alliance,
OneWorld, low-cost airlines, other airlines), by hub (direct flight, or one of the
134 DEVELOPMENT OF REGIONAL AIRPORTS

64 international hubs considered) and by access/egress airport (only if origin or

destination is within the catchment area). The utility of each alternative is
determined by the logarithm of the number of flights per week, by a generalized
cost term (determined by an assumed ticket fare and flight time (with an extra
penalty for an indirect flight)) and by an accessibility term for the airport (only in
catchment area). This accessibility term is the logsum of the access mode choice
model. Air level of service information including destinations served, flight
frequencies, fares, flight times, interconnections, etc. are taken from an input file.
This has been prepared using Official Airline Guide (OAG) [2] flight data and a
fares model.

3.1.3 Main mode choice

There are three choice alternatives here: car, train (high-speed) and aircraft. The
utilities for the first two modes are determined by travel cost (fuel or train fare) and
generalized travel time; the utility of the air alternative is determined by the logsum
of the route choice model. Level of service for car and (high-speed) rail is obtained
from international surface transport network data.

3.1.4 Air Freight model

In addition to the passenger demand model, a small air freight demand model can
also be used. In this model both the volume of air freight and its distribution
among alternative airlines and full freighters and belly-freight can be simulated.

3.2 Observed base year OD matrix

Another key component of the demand model system is the observed base year
pattern of air trips. This is derived from a combination of different sources:
● passenger surveys on airports (e.g. the ‘continuous survey’ carried out at
Amsterdam airport);
● detailed passenger statistics (annual volumes by destination) published by
airports;
● other international sources of information on surface transport.
In the ideal case complete and accurate information for all OD flows should be
available, for all airports, airlines/alliances, air routes, surface transport modes,
etc. In practice this is often not available, or only in the form of aggregate
information. Our pragmatic solution for this has been that we have used detailed
passenger survey data for a single airport in combination with the traveller
choice model described above, to estimate the missing OD information for the
other airports.
In short the procedure works as follows. The unobserved volume for an
airport k is calculated by multiplying the observed volume l by the ratio between
the simulated market share of alternative k and l:
Vunobserved (k ) = (Market share (k ) / Markets share (l ))Vobserved (l ).
If the simulated market shares for airport l are not too close to 0 this generally
gives credible results. This method is described in some more detail in [3].
 PRACTICAL AIRPORT DEMAND FORECASTING WITH CAPACITY CONSTRAINT 135

In practice, obtaining a reasonable accurate ‘observed’ base year OD matrix

is often a substantial activity, which requires a major effort. And the result will
inevitably be subject to some error. But when one makes sure that all available
information is taken into account, that at least a fairly accurate source of OD
information for one important airport is included, and that the marginal
distributions for all airports are checked against the published statistics (and
corrected where necessary), this forms a reasonable starting point for the
modelling procedure.

3.3 Growth factor model

The growth factor model provides factors which are applied to the observed base
year passenger volumes to obtain the future year OD matrix. The specification is
simple:
Factijp = ( Driverdtijp / Driverdbijp ) Elast ijp ,

where: Elastijp = Demand elasticity for Driver d, OD pair i-j and purpose p;
Drivertdijp = Driver of demand growth d, OD pair i-j and purpose p.
This procedure is applied on a year-by-year basis, for as many years as are
necessary. We use the following drivers of growth:
● population size for purpose leisure;
● GDP for purpose leisure;
● trade volume for purpose business;
● price for purposes leisure and business.
For OD relations, we take the average of the drivers’ values for origin and
destination zone.
Because the values of driver growth are often different for different OD pairs
(e.g. expected future GDP growth for Asian zones is much higher than for
European zones) the structure of the OD matrix is modified by the application of
this growth-factor procedure.

4 Supply forecasts
Often a substantial growth of air traffic is predicted. The resulting numbers of
aircraft movements in the coming 10 to 20 years often exceed the current runway
capacity. Furthermore, the amount of noise generated by aircraft may exceed
existing legal boundaries. And the same may hold for other environmental
emissions. To take these effects into account, we developed an airline choice
module that simulates the deployment of a mix of different aircraft to transport
the passenger volume as predicted by the traveller module. This module
distinguishes three dimensions:
1. the size of the aircraft (nine classes);
2. the technological status of the aircraft (five classes);
3. the time-period of departure/arrival (four periods per day).
136 DEVELOPMENT OF REGIONAL AIRPORTS

This results in 180 possible combinations.

We have used observed base year distributions, and foreseeable trends to
predict the future distribution of aircraft over these 180 combinations. One such
foreseeable trend is the future renewal of the aircraft fleet, based upon a simple
aging model. From this distribution, we infer the implicit preference utility
values for each of the combinations (the table is seen as a log-linear model,
where each cell has an associated utility which can be inferred from its share).
When for instance airport charges are introduced, these utilities are modified
(costs per seat are computed and converted into utility values, and added to the
base utility using an assumed cost coefficient). This type of application is similar
to the well-known incremental logit-modelling approach, described for instance
in [1]. After modification of the utilities new shares can be computed, and new
distributions over the possible combinations can be determined (Figure 4). This
enables us to simulate how autonomous developments (through aging of the
fleet), capacity constraints (through shadow cost) or policies (through actual cost
increases) modify the distributions of aircraft used.
An estimate of the total number of aircraft movements (per year and per
period of the day) and the total environmental burden (i.e. the amount of noise
generated by the departing and arriving aircraft) can be calculated using this
airline choice model.

Figure 4: Structure of the airline choices module.

4.1 Forecasting demand and supply

For the base year the passenger choice and the airline choice module are run
once to calculate a base scenario. The output values for the number of
passengers, the number of flights and the volume of noise produced are
calibrated using correction factors to match the observed values in the base year.
Typically only small corrections are necessary.
For the forecast year we specify:
 PRACTICAL AIRPORT DEMAND FORECASTING WITH CAPACITY CONSTRAINT 137

● the expected changes in air level-of-service (increase of flight frequencies,

change in air fares, flight times typically remain constant);
● the expected changes in level-of-service of the land modes (fuel cost, train
fares, travel times);
● the expected change in value-of-time (due to real increase of incomes);
● the expected changes in the airlines’ preferences for the deployment of
aircrafts of certain sizes (due to the availability of larger aircraft);
● the expected changes in the airlines’ preferences for the deployment of
aircrafts of certain technology (due to the availability of newer and more quite
aircraft). For this we use a simple fleet aging and replacement model.
The number of travellers in the forecast year travelling between an origin and a
destination zone are determined by applying a growth factor to the number of
travellers in the base year. For non-business travellers this growth factor depends
on population growth in the origin zone, real GDP per capita growth in the origin
zone and the price growth in both origin and destination zones. For business
travellers, this growth factor depends on trade growth between the origin and
destination zone and price growth. Price elasticities are within the ranges
indicated by Brons et al. [4].
The passenger choice module is then run again to determine the market shares
of the available alternatives in the forecast year. Consecutively, the airline choice
module is run again to calculate the number of aircraft movements and the amount
of generated noise in the forecast year.

5 Capacity constraint
If demand in the forecast year exceeds supply, during any of the time periods
considered, a capacity constraint procedure is necessary in order to establish a
consistent equilibrium solution. In the model system this works as follows. If the
total number of aircraft movements exceeds either the physical (runway)
capacity or the legal environmental noise limit, an iterative procedure is started
(Figure 5). In each iteration the passenger airfares are increased by a shadow
cost, so that demand is reduced and airlines that fly with larger aircraft and/or
from airports with less severe capacity constraints become relatively more
attractive. In parallel, charges for the airlines stimulate the use of larger and more
modern (i.e. less noisy) aircraft. This iterative procedure is repeated until the
demand (passengers converted into aircraft movements) can be accommodated
within the capacity limits.
The user of the AEOLUS model can choose between two options for the way
the shadow costs are allocated: slot allocation based upon slot trading, or a
system based on grandfathering rights. The first option allocates a charge to each
aircraft movement, independent of the airline. Since these costs are partly
transferred to the passengers by increased air fares, the final distribution of slots
will favour those airlines (and those passengers) that have the highest willingness
to pay for such a slot. This simulates a free slot trading system where airlines
may win and loose slots.
138 DEVELOPMENT OF REGIONAL AIRPORTS

Figure 5: Iterative procedure.

The second option keeps the number of flights per airline in the base year fixed.
This means that slots that have been allocated to an airline in the past, can not be
transferred to another airline (this is called the system of grandfathering rights).
Any remaining slots that have not been allocated in the base year are distributed
over the airlines proportional to their demand for additional slots. However, a
small number of slots are given to the smaller airlines to simulate the current
policy to stimulate new entrants in the market.
In case of slot trading, the shadow cost is (partly) dependent on the amount of
noise that an aircraft generates in case the noise limitations are exceeded. This
stimulates the choice for newer types of aircraft. In case of a slot allocation
system with grandfathering rights there is no dependency of the scarcity charges
on noise production, and hence no incentive to use newer and less noisy aircraft.
The model also takes the runway capacity limits on competing main airports
into account to prevent unrealistic predicted growth on these airports as a result
of the limited capacity on the main airport(s) under consideration.

6 Case study 1: Amsterdam airport capacity planning

The AEOLUS model has originally been developed and applied for the Dutch
Ministry of Transport, Public Works and Water Management, under supervision
of aviation experts from airports, airlines and the national economic research
centre. The case study reported here investigated what possible capacity
problems might arise for Amsterdam airport (Schiphol) for the years 2020 and
2040, under four different macro-economic and technological scenario
assumptions. The welfare effects were evaluated, and a series of 14 different
policy measures was investigated to mitigate the adverse societal effects.
The results of this study were used as an input for the scenario policy
assessment that the Ministry of Transport, Public Works and Water management
together with other ministries completed on the future of Amsterdam Schiphol
 PRACTICAL AIRPORT DEMAND FORECASTING WITH CAPACITY CONSTRAINT 139

airport. This assessment was a key input to the new Dutch government policy
decision concerning the future of the airport.
The scenario assessment is based on macro-economic scenarios for the
Netherlands that were developed by the Netherlands bureau for economic policy
analysis [5]. The implications of these four scenarios for air travel through
Amsterdam airport are summarized in Figure 6.

Figure 6: Four futures for air travel in 2020.

The two dimensions in Figure 6 represent possible orientations of economic

development for Europe and the Netherlands:
● Emphasis on private responsibilities or on public responsibilities?
● Emphasis on international cooperation or national sovereignty?
Figure 7 shows the predicted numbers of flights in 2020 for all four scenarios for
each of the three assumptions:
● no runway or noise restrictions, which represents the unconstrained potential
demand for air travel on Amsterdam airport;
● the current policy scenario, with existing runway and noise restrictions, and
with a slot allocation system based on grandfathering rights;
● an alternative capacity constrained scenario, with a new slot allocation system
based on slot trading.
For the two high economic growth scenarios (Global Economy and Transatlantic
Markets) the potential demand in 2020 exceeds the existing capacity constraints,
in particular the noise limits. The slot trading system is clearly more efficient in
that it is able to accommodate more flights within the existing noise capacity
constraints. This is mainly due to the fact that this system has noise-generation
dependent scarcity charges that stimulate the use of new and more quite aircraft.
140 DEVELOPMENT OF REGIONAL AIRPORTS

Figure 7: Model forecast of the number of flights per year in 2020 for the four scenarios
for three cases (no capacity constraints, slot allocation based on grandfathering
rights and slot allocation based on free trading).

