ADB

DEVELOPING BEST PRACTICES

FOR PROMOTING PRIVATE SECTOR
INVESTMENT IN INFRASTRUCTURE

AIRPORTS AND AIR TRAFFIC CONTROL

Asian Development Bank

The views, conclusions, and recommendations presented here
are those of the study consultants, and should not be considered
to represent the official views of the Asian Development Bank or
its member governments.

© Asian Development Bank 2000

ISBN No. 971-561-282-5

Stock No. 010500

Published by the Asian Development Bank

P.O. Box 789, 0980 Manila, Philippines

For more information on ADB, visit http:\\www.adb.org

 FOREWORD

This report is one of a series of five commissioned by the Asian Development Bank
(ADB) to identify and recommend best practices to be followed and specific steps to be taken, by
ADB’s developing member countries in order to encourage both private sector investment and
competition in infrastructure development. The study was financed through a $600,000 regional
technical assistance grant - RETA 5753: Developing Best Practices for Promoting Private Sector
Investment in Infrastructure. This report focuses on airports and air traffic control; the other
reports cover the power, water supply, road and port sectors.

Transport is central to achieving prosperity and the quality of life, to which all countries
aspire. Efficient and competitive airports and air traffic control are important for economic
development. This report develops best practices for promoting private sector investment in airports
and air traffic control. It examines the underlying economics and alternative models of private
sector participation, and then recommends best practice approaches. In addition, the study
discusses ADB’s role in facilitating private sector participation. It is hoped that the report will
help ADB’s developing member countries attract well-managed and cost-effective private
investment in airports and air traffic control.

The five reports have benefited from the support of and valuable contributions from
many individuals, both inside and outside ADB. The reports were prepared by a team of
individual consultants: Water Supply - Michael Porter of Tasman Asia Pacific; Power - Elliot
Roseman of PricewaterhouseCoopers; Ports - John Arnold, an independent ports specialist;
Airports and Air Traffic Control - Ian Jones of National Economic Research Associates; and
Roads - Roger Allport of Halcrow Fox. In ADB, Sean O’Sullivan, Senior Public/Private Sector
Specialist managed the technical assistance implementation with the help of Marcelo Minc,
Project Economist. ADB staff in the Energy; Transport and Communications; and Water