Other charging schemes and policies that have been simulated are reported
in [6].

7 Case study 2: The impact of air flight tax in the Netherlands

In 2007 Dutch Government decided to levy a tax on air flights to and from all
airports in the Netherlands. The objective was to generate 350 million Euro per
year. But the exact implementation was still under discussion: it was not yet
specified which travellers would pay and which would be exempt from this tax.
The Dutch Ministry of Finance commissioned a study to investigate the expected
effects, and to decide about the preferred specification of the tax. The AEOLUS
model was used for this. In this section, we discuss five of these alternative
implementations.
Sixteen alternative versions of the ticket tax have been studied. These
versions differed in the amount of tax that each of the segments (departing
passengers, transferring passengers, freight) had to pay. In all versions, the total
amount of tax collected per year was 350 million Euro. In the remaining of
this section, we discuss five of these versions. Since this chapter concentrates on
passenger choices, we have only selected versions with no tax on freight
(Table 2). The names of the versions correspond to the names in the original
report [7].
The AEOLUS model simulates the effects of the ticket tax in 2011 by
increasing the fare of air travel starting from the year of introduction of the tax,
2008. Four macro-economic scenarios were simulated, the same as the ones
mentioned in the previous section.
 PRACTICAL AIRPORT DEMAND FORECASTING WITH CAPACITY CONSTRAINT 141

Table 2: Effects of ticket tax (introduced in 2008) for the year 2011.
Version 1 Version 1E Version 1E- Version 2 Version
B 2E
Tax per departure
European € 23.00 € 16.67 € 12.50 € 13.75 € 9.50
destinations
Intercont. € 23.00 € 37.50 € 47.50 € 13.75 € 21.38
destinations
Tax per transfer
Europe–Europe – – – € 13.75 € 9.50
Europe–ICA – – – € 13.75 € 15.44
ICA–ICA – – – € 13.75 € 21.38
Amsterdam
Total passengers –10 to –8 to –11% –8 to –10% –19 to –20 to
–12% –22% –26%
Dep. total –13 to –11 to –10 to about – about –9%
–14% –12% –11% 10%
Dep. Europe –15 to about –12% –9 to –10% –11 to about –9%
–16% –12%
Dep. interc. –8 to –9% –11 to –14 to –6 to –9 to
–14% –18% –7% –10%
Transferring –5 to –8% –5 to –7% –4 to –8% –37 to –44 to
–39% –48%
Total flights –9 to –17 to –17 to
–12% –8 to –9% –8 to –9% –20% –23%
Regional airports
Dep. passengers –18 to –14 to –11 to –13 to –9 to
–20% –16% –13% –15% –12%
Emissions (Amsterdam)
Noise (dBA) about –0.3 –0.2 to –0.3 –0.2 to –0.7 to –0.9 to
–0.3 –0.8 –1.0
Particles –5 to –5 to –9% –3 to –9% –14 to –17 to
–10% –19% –23%
For the discussion of the effects of the ticket tax, we distinguish between
departing and transferring passengers. Note that a transfer passenger has to pay
the tax twice per journey, since he makes a transfer both during the outward trip
and the return trip. Arriving passengers do not pay a tax. However, since most
passengers buy a round-trip ticket, we assume that half of the tax applies to the
outward journey and half of it applies to the return journey. Therefore, the effects
on arriving passengers are in the model identical to the effects on departing
passengers.
142 DEVELOPMENT OF REGIONAL AIRPORTS

Version 1: Tax on departing passengers only

In this version, each passenger departing from a Dutch airport (except transfer
passengers) has to pay a tax of € 23. As a result, less travellers will use a Dutch
airport as their departure airport. The number of departing passengers at Schiphol
airport decreases by 10–12% in 2011 (depending on the macro-economic
scenario, see Table 2). As a result, the number of flights will be reduced. This
affects transfer passengers since they have fewer options to travel via
Amsterdam. This results in a decrease of transfer passengers of 5–8%.
For European destinations the relative increase in air fare is larger than for
intercontinental destinations. Hence, the decrease of the number of travellers that
depart from Amsterdam to a European destination is larger than for an
intercontinental destination (15–16% vs. 8–9%). Since the regional airports offer
mainly European destinations, and they lack the segment of transfer passengers
that do not have to pay a tax, regional airports are stronger affected than
Amsterdam airport (decrease of total number of passengers of 18–20% for
regional airports vs. a decrease of 10–12% for Amsterdam).

Version 1E: Differentiation between European and intercontinental

destinations

In this version of the ticket tax departing passengers with a European destination
have to pay a tax of € 16.67, while passengers with an intercontinental
destination have to pay a tax of € 37.50. As a result, the decrease of the European
market at Amsterdam is similar to the decrease of the intercontinental market
(about 12%).

Version 1E-B: Further differentiation between European and intercontinental

destinations

In this version of the ticket tax departing passengers with a European destination
have to pay a tax of € 12.50, while passengers with an intercontinental
destination have to pay a tax of € 47.50. As a result, the decrease of the European
market at Amsterdam is less than the decrease of the intercontinental market (9–
10% vs. 14–18%). Regional airports are less affected than in versions 1 and 1E:
the decrease of the total number of passengers for regional airports is about the
same as for Amsterdam airport (11–13% for regional airports vs.8–10% for
Amsterdam).

Version 2: Tax on departing and transferring passengers

Transfer passengers pay the same amount of tax (per transfer) as departing
passengers. In order to raise 350 million Euro per year, the tax level is set at
€ 13.75. This results in a very strong decrease in the number of transfer
passengers (37–39%). This is due to the fact that these passengers have to pay
the tax twice per round journey, since they will make a transfer both during the
outward and the return trip. Furthermore, these passengers have a large number
 PRACTICAL AIRPORT DEMAND FORECASTING WITH CAPACITY CONSTRAINT 143

of good alternatives, because most of them can also choose to make a transfer at
London Heathrow, Frankfurt or Paris Charles de Gaulle without paying extra tax
or having to make a detour.

Version 2E: Differentiation between European and intercontinental

destinations

This version is similar as version 2, but the amount of tax depends on the
destination (tax for intercontinental destinations is about 2.25 times as high as for
Europeans destinations). This has an even larger effect on transferring
passengers (that are predominantly passengers with an intercontinental origin or
destination). The decrease of the total number of passengers at Amsterdam
airport is 20–26%, while the decrease at regional airports is limited to 9–12%.

Effects of ticket tax on departing passengers

In all versions presented, the number of passengers departing from Dutch

airports will decrease. Instead, they will:
● depart from a foreign airport, where they do not have to pay the ticket tax;
● travel using another mode (either train or car). This is only an alternative for
travel within Europe;
● decide not to travel at all.
Figure 8 displays the number of passengers (absolute number of passengers, as
determined after averaging over the four macro-economic scenarios), who
change their travel behaviour for each version of the ticket tax (departure and
arriving together). The total number of passengers that no longer depart
from/arrive at a Dutch airport (either Amsterdam, or at one of the regional
airports) is equal to the number of passengers that either shift their
departure/arrival to a foreign airport (about 45%), or shift to a different mode
(about 10%) or no longer travel (about 45%).

Figure 8: Effect of ticket tax on passengers departing/arriving at Dutch airports (2011).

144 DEVELOPMENT OF REGIONAL AIRPORTS

Effects of ticket tax on transfer passengers

Travellers that stop transferring at Amsterdam airport as a result of the

introduction of the ticket tax will instead:
● transfer at another airport;
● shift to a direct flight;
● decide not to travel at all.
Figure 9 displays the effect on the number of transfer passengers at Amsterdam,
Frankfurt and Paris Charles de Gaulle. In versions 1, 1E and 1E-B, the number of
transfer passengers at Frankfurt and Charles de Gaulle reduces as well. These are
travellers that would have departed from a Dutch airport and would have
transferred at FRA or CDG if there would not have been a ticket tax. In versions 2
and 2E, the number of transfer passengers at Frankfurt and Charles de Gaulle
increases. These are travellers that have diverted their route due to the ticket tax at
Amsterdam airport.

Figure 9: Effect of ticket tax on passengers transferring at Amsterdam airport (2011).

Final implementation

In order to mitigate the effects of the ticket tax on the airlines and airports
(particularly on the regional airports), the Dutch government decided to
implement a version that is very similar to version 1E-B. It was decided that the
tax would be € 11.25 for all destinations within 2500 kilometres, including all
EU member countries) and € 45 for other destinations. An exception is made for
countries with destinations on both sides of the 2500 kilometres border. The low
tax of € 11.25 also applies for other destinations in those countries, provided that
they are not further away than 3500 kilometres.
This tax was implemented in the Netherlands on 1 July 2008, and its effects
were quickly noticeable. The number of passengers departing from foreign
airports increased substantially according to travel agencies. One of these
agencies (D-reizen) reported an increase of 350% for their customers [8].
Amsterdam airport announced that their passenger growth would stagnate [9].
 PRACTICAL AIRPORT DEMAND FORECASTING WITH CAPACITY CONSTRAINT 145

KLM expected to loose half a million to a million passengers in 2008 [10]. These
effects were reinforced by the impact of the economic crisis that followed in
2008. Early 2009 the Dutch government decided to abandon the tax from 1 July
2009 onwards.

8 Conclusion
In this chapter, we have described a practical demand forecasting model that has
been developed for strategic planning purposes. The model has been applied in
several studies in the Netherlands, two of which have been described. But, the
concept is generic and can easily be applied to simulate demand at airports
elsewhere; for instance in other countries, both for main and regional airports.
As an example, the model has been successfully implemented in 2007 for
forecasting air passenger demand in France, in particular for the various airports
in the Paris region.
Having discussed some of the main features and capabilities of the AEOLUS
model system, it is also useful at this point to discuss some of its limitations.

1. First we would like to emphasize that the AEOLUS model presented here is a
simple, pragmatic forecasting model that uses fairly straightforward methods.
It is far from the state-of-the-art methodologies that are described for instance
by [11]. That is related to its key objective, and to the requirements of
transparency and intuitivity of operation.
2. Another limitation is the fact that the coefficients of several of the models
have not been formally calibrated, but were ‘imported’ from other similar
studies. To our defence, we can say that we have, of course, extensively tested
the response characteristics of the model, and the resulting demand
elasticities. Also our model has been audited by the Netherlands Bureau
for Economic Policy Analysis, with favourable outcome. But empirically
calibrated coefficients would still add further credibility.
3. Then the supply model, which is a very simple, again highly pragmatic
heuristic that has no solid foundation in economic theory. Particularly the way
in which the observed multidimensional distribution of aircrafts is modified in
response to cost increases would benefit from further work, both in terms of
methodology and use of marginal cost functions.
4. The equilibrium procedure: we use a heuristic iterative procedure that adjusts
(shadow) costs and passes these on to demand (generalized cost of air
passengers) and supply (marginal cost for airlines) until demand and supply
are more or less in equilibrium. This problem could be re-specified as a
multidimensional optimization problem that could be solved by means of
dedicated solvers. In fact we have now implemented a new version of
AEOLUS using the General Algebraic Modelling System (GAMS) [12]
package.

A final issue that we want to raise is the fact that this type of model is quite
demanding in terms of availability of data. Firstly, it requires extensive
passenger survey information (including detail about trip origin, destination,
146 DEVELOPMENT OF REGIONAL AIRPORTS

journey purpose, air route and socio-economic information) for at least one
important airport under consideration, and ideally more. But also it needs
detailed airside level of service information, which can be derived from the OAG
database [2] and airport statistics. And landside level of service, which can be
derived from surface transport networks for road and rail. In return, however, it
also provides a mass of information, thus enabling a detailed assessment of
future developments and possible impacts of policy measures.

Acknowledgements
The development, implementation and application of the AEOLUS model
system has been commissioned by the Dutch Ministry of Transport, Directorate
General of Transport and Aviation. Also I want to acknowledge the substantial
contributions to the development of the AEOLUS model system that have been
provided by Marco Kouwenhoven (Significance) and Jan Veldhuis (SEO).

References
[1] Ben-Akiva, M. & Lerman, S.R., Discrete Choice Analysis, Theory and
Application to Travel Demand, The MIT Press: Cambridge, MA, 1985.
[2] See www.oag.com.
[3] Kroes, E., Lierens, A. & Kouwenhoven, M., The Airport Network and
Catchment area Competition Model ERSA Conference Series, 2005.
Available via http://www.ersa.org/ersaconfs/ersa05/papers/521.pdf
[4] Brons, M., Pels, E., Nijkamp, P. & Rietveld, P., Price Elasticities of
Demand for Passenger Air Travel, Tinbergen Institute Discussion Papers
01-047/3, Tinbergen Institute, 2001.
[5] Mooij, R. de & Tang, P. Four futures of Europe CPB special publication nr.
49. Available via http://www.cpb.nl/nl/pub/cpbreeksen/bijzonder/49/
download.html
[6] SEO and RAND Europe, Ontwikkeling Schiphol 2020–2040 bij
ongewijzigd beleid – Eindrapport, in opdracht van het ministerie van
Verkeer en waterstaat, Directoraat Generaal Transport en luchtvaart (Dutch
version only), 2006.
[7] Significance, SEO Economisch Onderzoek, Effecten van verschillende
heffingsvarianten op de Nederlandse Luchtvaart, Significance report
07014, 2007 (Dutch version only). Available via http://www.
significance.nl/reports/2007-MINFIN-07014.pdf.
[8] NRC, Schiphol verwacht stagnatie door vliegbelasting, news paper article
published on 4 January 2008 (Dutch version only). Available via
http://www.nrc.nl/economie/article881801.ece/Schiphol_verwacht_stagnati
e_door_vliegbelasting
[9] Volkskrant, Duur Schiphol verkeert in crisis, news paper article published
on 3 July 2008 (Dutch version only). Available via www.volkskrant.nl/
economie/article1039362.ece/Duur_Schiphol_verkeert_in_crisis
 PRACTICAL AIRPORT DEMAND FORECASTING WITH CAPACITY CONSTRAINT 147

[10] Volkskrant, Vliegtaks kost KLM ‘miljoen passagiers’, newspaper article

published on 17 July 2008 (Dutch version only). Available via www.
volkskrant.nl/economie/article1043932.ece/Vliegtaks_kost_KLM_miljoen_
passagiers.
[11] Hess, S. & Polak, J.W., Mixed logit modelling of airport choice in multi-
airport regions. Journal of Air Transport Management, 11(2), pp. 59–68,
2005.
[12] See www.gams.com.
This page intentionally left blank
7

Regional airports and opportunities for

low-cost carriers in Australia
A. Collins, D. A. Hensher & Z. Li
The University of Sydney, Australia

Abstract
Australia is vitally dependent on aviation services for delivering passenger
accessibility to many rural and remote locations. The majority of airports in
Australia are regional airports. There are real opportunities for a number of
regional airports to improve their services for the region through the introduction
of low-cost carriers (LCCs). The aim of this paper is to investigate this potential,
through a formal model system of the entire aviation network in Australia,
focusing on identifying influences on passenger demand and flights offered, and
the role of air fares and number of competitors on each route.

Keywords: regional airports; low-cost carriers; regular passenger transport;

structural equation system; three stage least squares (3SLS)

1 Introduction
Australia is vitally dependent on aviation services for delivering passenger
accessibility to many rural and remote locations. In 2005–06, over 40.93 billion
passenger kilometres or 11.47% of the total domestic passenger transport task
(including metropolitan travel) was serviced by aviation [1]. Conservatively this
represents over 46% of all intra- and inter-state aircraft movements and 13% of
revenue passenger activity. As the Australian population progressively, albeit
slowly, migrates away from the capital cities along the coast and inland, a
number of towns that were once small centres servicing a hinterland have grown
to become sizeable hubs for substantial regional activity. The role of aviation has
grown in response to the need for improved accessibility to these regional hubs.
Some of these centres already enjoy one or more low-cost carriers (LCCs),
150 DEVELOPMENT OF REGIONAL AIRPORTS

whereas a number of them are yet to benefit from LCCs, often with a single
carrier with relatively high fares and poor service frequency.
There are real opportunities for a number of regional airports to improve their
services for the region through the introduction of LCCs. The aim of this paper is
to investigate this potential, through a formal model system of the entire aviation
network in Australia, focusing on identifying influences on passenger demand
and flights offered, and the role of air fares and number of competitors on each
route.
The chapter is organized as follows. Section 2 provides an overview of airline
activity and regional airports in Australia, with a particular focus on the growth
in LCCs, expansion by regional airports and the interplay between the two.
Section 3 establishes a formal modelling framework through which operational
changes by airlines and airports can be evaluated. Section 4 outlines the data that
was collected for model estimation, and provides some summary statistics using
this data. Section 5 provides results for the base year for the various models.
Section 6 assesses the impact of new LCC entrants on patronage and regional
airport activity. The chapter concludes with a summary of major findings.

2 An overview of airline activity and regional airports in

Australia
2.1 Airline activity

Prior to 1990, the Australian domestic aviation market was a regulated duopoly
on the trunk routes. The two permitted airlines, Ansett Airlines and Australian
Airlines, were similar in their operation and engaged in little competition. Entry
by other airlines was prohibited, capacity constraints were applied by the
government, and fares were determined on a cost-plus basis. The first hint of
liberalization came in 1981 with an amendment of the Airlines Agreement Act
that allowed regional airlines to expand their operations on non-trunk routes and
operate jet aircraft. On 1 November 1990 the entire industry was deregulated.
Restrictions on capacity, fares and entry were lifted, opening the way for new
airlines to enter the market and compete.
Compass Airlines was the first new entrant to the market, commencing
flights just 1 month after economic deregulation. Compass intended to compete
as a low-cost carrier and undercut the bloated costs of the incumbent airlines. Its
fleet comprised of a single aircraft type, the 266 seat A300–600. The network
was simple, linking only seven major airports. Interestingly, this approach
contrasts with later entrants, who have relied on smaller Boeing 737 and A320
aircraft, and formed networks that extend well beyond the major airports
(although admittedly while being significantly more capitalized). Despite having
lower costs than the incumbent airlines, Compass Airlines collapsed barely a
year after commencing operations. Nyathi et al. [2] provide an extensive analysis
of why Compass failed, and consider the implications of undercapitalization,
poor pricing including crude discounting strategies, the lack of a yield
management system, poor marketing and management and a lack of access to
 REGIONAL AIRPORTS AND OPPORTUNITIES FOR LOW-COST CARRIERS IN AUSTRALIA 151

adequate terminal space. The Compass brand was revived in 1992 as Compass
Mk II when a new startup called Southern Cross Airlines decided to trade under
the same name. It too failed, and no LCCs operated for the remainder of the
decade.
Despite the absence of further LCC entrants, Australian domestic aviation did
not remain static in the decade that followed deregulation. In August 1992,
Australian Airlines was purchased by Qantas. Qantas in turn was privatized by
the Australian government in March 1993. Forsyth [3] determined that Qantas
and Ansett increased their total factor productivity during the 1990s, but not to
the levels of equivalent overseas airlines. Hence, there was scope for new
entrants to compete at a lower cost. Ansett however struggled in the more
competitive environment. A lack of capital under the full ownership of Air New
Zealand, few changes to labour arrangements, maintenance problems, high costs
and the entrance of Virgin Blue in August 2000 all placed pressure on the airline
(Forsyth [4]). Ansett went into administration in September 2001 and ceased
operation in March 2002.
Virgin Blue, the first successful Australian LCC, was a major beneficiary of
the Ansett collapse. The sudden decline in domestic seat capacity allowed Virgin
Blue to expand rapidly with a competitive low-cost model. Old Ansett terminal
space was also easily acquired at most airports. A lack of adequate terminal
space had played a key role in the demise of Compass Airlines. The case against
Qantas claiming anti-competitive behaviour disadvantaging Compass in respect
of adequate information at Sydney Airport on the location of the Compass gates,
was won by Compass but only after the airline had ceased operations. Virgin
Blue commenced operations in August 2000 between Brisbane and Sydney. By
April 2003, it was operating 24 aircraft and moving 6.6 million passengers per
year. By the end of 2007, Virgin Blue was operating 53 aircraft and moving 15.3
million passengers per year. In May 2007 Virgin Blue had a domestic market
share of 31.7%, making it the second largest domestic airline in Australia. Virgin
Blue has also proven very profitable, with a profit every year since its formation,
and a 2007 profit of A$215.8 million.
While Virgin Blue is recognized as a LCC, some signs suggest that it lies
somewhere between the LCC and full-service models. The airline has introduced
various services that LCCs have typically eschewed. Airport lounges were
opened in April 2003 and a frequent flier programme commenced in November
2005. A two class configuration was introduced in March 2008, with a premium
economy fare providing extra seat pitch and the use of the middle seat as a table.
These new features suggest that Virgin Blue is attempting to both cater to the
leisure market and compete with Qantas in the high yield business market.
Virgin Blue has expanded beyond the trunk routes linking the capital cities,
and beyond the traditional tourist routes linking the capitals with the larger
coastal tourist destinations such as Cairns and Townsville (see Figure 1). Recent
route additions have previously been served only by regional airlines, including
Sydney-Albury and Sydney-Port Macquarie. To make these thin routes viable,
Virgin Blue is in the process of supplementing its core Boeing 737 fleet with 24
Embraer 170 and 190 regional jets, which carry 76 and 104 passengers,
152 DEVELOPMENT OF REGIONAL AIRPORTS

respectively. The Albury destination was selected from a short list of 20 regional
airports, which were considered with the Embraer jets in mind [5]. The size of
both the Embraer acquisition and the short list suggest that further expansion to
regional airports can be expected, and boosts the relevance of the analysis in this
chapter. In a further indication of the airline’s desire to grow business patronage,
one of the stated aims of the Embraer acquisition is to boost frequency on key
business routes [6], while presumably retaining load factors (LFs) and
profitability.

Figure 1: Network structure for the top 46 city pairs in Australia (ITLS mapping 2008).

Qantas was not prepared to let the entry of a LCC go unchecked, and in response
re-branded Impulse Airlines as Jetstar and commenced operations in May 2004.
Impulse had originally been an independent regional airline, but in 2000
commenced operations as a LCC using Boeing 717s. The airline encountered
cash flow problems, and by April 2001 Impulse was wet leasing its aircraft to
Qantas. In November 2001 Impulse was acquired by Qantas. Wholly owned by
Qantas, Jetstar is an LCC that operates independently and is run by a different
management team. Unlike Virgin Blue, Jetstar does not attempt to lure business
travellers; they are served by Qantas. Jetstar does not have airport lounges or
multiple classes on domestic routes. Qantas’ frequent flier points can be earned
on the more expensive fares, but Jetstar does not run its own programme.
In the 2006/2007 financial year, Jetstar carried 7.6 million passengers with 24
A320s, and held 15% of the domestic market share. The average cost per
available seat kilometre (CASK) is a common measure used to identify how low
cost an airline is. Jetstar had the lowest CASK of all Australian airlines, at 7.5
cents in 2007. By comparison, Virgin Blue, the other LCC, had a CASK of 8.2
 REGIONAL AIRPORTS AND OPPORTUNITIES FOR LOW-COST CARRIERS IN AUSTRALIA 153

cents in 2007. Compared with LCCs, traditional carriers, particularly regional

airlines, have much higher CASKs. For example, the CASK was 22.1 cents for
Regional Express (REX) and 19.7 cents for Skywest. The Qantas Group
(including Qantas, QantasLink and Jetstar) had a CASK of 11.5 cents in 2007.
This figure would be higher if Jetstar was excluded from the calculation (figures
sourced from various annual reports).
In 2007, Jetstar was awarded the title of ‘World’s Best Low-Cost Airline’,
based on a survey run by Skytrax Research [7]. Air Berlin and EasyJet were
awarded second and third place, respectively. This survey also listed Virgin Blue
as the second best LCC in the Australia/Pacific region.
Tiger Airways Australia is an LCC that commenced operations in November
2007. The airline is a part of the Tiger Aviation group, which also includes Tiger
Airways Singapore. Initial routes have centred on Melbourne Tullamarine airport
as a hub, with 13 destinations served by five A320s as of April 2008. Sydney and
Brisbane are notable for their absence in the list of initial destinations, with the
nearest served destinations to each being Newcastle and the Gold Coast,
respectively. Tiger Airways’ chief executive has suggested that these major
airports have not been serviced as they do not offer sufficiently low costs or the
required level of efficiency [8]. By contrast, Tullamarine was able to offer Tiger
a low-cost terminal. This is evidence that LCCs are actively considering airport
charges as they choose destinations, and a motivation for the inclusion of airport
charges in our models.
The full-service and LCCs are complemented by a variety of regional
airlines, which in recent years have undergone a degree of consolidation.
QantasLink is operated by Qantas and is comprised of three regional airlines:
Airlink, Eastern Australia Airlines and Sunstate Airlines. Regional Express, also
known as Rex, was formed as a merger of Hazelton Airlines and Kendell
Airlines, two profitable airlines that were owned by Ansett at the time of its
collapse. Skywest operates regional routes primarily in Western Australia, as
well as charter operations that cater to the booming mining industry in that state.
The Australian domestic aviation market is now highly competitive, with one
full-service carrier and three LCCs. Competition has been fierce on existing
routes, but the LCCs are also seeking to expand by introducing new routes and
destinations that have previously only been served by regional airlines. There is
growing evidence of the potential of Australia’s regional airports to grow
passenger traffic with additional LCC’s. The model developed below addresses
this issue.

2.2 Regional airports

Only Sydney, Melbourne, Brisbane, Perth and Adelaide have over five million
passengers per annum. Using this number of passengers per annum as the
threshold for the definition of a regional airport, all remaining airports can be
defined as regional. There is a great variability in the regional airports with
regards to the mix of inbound and outbound passenger flows. Some are
significant population centres and trip generators in their own right. Other
154 DEVELOPMENT OF REGIONAL AIRPORTS

regional airports largely serve inbound tourist flows, including many of the
airports along the Queensland coast.
In the late 1990s, many Australian airports were privatized. For example,
Townsville airport is now owned by Queensland Airports Limited, a regional
airport investment company. Privatization has allowed for a capital injection into
many airports, and has led to a greater focus on passenger growth and airport
profit. Other regional airports are owned by the surrounding local council(s),
either directly, as with Albury and Port Macquarie airports, or indirectly through
a company, as with Newcastle airport. Here the motivation is typically to support
and grow tourism and business in the local area.

Figure 2: Monthly revenue passenger movements, seats and aircraft trips for Newcastle–
Brisbane route.
Newcastle airport is a good case study example of a regional airport that has
been a beneficiary of LCC operations. Figure 2 details revenue passenger
movements, seat capacity and aircraft trips between Newcastle and Brisbane
from July 1994 to May 2007. From 1997 to 1999, when the route was serviced
by regional airlines, the route is notable for its high frequency of service, steady
passenger flows and low LF (an average of 48.8%). In May 2000, Impulse
Airlines commenced operations on the route with Boeing 717s, resulting in
modest increases in passenger flow but little change in LF. At that time, the
airline was moving from a regional to a low-cost model. However, by November
2001, Impulse was integrated into QantasLink, Qantas’ regional operator. Seat
capacity and passenger flows receded to pre-2000 levels. The boost in passenger
movements created by Impulse Airlines’ presence was small compared to the
almost immediate quadrupling of passenger movements to 20,000 per month
 REGIONAL AIRPORTS AND OPPORTUNITIES FOR LOW-COST CARRIERS IN AUSTRALIA 155

following the entry of both Virgin Blue and Jetstar on the route in May 2004. As
of mid-2007, passenger movements had doubled again to an average of 40,000
per month. The number of monthly aircraft trips following the entry of the LCCs
was not unprecedented, however the use of larger aircraft lead to substantially
greater seat capacity.
The substantial increase in passenger movements at Newcastle airport is
likely to have numerous causes. In addition to increased capacity and reduced
fare price resulting from the entry of Virgin Blue and Jetstar, the nature of
Newcastle as a trip origin and destination must be considered. Located 20
kilometres north of Newcastle (and 150 kilometres north of Sydney), Newcastle
airport primarily serves the Hunter region, which has a population of 573,000.
Additionally the Central Coast, a sizeable and growing population region, has the
potential to be a part of the catchment area. For many Central Coast residents, a
road journey to Newcastle airport would involve a similar distance, similar or
shorter travel time and less expensive parking than Sydney airport. Therefore,
the airport has an extensive outbound market. The Hunter region also boasts
numerous tourist attractions, including wineries and coastal holiday destinations.
The combination of increased service by LCCs, reduced fares and a strong
potential for trip production through a populous catchment area makes Newcastle
airport a great example of the growth that regional airports and LCCs can
experience by working in tandem.
Given the background on airline and airport activity in Australia, and the
growing role of LCCs, the rest of the paper focuses on the development and
application of a model system to represent the key demand and supply elements
of aviation activity, with a specific interest in identifying the opportunities to
grow passenger activity at specific airports through the introduction of LCCs.

3 Establishing a framework in which to investigate the role of

regional airports and airline activity
The previous section described both the growth in LCC operations in Australia,
and the impact this increase has had on some Australian regional airports. In this
section, we establish a formal framework through which we can analyse the
underlying drivers supporting the development of a regional airport or entry of a
new operator such as a low-cost airline.
The development of a demand model to predict base regular passenger
transport (RPT) flows on each route between points A and B in a network begins
with a theoretical definition of the potential influences on RPT flows. The
literature on factors influencing airline travel by a specific carrier is extensive
(see for example [9], Chapter 3). The major influences can be synthesized under
the broad headings of fare, service levels, the nature of the end points (i.e.
productions and attractions), presence of competitors and the capacity of an
airline to serve a market. Using a framework proposed by Dresner and Windle
[10], formally we can specify a demand function as in equation (1) as follows:
PASSj,AB = f {FAREj,AB, COMPj,AB,SERVICEj,AB, MARKETj,A, MARKETj,B}. (1)
156 DEVELOPMENT OF REGIONAL AIRPORTS

PASS is the annual passengers carried by airline j between points A and B

including both origin and destination passengers and flow-through passengers.
FARE is a vector of fares offered by airline j between points A and B. These
fares are not the full fare paid by a flow-through passenger, and would mainly
influence the origin–destination passengers. COMP acknowledges the role of
competitors on the route that affects both total demand and airline share. COMP
could be generalized to include competition from other modes such as the car,
train or coach. In the current study we confine the demand context to the airline
market. We can distinguish between LCCs and other carriers. SERVICE is a
vector of service (quality) attributes such as headways between flights, on-board
service, type of aircraft and airline image/reputation. MARKET refers to the
characteristics at the production and attractions ends of the AB endpoints. Total
population and its wealth as measured by per capita and household income are
often used as indicators of production and attraction.
A closer assessment of equation (1) will suggest that some of the explanatory
variables are endogenous. For example, air fare can be considered endogenous
because changes in passenger levels may trigger changes in prices, especially in
a liberalized competitive market. Endogenous fares are themselves a function of
a number of potential influences as summarized in equation (2).
FAREj,AB = f {PASSj,AB,SERVICEj,AB, MARKETj,A, MARKETj,B, COMPj,AB, LFj,AB, DISTAB}. (2)
The additional attributes on the right hand side are stage length distance (DIST)
of a route, which is a useful proxy for the cost of flying that route, and LF which
gives some idea of flight productivity and hence is linked to cost to the airline of
servicing each passenger. We also postulate that the number of competitors on a
route is also endogenous, dependent on the size of the market as shown in
equation (3).
COMPj ,AB = f {MARKET j ,A , MARKET j ,B }. (3)
Finally, the number of flights between each city pair is endogenous to the extent
that it is influenced by patronage on the demand side and competition on the
supply side. It is potentially influenced by airport landing and passenger charges.
FLIGHTS j ,AB = f {PASS j ,AB , COMPj ,AB , AIRPORTCHG j ,AB }. (4)
As no airline-specific passenger and fare data are available, j is neglected for this
study. Thus, instead of airline specific data, city-pair data is used for modelling
such as the passenger number for all airlines serving on a pair and the average
fare for all operating airlines. We have taken the natural logarithm of each
continuous variable as one possible functional specification for assessment. This
enables us to obtain mean estimates of direct elasticities from the parameter
estimate of the explanatory variable.
The structural equation system of four interdependent equations is a set of
simultaneous equations. We have chosen three stage least squares (3SLS) to obtain
parameter estimates, which starts with either two stage least squares (2SLS) or
seemingly unrelated regression (SUR). The rationale is set out below. 2SLS is a
single equation method, which means that over identifying restrictions in other
 REGIONAL AIRPORTS AND OPPORTUNITIES FOR LOW-COST CARRIERS IN AUSTRALIA 157

equations are not taken into account in estimating parameters in a particular

equation. As a result, 2SLS estimates are not asymptotically efficient. The system
method of 3SLS uses information concerning the endogenous variables in the
system and takes into account error covariances across equations, and hence is
asymptotically efficient in the absence of specification error. The SUR method
uses information about contemporaneous correlation among error terms across
equations in an attempt to improve the efficiency of parameter estimates.
The 2SLS method uses instrumental variable methods which involve
substituting a predicted variable for the endogenous variable Y when it appears
as a regressor. Hence the predicted variables are linear functions of the
instrumental variables and the endogenous variable substitutes Ŷ or Y, which
results in consistent estimates. Normally, the exogenous variables of the system
are used as the instruments. It is possible to use variables other than exogenous
variables from the system of equations as instruments; however, the estimation
may not be as efficient. For consistent estimates, the instruments must be
uncorrelated with the residual and correlated with the endogenous variable.
SUR may improve the efficiency of parameter estimates when there is
contemporaneous correlation of errors across equations. In practice, the
contemporaneous correlation matrix is estimated using ordinary least squares
(OLS) residuals. Under two sets of circumstances, SUR parameter estimates are
the same as those produced by OLS: when there is no contemporaneous
correlation of errors across equations (the estimate of contemporaneous
correlation matrix is diagonal); and when the independent variables are the same
across equations.
Theoretically, SUR parameter estimates will always be at least as efficient as
OLS in large samples, provided that the equations are correctly specified.
However, in small samples the need to estimate the covariance matrix from the
OLS residuals increases the sampling variability of the SUR estimates, and this
effect can cause SUR to be less efficient than OLS. If the sample size is small
and the across-equation correlations are small, then OLS should be preferred to
SUR. The consequences of specification error are also more serious with SUR
than with OLS.
The 3SLS method combines the ideas of the 2SLS and SUR methods. Like
2SLS, the 3SLS method uses Ŷ instead of Y for endogenous regressors which
results in consistent estimates. Like SUR, the 3SLS method takes the cross-
equation error correlations into account to improve large sample efficiency. For
3SLS, the 2SLS residuals are used to estimate the cross-equation error
covariance matrix. 3SLS is at least as efficient as any other estimator which uses
the same amount of information.

4 Model data
4.1 Sourcing of data for model estimation

The publicly available data reported by The Federal Bureau of Infrastructure,

Transport and Regional Economics (BITRE) aggregates revenue passenger
158 DEVELOPMENT OF REGIONAL AIRPORTS

movements for each city pair, such that we are unable to identify passengers
carried by each operator between a pair of end points. Further, data is not
available for routes with only one operator. No reasonable basis was found for
either of these two restrictions; the airlines claim the airline-specific information
is commercially sensitive. Thus, instead of airline-specific passenger data, city
pairs at the airport level are considered in this paper to identify the mutual
relationships between passenger numbers, fares and competition.
Traffic data for domestic and regional airline activity in Australia for the top
46 city pairs served by multiple airlines in the financial year 2006/2007 was
provided by BITRE [11]. All data at the city pair level is an aggregation of both
directions. That is, the city pair of Albury–Sydney aggregates the traffic
information both from Albury to Sydney and from Sydney to Albury. All city
pair figures are for direct flights between the two cities. For example, a flight
from Melbourne to Brisbane via Sydney will count both for the Melbourne–
Sydney pair and the Sydney–Brisbane pair. The data provides scheduled RPT
information including revenue passenger movements, aircraft movements,
available seats, LFs, revenue passenger kilometres (RPKs) and available seat
kilometres (ASKs). The 46 city pairs cover over 84.8% of total domestic
passenger movements and around 60.8% of total domestic aircraft trips for
Australia in 2007.
For each city pair, the number of traditional and LCCs was obtained from the
Airline On-time Performance Annual Report for the 2007 financial year [12],
which indicated the competition on each city pair. This report also has airline-
specific trip data (e.g. sectors flown, cancellations, departures/arrivals on time,
departures/arrivals delayed) for different routes; thus the proportion of each
airline with respect to frequency in those city pairs can be calculated and the
competition on routes can be revealed. However, airline-specific passenger
numbers have not been found at the city pair level and requests for such data
from each airline were singularly unsuccessful. Monthly lowest fare information
on the top 70 city pairs were provided by BITRE, in four categories where
available. The categories are business, full economy (transferable and fully
refundable), restricted economy (transferable and non-refundable) and best
discount (cheapest fare). The lowest restricted economy fares were averaged into
annual figures for modelling. In addition to traffic related information, distances
in kilometres between airports for 46 city pairs were obtained from BITRE [13].
The characteristics of an airport and its surrounding region are crucial.
Providing low charges may help airports attract LCCs. Landing charges and
passenger charges are two main categories for RPT services. Landing charges (or
runway charges) are based on per tonne of maximum take-off or landing weight
(MTOW). Passenger charges are levied per arriving/departing passenger through
the domestic terminal including a terminal usage charge and an aeronautical
passenger charge. For freight services, a freight (or cargo) charge is calculated on
per tonne of goods discharged from or loaded into aircraft. While many
Australian airlines typically transport freight in addition to passengers, freight
charges have not been included in the models. Airport charging rates were
obtained by directly contacting individual airports or local councils. All charges
 REGIONAL AIRPORTS AND OPPORTUNITIES FOR LOW-COST CARRIERS IN AUSTRALIA 159

and fares include a goods and services tax (GST) of 10%. Annual inbound and
outbound passenger movements from all destinations were obtained for each
airport from BITRE [14]. All airports are identified by International Air
Transport Association (IATA) codes.
Information describing the area surrounding the airport was gathered,
including (1) population and weekly per capita income at the local area or district
where an airport is located [15], and (2) accommodation statistics (i.e. the total
number of beds in hotels, motels and serviced apartments with five or more
rooms) at the tourism regional level [16].
Table 1 provides a summary of all city pair level data available for model
estimation.
Table 1: Data available for model estimation.
Variable Description
PopulatA Local area population at Airport A
PopulatB Local area population at Airport B
IincomeA Median individual weekly income for the local area where
Airport A is located
IincomeB Median individual weekly income for the local area where
Airport B is located
HincomeA Median household weekly income for the local area where
Airport A is located
HincomeB Median household weekly income for the local area where
Airport B is located
NobedsA Number of beds in the tourism region where Airport A is located
NobedsB Number of beds in the tourism region where Airport B is located
PaxA Annual total passengers at Airport A
PaxB Annual total passengers at Airport B
LandCA Landing charge for Airport A
LandCB Landing charge for Airport B
PaxCA Passenger charge for Airport A
PaxCB Passenger charge for Airport B
Distance Distance between airports (kilometers)
AvFare Average restricted economy fare for a city pair in the 2007
financial year
Pax Annual total passenger movement number for a city pair
Seat Annual total seat number for a city pair
RPK Annual total Revenue passenger kilometers (000s)
LF Load factor
ASK Annual total available seat kilometers (000s)
Nocomp Number of competitors for a city pair (traditional and LCCs)
NoLCA Number of LCCs for a city pair
Alltrip Annual total aircraft trip number for a city pair
160 DEVELOPMENT OF REGIONAL AIRPORTS

4.2 Overview of available data

The approach to model estimation is strictly cross-sectional. Thus, the predicted

passenger flows are essentially the long-run profile under a specified network
configuration. When developing an appropriate data set for model estimation, we
must ensure that we have a sufficiently large sample that has a rich variability in
the set of variables that we wish to test as possible sources of influence on
passenger flows. We have a total of 46 city pair observations, serviced by six
airlines (two LCCs, three regional carriers and one traditional carrier). Table 2
lists the airlines that serve each of the 46 city pairs. Tiger Airways began
operations in November 2007, and so is not included in the data.
Table 2: Routes under study.
Port A Code A Port B Code B Airlines operating on route
Albury ABX Sydney SYD QantasLink and Regional Express
Adelaide ADL Brisbane BNE Jetstar*, Virgin Blue* and Qantas
Adelaide ADL Canberra CBR Virgin Blue* and Qantas
Adelaide ADL Melbourne MEL Virgin Blue* and Qantas
Adelaide ADL Gold Coast OOL Jetstar* and Virgin Blue*
Adelaide ADL Perth PER Virgin Blue* and Qantas
Adelaide ADL Sydney SYD Jetstar*, Virgin Blue* and Qantas
Virgin Blue*, Qantas, QantasLink
Broome BME Perth PER and Skywest
Brisbane Virgin Blue*, Qantas and
BNE Canberra CBR QantasLink
Brisbane BNE Cairns CNS Jetstar*, Virgin Blue* and Qantas
Brisbane BNE Darwin DRW Jetstar*, Virgin Blue* and Qantas
Brisbane BNE Hobart HBA Jetstar* and Virgin Blue*
Brisbane BNE Hamilton Island HTI Jetstar* and Virgin Blue*
Brisbane BNE Melbourne MEL Jetstar*, Virgin Blue* and Qantas
Brisbane Jetstar*, Virgin Blue* and
BNE Mackay MKY QantasLink
Brisbane Jetstar*, Virgin Blue* and
BNE Newcastle NTL QantasLink
Brisbane BNE Perth PER Virgin Blue* and Qantas
Brisbane BNE Proserpine PPP Jetstar* and Virgin Blue*
Brisbane Jetstar*, Virgin Blue* and
BNE Rockhampton ROK QantasLink
Brisbane BNE Sydney SYD Virgin Blue* and QantasLink
Brisbane BNE Townsville TSV Jetstar*, Virgin Blue* and Qantas
 REGIONAL AIRPORTS AND OPPORTUNITIES FOR LOW-COST CARRIERS IN AUSTRALIA 161

Table 2: Routes under study (continued).

Port A Code APort B Code BAirlines operating on route
Ballina Jetstar*, Virgin Blue* and Regional
BNK Sydney SYD Express
Canberra CBR Melbourne MEL Virgin Blue*, Qantas and QantasLink
Canberra CBR Sydney SYD Virgin Blue*, Qantas and QantasLink
Coffs
Harbour CFS Sydney SYD Virgin Blue* and QantasLink
Cairns CNS Melbourne MEL Jetstar*, Virgin Blue* and Qantas
Cairns CNS Sydney SYD Jetstar*, Virgin Blue* and Qantas
Dubbo DBO Sydney SYD QantasLink and Regional Express
Darwin DRW Melbourne MEL Jetstar* and Virgin Blue*
Hobart HBA Melbourne MEL Jetstar*, Virgin Blue* and Qantas
Hobart HBA Sydney SYD Jetstar*, Virgin Blue* and Qantas
Hervey Bay HVB Sydney SYD Jetstar* and Virgin Blue*
Kalgoorlie KGI Perth PER Qantas and QantasLink
Jetstar*, Virgin Blue*, Qantas and
Launceston LST Melbourne MEL QantasLink
Launceston LST Sydney SYD Jetstar* and Virgin Blue*
Sunshine
Coast MCY Melbourne MEL Jetstar* and Virgin Blue*
Sunshine
Coast MCY Sydney SYD Jetstar* and Virgin Blue*
Melbourne MEL Mildura MQL QantasLink and Regional Express
Melbourne MEL Newcastle NTL Jetstar*, Virgin Blue* and QantasLink
Melbourne MEL Gold Coast OOL Jetstar* and Virgin Blue*
Melbourne MEL Perth PER Jetstar*, Virgin Blue* and Qantas
Melbourne MEL Sydney SYD Jetstar*, Virgin Blue* and Qantas
Gold Coast OOL Sydney SYD Jetstar*, Virgin Blue* and Qantas
Perth PER Sydney SYD Virgin Blue* and Qantas
Sydney SYD Townsville TSV Jetstar*, Virgin Blue* and Qantas
Wagga
Sydney SYD Wagga WGA QantasLink and Regional Express
Note: *Denotes a LCC.
The network structure for those 46 city pairs is visualized in Figure 1. State capitals
are in a larger font. A large percentage of cities are located on the Eastern
seaboard, which is reflective of Australia’s geographic population distribution.
Revenue passenger movements for all city pairs in the financial year
2006/2007 are given in Figure 3, with movements varying from a high of
6,624,665 for Melbourne–Sydney to a low of 121,056 for Darwin–Melbourne.
The top three pairs link Sydney, Melbourne and Brisbane, and together account
162 DEVELOPMENT OF REGIONAL AIRPORTS

for approximately 40% of all revenue passenger movements. Of the top 46 pairs,
36 have less than 1,000,000 annual passenger trips. Gold Coast–Sydney is the
fourth largest pair by revenue passenger movements, and the largest that includes
a regional airport (Gold Coast). Figure 4 indicates the annual number of aircraft
trips for each city pair, with flights varying from 41,907 for Melbourne–Sydney
to 835 for Darwin–Melbourne.

Figure 3: Revenue passenger movements for city pairs.

Figure 4: Total aircraft trips for city pairs.

 REGIONAL AIRPORTS AND OPPORTUNITIES FOR LOW-COST CARRIERS IN AUSTRALIA 163

The LF indicates the proportion of total aircraft seats that are filled by revenue
passengers [17], which is a key parameter to establish seat utilization over routes.
Figure 5 presents LFs for the top 46 city pairs in Australia. The average LF is
0.78; Melbourne–Gold Coast has the highest LF (0.87), while the LF for
Sydney–Wagga Wagga is the lowest (0.66). The majority (76.1%) of pairs
generate LFs between 0.75 and 0.80, which reveals efficient utilization on those
top pairs in general.

Figure 5: Load factors for city pairs.

Average lowest air fares and distances between airports are shown in Figure 6.
Distances range from 236 kilometres to 3615 kilometres, while fares range from
A$199.00 to A$598.70. Generally, there is consistency in price for any given
distance, with fares increasing with distance. However, some routes have notably
higher costs per kilometre than other routes of similar distance. Examples include
Albury–Sydney, Kalgoorlie–Perth and Sydney–Wagga Wagga, all of which have
no low-cost airline. The absence of LCC competition might be contributing to the
high fares. However, high costs per kilometre for Broome–Perth, which is serviced
by the LCC Virgin Blue, suggests that other factors are also at play.
A total of 27 airports are in the domestic and regional network under study.
Table 3 shows the airports served by each of the airlines in the study.
From Table 3, only Sydney, Melbourne, Brisbane, Perth and Adelaide have
over five million passengers per annum. Using this number of passengers per
annum as the threshold for the definition of a regional airport, all remaining
airports can be defined as regional. Thus, 10 of 46 routes link primary airports,
while the remaining 36 routes link a primary airport with a regional airport.
Given that the total number of passengers for all of Australia’s airports was
112,068,399 for the financial year 2006/2007 [14], the 27 airports analysed cover
94.9% of total airport passenger movements.
164 DEVELOPMENT OF REGIONAL AIRPORTS

Figure 6: Average fares and distances between airports.

Figure 7: Profile of competition on routes.

Figure 8: Profile of competition on routes by low-cost airline.

 REGIONAL AIRPORTS AND OPPORTUNITIES FOR LOW-COST CARRIERS IN AUSTRALIA 165

Of the 46 city pairs involved in this study, which are all served by more than one
airline, the majority (52.2%) of them are served by three airlines (see Figure 7).
With respect to LCCs, Virgin Blue and Jetstar compete with each other on 34
pairs or 73.9% of total city pairs, and only 10.9% of them are not served by any
LCC (see Figure 8). These figures reveal the high level of competition on the top
routes in Australia.
Table 3: Ports served by airlines in the study.
Airport Passengers Airlines
A B C D E F
Sydney 31,016,186 9 9 9 9 9
Melbourne 22,156,871 9 9 9 9 9
Brisbane 17,379,809 9 9 9 9
Perth 7,977,091 9 9 9 9 9
Adelaide 6,181,390 9 9 9
Cairns 3,782,183 9 9 9
Gold Coast 3,777,856 9 9 9
Canberra 2,687,336 9 9 9
Hobart 1,629,417 9 9 9
Darwin 1,403,685 9 9 9
Townsville 1,271,649 9 9 9
Launceston 995,664 9 9 9 9
Newcastle 958,087 9 9 9
Sunshine Coast 880,822 9 9
Mackay 743,321 9 9 9
Rockhampton 638,602 9 9 9
Hamilton Island 465,941 9 9
Broome 344,790 9 9 9
Ballina 323,791 9 9
Coffs Harbour 323,565 9
Proserpine 256,282 9 9 9
Albury 212,264 9 9
Kalgoorlie 211,857 9 9
Wagga Wagga 203,798 9 9
Hervey Bay 189,429 9 9
Dubbo 171,026 9 9
Mildura 167,983 9 9
Total number of airports served 10,635,0695 19 22 14 17 8 2
Note: For airlines, where: A=Jetstar; B=Virgin Blue; C=Qantas; D=QantasLink;
E=Regional Express; F=Skywest.

5 Results of base year models

The passenger demand model, the air fare model, the competition model and the
flight supply model are estimated simultaneously, addressing feedback between
the left-hand side variables. The final passenger demand model for the financial
166 DEVELOPMENT OF REGIONAL AIRPORTS

year 2007 is presented in Table 4. A number of alternative functional forms were

investigated in the process leading to the selection of the preferred model. All the
explanatory variables have the expected sign and are statistically significant at
the 5% level of significance. The measure of overall goodness-of-fit (R2) must be
treated with caution in simultaneous equations. For two-stage least squares, some
of the regressors enter the model as instruments when the parameters are
estimated. However, since our objective is to estimate the structural model, the
actual values, not the instruments for the endogenous right-hand-side variables,
are used to determine the model sum of squares (MSS). The model’s residuals
are computed over a set of regressors. We have been able to explain 44.3% of the
single-equation variation in passenger numbers between the 46 pairs by the
differences in five right hand variables and a constant.
What we find is that the number of passengers travelling between each location
pair is inversely related to average fare and directly related to population at
origin and destination, the number of competitor airlines on the route and the
presence of one or more low-cost airlines (1,0).
As a double logarithmic form, the parameter estimates can be directly
interpreted as elasticities. For example, –1.18778 indicates that a 1% increase in
average fare, all other things being equal, results in a 1.18778% reduction in the
demand for passenger movements per annum. The key drivers of passenger
demand are total population at two end points of a pair and the competition on a
route. The presence of LCCs is also expected to stimulate more passengers. There
are two main reasons for this. First, capacity for that route increases with more
airlines in service, and secondly the air fare decreases due to increasing
competition, particularly from a low-cost airline. This is also supported by the air
fare model given in Table 5. All right-hand side variables are statistically
significant at the 5% level of significance, except the presence of LCCs which is
significant at the 10% level. 38.8% of the variation in fares can be explained by the
five influences. The average fare varies inversely with the number of passengers
(the quantity effect) and the presence of one or more LCCs; and directly by the
median personal income at the origin and destination as well as the LF.
Table 4: Preferred passenger demand model.
Right hand side variables Acronym Final full model
(t-ratio in brackets)
Ln (average fare) LNAVFARE –1.18778 (–2.84)
Ln (the number of competitors) LNCOMP 1.54455 (2.46)
Ln (product of populations at the local LPOPAB 0.21791 (3.96)
area for airports A and B)
Dummy variables (1,0) for the LCC 0.72730 (2.07)
presence of one more LCCs
Constant 12.0174 (4.54)
R2 0.443
Adjusted R2 0.388
No. of observations 46
Note: Dependent variable: Natural logarithm of the number of passenger movements.
 REGIONAL AIRPORTS AND OPPORTUNITIES FOR LOW-COST CARRIERS IN AUSTRALIA 167

Table 5: Preferred air fare model.

Right hand side variables Acronym Final full model
(t-ratio in
brackets)
Ln (the number of passengers) LNPAX –0.0939 (–2.54)
Ln (product of weekly per capita income LNPINCAB 0.8256 (3.13)
at the local area for Airports A and B)
Ln (load factor) LNLF 2.71243 (3.62)
Dummy variables (1,0) for the presence of LCC –0.25676 (–1.87)
one or more LCCs
Constant –2.4103 (–0.75)
R2 0.388
Adjusted R2 0.328
No. of observations 46
Note: Dependent variable: Natural logarithm of the average fare.
Table 6 summarizes the findings for the competition model. All right hand side
variables except the constant are statistically significant at the 5% level of
significance. The number of airlines competing on the route is positively related
to the presence of one or more low-cost airlines (1,0) and the median weekly
household income at the origin and destination.
Finally, as summarized in Table 7, the number of flights between a pair of
airports is positively influenced by patronage demand. As passenger demand
grows, airlines respond by adding more capacity on the route. The implied mean
elasticity of 0.7842 indicates that a 10% increase in passengers, all other factors
held constant, lead to an average 7.84% increase in flights. Importantly, the
number of flights is inversely related to the product of landing and passenger
charges at each airport in the pair, supporting the position that airport charges do
have a statistically significant influence on attracting air services. This could be
more significant to low-cost airlines, as they are more cost-sensitive than
traditional carriers.
Table 6: Preferred competitor airlines model.
Right hand side variables Acronym Final full model
(t-ratio in
brackets)
Ln (product of median weekly household LNHINC 0.34978 (1.96)
income at the local area for airports A and B) AB
Dummy variables (1,0) for the presence of LCC 0.30362 (3.17)
one or two LCCs
Constant –4.2007 (–1.64)
R2 0.208
Adjusted R2 0.148
No. of observations 46
Note: Dependent variable: Natural logarithm of competitor number on each route.
168 DEVELOPMENT OF REGIONAL AIRPORTS

Table 7: Preferred air trip supply model.

Right-hand-side variables Acronym Final full model

(t-ratio in
brackets)
Ln (the number of passengers) LNPAX 0.78421 (–2.16)
Ln (the product of landing and passenger LNAPCHAB –0.00046 (–2.13)
charges at each airport in the pair)
Constant –1.8719 (–2.11)
R2 0.758
Adjusted R2 0.747
No. of observations 46
Note: Dependent variable: Natural logarithm of the number of flights between a pair.

6 Scenario assessment – introduction of low-cost carriers

(LCCs)
The model system developed in the previous section can be used to undertake
scenario analysis. In particular, we are interested in what would be the impact of
key policy instruments such as fares, new entrants on a route, especially LCCs
on patronage and hence on regional airport activity; and also what influence does
airport charges and patronage growth have on the amount of flight activity to and
from a regional airport.
We have set up a scenario model to evaluate a range of ‘what if’ applications.
A reduced form of the four structural equations is used in scenario applications.
The reduced-form models of interest are the natural log of passenger movements
between each airport pair and the natural log of the number of flights between
airport pairs.
The patronage prediction model is:
Ln(Pax) −[(−1.18778) ×(−0.09389) × Ln(Pax)] =12.0174
−1.18778×[−2.41027 +1.187786× Ln(pincAB)] + 2.71243× Ln(LF) − 0.25676× LCC]
+1.54455×[−4.20066 + 0.34978× Ln(hincAB) + 0.30362× LCC]
+ 0.21791× Ln(popAB) + 0.72730× LCC,
Ln(No. of Flights) = −1.87193 − 0.00046 × Ln(landchg × passchj) + 0.78421× Ln(Pax).
The patronage model can be simplified by rearranging terms to result in:
Ln(Pax) = {8.391 − 0.98057 × Ln(pincAB) − 3.22177 × Ln(LF) + 1.50122 × LCC
+ 0.540253 × Ln(hincAB) + 0.21791× Ln(popAB)}/ 0.888479.
These models have been calibrated (via the constant) to produce the relationship
shown in Table 8 between actual patronage and predicted patronage on each
airport pair. Table 9 is the relationship between actual and predicted flights per
annum. The discrepancy is less than 1000 passenger trips per annum.
REGIONAL AIRPORTS AND OPPORTUNITIES FOR LOW-COST CARRIERS IN AUSTRALIA 169

Table 8: Calibrated airport pairs patronage.

Airport pair Actual pass Pred pass Diff pred vs. act
ABX–SYD 155,609 154,935 –674
ADL–BNE 643,579 644,297 718
ADL-CBR 201,511 201,215 –296
ADL–MEL 1,862,257 1,862,220 –37
ADL–OOL 173,876 174,414 538
ADL–PER 533,226 533,895 669
ADL–SYD 1,483,330 1,483,515 186
BME–PER 268,445 268,106 –339
BNE–CBR 595,824 596,154 331
BNE–CNS 1,191,234 1,191,173 –61
BNE–DRW 330,579 330,168 –411
BNE–HBA 137,256 137,081 –175
BNE–HTI 164,358 164,129 –228
BNE–MEL 2,632,646 2,632,322 –324
BNE–MKY 626,685 627,126 441
BNE–NTL 463,982 464,348 366
BNE–PER 563,713 563,532 –181
BNE–PPP 202,663 203,035 372
BNE–ROK 563,093 563,532 439
BNE–SYD 3,934,916 3,935,407 492
BNE–TSV 851,283 851,746 463
BNK–SYD 284,532 285,227 695
CBR–MEL 954,650 955,363 714
CBR–SYD 829,103 829,047 –57
CFS–SYD 300,740 301,060 320
CNS–MEL 458,447 458,119 –328
CNS–SYD 946,381 945,735 –646
DBO–SYD 181,099 181,173 74
DRW–MEL 121,055 120,846 –209
HBA–MEL 1,001,491 1,001,610 119
HBA–SYD 426,258 426,279 21
HVB–SYD 150,317 149,828 –489
KGI–PER 211,145 211,855 710
LST–MEL 725,924 725,122 –802
LST–SYD 202,016 202,806 791
MCY–MEL 410,118 409,813 –305
MCY–SYD 444,053 444,406 353
170 DEVELOPMENT OF REGIONAL AIRPORTS

Table 8: Calibrated airport pairs patronage (continued).

Airport pair Actual pass Pred pass Diff pred vs. act
MEL–MQL 152,711 153,373 663
MEL–NTL 339,083 338,827 –256
MEL–OOL 1,273,399 1,272,959 –440
MEL–PER 1,539,857 1,539,652 –205
MEL–SYD 6,624,596 6,624,613 17
OOL–SYD 2,051,743 2,051,485 –257
PER–SYD 1,321,133 1,320,831 –302
SYD–TSV 158,816 159,575 760
SYD–WGA 167,745 168,392 647
Average 844,706 844,790 84

Table 9: Calibrated airport pairs flights.

Airport pair Actual flights Pred flights Diff Pred vs. Act
ABX–SYD 5390 4855 –535
ADL–BNE 4992 5492 500
ADL–CBR 1853 2209 356
ADL–MEL 14,334 14,828 494
ADL–OOL 1357 1968 611
ADL–PER 4050 4739 689
ADL–SYD 11,610 11,683 73
BME–PER 2504 2766 262
BNE–CBR 5614 5169 –445
BNE–CNS 7950 8900 950
BNE–DRW 2085 2536 451
BNE–HBA 1058 1635 577
BNE–HTI M 1883 –
BNE–MEL 19,157 19,845 689
BNE–MKY 5713 5378 –335
BNE–NTL 4536 4249 –287
BNE–PER 3257 3667 410
BNE–PPP 1449 2219 770
BNE–ROK 7493 8154 661
BNE–SYD 27,648 28,025 378
BNE–TSV 6292 7189 897
BNK–SYD M 2896 –
CBR–MEL 10,308 9605 –703
 REGIONAL AIRPORTS AND OPPORTUNITIES FOR LOW-COST CARRIERS IN AUSTRALIA 171

Table 9: Calibrated airport pairs flights (continued).

Airport pair Actual flights Pred flights Diff Pred vs. Act
CBR–SYD 15,587 14,905 –682
CFS–SYD M 3024 –
CNS–MEL 3235 3733 498
CNS–SYD 5520 6395 876
DBO–SYD 5818 5253 –564
DRW–MEL 835 1481 646
HBA–MEL 7331 7769 437
HBA–SYD 3351 3976 625
HVB–SYD 1144 1756 612
KGI–PER 2700 2291 –409
LST–MEL 6572 6036 –536
LST–SYD 1539 2214 675
MCY–MEL 2880 3490 610
MCY–SYD 3081 3752 671
MEL–MQL M 5235 –
MEL–NTL 3976 3323 –653
MEL–OOL 8436 8658 222
MEL–PER 8577 9369 792
MEL–SYD 41,907 41,747 –159
OOL–SYD 15,081 15,067 –14
PER–SYD 6159 6802 643
SYD–TSV 1283 1833 550
SYD–WGA 6453 6043 –410
Average 6880 7044 164
Note: M = missing in data.
If we increase the number of LCCs on routes where there is currently no low-
cost carrier, five in total, at flight levels typically provided on other regional
routes, then we observe a substantial increase in potential patronage as
summarized in Table 10 (i), and the prediction of additional flights (Table 10
(ii)) to accommodate the extra demand and existing LFs. Selecting the Albury–
Sydney pair as an example, there is the potential to grow 180,815 passenger trips
in both directions per annum with 11 LCC flights daily. This amounts on average
to 5.5 flights in each direction per day for an extra 247 passengers in each
direction per day, a payload per flight of close to 45 passengers. This seems a
sensible set of estimates to attract a LCC.
We investigated the influence of airport charges on the number of flights
offered to establish if it was having a significant influence on flights offered out
of specific airports. Table 11 shows that the sensitivity to airport landing and
passenger charges is very small indeed, consistent with the elasticity parameter
172 DEVELOPMENT OF REGIONAL AIRPORTS

in Table 6. Hence on our evidence we can discount this as a major determinant of

airline choice of activity to regional airports.

Table 10: Impact of introducing an LCC on routes without LCCs

(i) Patronage impact.
Airport pair (two-way) Pred pass Base pred pass Diff
ABX–SYD 335,750 154,935 180,815
DBO–SYD 392,609 181,173 211,436
KGI–PER 459,097 211,855 247,242
MEL–MQL 332,366 153,373 178,993
SYD–WGA 364,911 168,392 196,520
(ii) Number of new flights required.
Extra one- Pass per
New pred Base pred Diff pred way flights Extra extra
Airport pair flights flights vs. actual daily pass/day flight
ABX–SYD 8803 4816 3987 11 495 45
DBO–SYD 9563 5213 4350 12 579 49
KGI–PER 4183 2275 1908 5 677 130
MEL–MQL 15,337 8335 7003 19 490 26
SYD–WGA 11,028 5995 5033 14 538 39

Table 11: Impact of 10% reduction in airport charges at key regional airports.

City Pair Flights pred Flights actual Increase

BNE–MKY 5226 5219 6
BNE–NTL 9288 9277 11
BNE–PPP 2404 2402 3
BNE–ROK 7600 7590 9
BNE–TSV 6320 6313 8
MEL–NTL 9680 9668 12
SYD–TSV 5509 5503 7
SYD–WGA 2150 2148 3

7 Conclusions
This paper has reviewed the progress of aviation activity and development of
services to and from regional airports in Australia, as background to the
development of a model system capable of identifying the factors influencing the
demand for passenger movements and number of flights between airport pairs in
Australia. The modelling framework also recognized the endogeneity of air fares
and the number of competitors in the determination of the market for regional
aviation activity.
 REGIONAL AIRPORTS AND OPPORTUNITIES FOR LOW-COST CARRIERS IN AUSTRALIA 173

For regional airports that currently are not serviced by LCCs, five in total, we
investigated the opportunities to grow patronage and increase flights in the
context of ensuring an acceptable payload per flight. The opportunity gap in the
market is shown to exist. This is encouraging and indeed we are aware that a
LCC is planning to service Albury in 2008.

Acknowledgements
Jonathan Firth of the Bureau of Infrastructure, Transport and Regional
Economics (BITRE) provided extensive support in sourcing published and
unpublished data, as did Theo Koo of the University of New South Wales.
Thanks also to Larry Dwyer and Peter Forsyth for advice.

References
[1] Department of Infrastructure, Transport, Regional Development and Local
Government, Australian Transport Statistics Yearbook 2007, Canberra:
Australian Government, 2008.
[2] Nyathi, M., Hooper, P. & Hensher, D.A., Compass Airlines: 1 December
1990 to 20 December 1991: What went wrong? Parts I and II. Transport
Reviews, 13, pp. 119–149, 185–206, 1993.
[3] Forsyth, P., Total factor productivity in Australian domestic aviation.
Transport Policy, 8, pp. 201–207, 2001.
[4] Forsyth, P., Low-cost carriers in Australia: Experiences and impacts.
Journal of Air Transport Management, 9, pp. 277–284, 2003.
[5] Virgin Blue, Ho Ho Ho – It’s off to Albury we go! http://www.virginblue.
com.au/AboutUs/Media/NewsandPressReleases/P_002088.htm on 3 April
2008.
[6] Virgin Blue, Virgin Blue Orders 20 Embraer ‘E-Jets’ – New Fleet Strategy
for Australia’s ‘New World Carrier’, http://www.virginblue.com.au/
AboutUs/Media/NewsandPressReleases/U_001189.htm.
[7] Skytrax Research, World Airline Awards, http://www.worldairlineawards.
com/Awards_2007/Lowcost.htm.
[8] The Australian, Tiger vows to shun high-cost airports, http://www.
theaustralian.news.com.au/story/0,25197,22687139-23349,00.html.
[9] Trethaway, M.W. & Oum, T.H., Airline Economics: Foundations for
Strategy and Policy, Centre for Transportation Studies, University of British
Columbia: Vancouver, 1992.
[10] Dresner, M. & Windle, R., Are US air carriers to be feared: Implication of
hubbing to North Atlantic competition. Transport Policy, 2(3), pp. 195–202,
1995.
[11] BITRE, Domestic Airline Activity, http://www.btre.gov.au/info.aspx?
NodeId=101.
[12] BITRE, Airline on Time Performance Annual Reports, http://www.
bitre.gov.au/publications/17/Files/Master200607.xls.
174 DEVELOPMENT OF REGIONAL AIRPORTS

[13] BITRE, Australian Air Distances, http://www.btre.gov.au/Info.aspx?

NodeId=97.
[14] BITRE, Airport Traffic Data, http://www.btre.gov.au/Info.aspx?NodeId=96.
[15] Census, Census of Population and Housing, http://www.abs.gov.au/
websitedbs/d3310114.nsf/Home/census.
[16] ABS, Tourist Accommodation, Small Area Data, http://www.abs.gov.au
/AUSSTATS/[email protected]/DetailsPage/8635.3.55.001Jun%202007?OpenDocu
ment.
[17] BITRE, Avline, http://www.btre.gov.au/Info.aspx?ResourceId=606&
NodeId=92.
Subject Index
Accessibility Choice
Land, 28, 40, 81, 82 Airline, 5, 7, 42, 79, 93, 94, 109, 113, 117,
Index, 80, 81 129, 135–137
Size, 149 Access mode, 7, 109, 133, 134
Aggregate, 10, 11, 14, 18, 84, 85, 87, 91, 98, Discrete, 83–88, 91–94, 109, 115, 127, 128
105, 114, 115, 120, 134, 157, 158 Main mode, 112, 133, 134
Air fare, 4, 8, 43, 46, 112–114, 119–121, 123, Route, 133, 134
131, 137, 142, 149, 150, 156, 165, 166 Traveller, 129–135
Aircraft movements, 5, 31, 42, 57, 67, 127, 128, User, 7, 83, 93, 97
135, 136, 137, 149, 158 Competitive
Airport Airport, 78, 80, 82, 84, 97, 98
Classification, 3, 65–66, 79, 96 Demand
Community, 2, 28, 29, 32, 45, 95 Forecast, 77, 79, 84, 85, 127, 128, 130, 131,
Feeder, 2, 3, 123 145
Heavy, 2, 63 Level, 1, 2, 4, 5, 6, 42, 78, 79, 81, 85, 88
Hub, 2, 6, 28, 31, 39, 42, 43, 56, 61, 68, 75, Model, 77, 79, 83–85, 87, 90, 92, 93, 103,
131 127, 129, 134, 155, 165, 166
Light, 2 Disaggregate, 11, 84, 91, 92, 112, 114, 115
Medium, 2, 44, 64 Econometric, 11, 13
National, 2, 3, 9, 11, 16, 18, 44, 60, 69, 74, Economies
82, 95, 97, 120 Local, 9, 12
Pair, 83, 167–171 Regional, 9, 11
Primary, 2, 39, 48, 49, 81, 163 Economic
Approach Development, 1, 9–11, 17, 21, 22, 28, 44, 55,
Behavioural, 80, 94, 121 63, 74, 104, 128, 139
Multi-mode, 83, 84 Productivity, 11
Uni-mode, 83, 84 Effects
Benefit(s), 10, 12, 29, 31, 32, 35, 37, 38, 45, 78, Primary, 12, 14
106, 146, 150 Secondary, 12, 14
Capacity Tertiary, 12
Constraint, 7, 127–129, 136, 137, 139, 140, Perpetuity, 13
150 Elasticity(ies), 94, 130, 135, 137, 145, 156, 166,
Environmental, 30, 31, 127 167, 171
Runway, 135, 137–138 Equilibrium, 5, 7, 8, 12, 44, 128, 137, 145
Capital(s) Freight, 1, 36, 45, 49, 134, 140, 158
Cities, 57, 60, 61, 63, 65, 67, 68, 70, 73–75, Flow(s), 2, 5, 15, 17, 18, 34, 36, 95, 97, 127–
149, 151, 161 130, 134, 152–155, 160
Monetary, 10, 11, 13, 14, 151 Grandfathering rights, 32, 137–140
Carrier Growth
Low-cost (LCCs), 3, 4, 8, 22 27–29, 36, 40, New growth theory, 11
46–49, 61, 96, 110, 123, 149, 150, 171 Hedonic, 88, 102
Full-service (FSCs), 40, 43, 46, 47, 69, 153 Impact(s)
Catchment area Economic, 11, 13, 22
Primary, 81, 82 Labour, 1, 10, 11, 14–16, 23, 151
Secondary, 81 Least squares
176 SUBJECT INDEX
Three stage (3SLS), 149, 156 Revenue, 12, 17, 18, 21, 41, 42, 48, 50, 149,
Low-fare, 15, 42, 47 154, 157–159, 161–163
Model Scenario, 6, 78, 85, 128, 136, 138–140, 142,
Competition, 127, 165, 167 143, 168
Discrete choice, 83–86, 88, 91–94, 109, 115, Slot trading, 137–139
128 Strategic, 18, 21, 25, 32, 84–86, 91, 127, 128,
Model 145
Multivariate, 91, 103, 104, 106 Tactical, 84, 85, 127
Patronage, 168 Time series, 83–86, 88–92, 96, 97, 100, 103,
Random utility (RUM), 92, 115, 133 105, 106, 127, 129
Structural, 166 Traffic
Univariate, 84, 89, 91, 100, 106 Business, 16
Migration, 11, 13–16, 23 Ethnic, 16
Parameter(s), 87–91, 94, 103, 115–117, 119, Leisure, 16, 97
121, 156, 157, 163, 166, 171 Zone, 80, 81, 83, 84, 89–92
Point-to-point, 2–4, 21, 39, 43, 37, 43, 44, 46, Variables
47, 49, 53, 63, 74, 82, 83, 95, 123 Dependent, 79, 84, 85, 90, 166–168
Price Independent, 79, 84, 85, 90–92, 157
Shadow, 127
This page intentionally left blank
This page intentionally left blank
 ...for scientists by scientists

Railway Timetabling, Operations Analysis and

Rescheduling
Edited by: I. HANSEN, Delft University of Technology, The Netherlands

The book comprises a number of research papers presented at several

Computers in Railways Conferences. It has been compiled by Ingo A. Hansen,
President of the International Association of Railway Operations Research
(IAROR) and comprises selected papers originating from different countries,
such as Denmark, France, Germany, Japan, Italy, Netherlands, Sweden and
Switzerland. The papers give an overview of the current state-of-the-art
analytical approaches, methods and simulation tools for the modelling and
analysis of network timetables, the distribution of train delays and real-time
rescheduling of perturbed operations. The topics include e.g. railway capacity
estimation according to the UIC norm 406, train punctuality analysis based
on standard track occupation and clearance data, and boarding, alighting and
distribution of passengers along suburban trains, as well as fast recognition
and resolution of conflicts between train movements in case of disturbances
by means of real-time speed adaptation, re-ordering or re-routing. The book
can serve as an introduction to the theory of railway traffic, timetable design,
operations analysis, simulation, safety and control for Master and PhD students
from engineering faculties and professionals working in the railway industry.
ISBN: 978-1-84564-500-7 eISBN: 978-1-84564-501-4
Forthcoming apx 208pp apx £79.00

All prices correct at time of going to press but

subject to change.
WIT Press books are available through your
bookseller or direct from the publisher.
 ...for scientists by scientists

Urban Transport XVI

Urban Transport and the Environment in the 21st Century
Edited by: A. PRATELLI, University of Pisa, Italy and C.A. BREBBIA, Wessex Institute of
Technology, UK

This International Conference on Urban Transport and the Environment has

been successfully reconvened annually for the last fifteen years. Transportation
in cities, with related environmental and social concerns, is a topic of the
utmost importance for urban authorities and central governments around the
world. Urban Transport systems require considerable studies to safeguard their
operational use, maintenance and safety. They produce significant
environmental impacts and can enhance or degrade the quality of life in urban
centres. The emphasis is to seek transportation systems that minimize any
ecological and environmental impact, are sustainable and help to improve the
socio-economic fabric of the city. Another area of concern addressed by the
conference is that of public safety and security, seeking ways to protect
passengers while retaining the efficiency of the systems.
The sixteenth conference topics are: Transport Modelling and Simulation;
Transport Security and Safety; Transport Technology; Land Use and Transport
Integration; Intelligent Transport Systems; Public Transport Systems; Road
Pricing; Inter-Model Transport Systems; Transport Automation; Traffic
Management; Urban Transport Strategies; Urban Transport Management;
Environmental Impact, Including Air Pollution and Noise; Information
Techniques and Communications; Mobility Behaviour; Policy Frameworks;
Environmentally Friendly Vehicles; Transport Sustainability; Safety of Users
in Road Evacuation.
WIT Transactions on The Built Environment, Vol 111
ISBN: 978-1-84564-456-7 eISBN: 978-1-84564-457-4
Forthcoming apx 700pp apx £266.00
 ...for scientists by scientists

Power Supply and Energy Management

Edited by: E. PILO, Pontifical Comillas University of Madrid, Spain

In latter years, energy efficiency has become a crucial concern for every
transportation mode, but it is in electrified railways where energy savings have
shown a bigger potential due to (i) regenerative braking, that allows converting
kinetic energy into electric power, and (ii) vehicle interconnection, that allows
other trains to use regenerated power. Power supply and energy management
will continue to develop in the future.
This book gathers under a single cover several papers published in the
Computer on Railways series (IX, X and XI) and focuses on power supply and
energy management. Some of the discussed themes are: modelling, simulation
and optimisation of AC and DC infrastructure, analysis of rolling stock
consumption, and innovative approaches in power supply operation.
This book will be invaluable to management consultants, engineers,
planners, designers, manufacturers, operators and IT specialists who need to
keep abreast of the latest developments in the field.
ISBN: 978-1-84564-498-7 eISBN: 978-1-84564-499-4
Forthcoming apx 192pp apx £73.00

WIT eLibrary
Home of the Transactions of the Wessex Institute, the WIT electronic-
library provides the international scientific community with immediate
and permanent access to individual papers presented at WIT
conferences. Visitors to the WIT eLibrary can freely browse and search
abstracts of all papers in the collection before progressing to download
their full text.
Visit the WIT eLibrary at
http://library.witpress.com
 ...for scientists by scientists

Computers in Railways XII

Computer System Design and Operation in the Railway and Other Transit
Systems
Edited by: B. NING, Beijing Jiaotong University, China and C.A. BREBBIA, Wessex, Institute
of Technology, UK

This volume features the proceedings of the Twelfth International Conference

on Computer System Design and Operation in the Railway and other Transit
Systems. This book updates the use of computer-based techniques, promoting
their general awareness throughout the business management, design,
manufacture and operation of railways and other advanced passenger, freight
and transit systems. It will be of interest to railway managers, consultants,
railway engineers (including signal and control engineers), designers of
advanced train control systems and computer specialists.
The COMPRAIL series has become the world forum for all major advances
in this important field, and this latest conference volume highlights themes of
great current interest. These are: Planning; Safety and Security; Advanced
Train Control; Drivers Operations; Communications; Energy Supply and
Management; Operations Quality; Timetable Planning; Level Crossing and
Obstacle Detection; Computer Techniques; Dynamics and Wheel/Rail
Interface; Maintenance; Rolling Stocks; Training Tools and Technology;
Condition Monitoring; Asset Management; Maglev and High Speed Railway;
Passenger Information Systems; Train Regulations; Metro and Other Transit
Systems; Advanced Train Control.
WIT Transactions on The Built Environment, Vol 114
ISBN: 978-1-84564-468-0 eISBN: 978-1-84564-469-7
Forthcoming apx 1000pp apx £380.00

WITPress
Ashurst Lodge, Ashurst, Southampton,
SO40 7AA, UK.
Tel: 44 (0) 238 029 3223
Fax: 44 (0) 238 029 2853
E-Mail: [email protected]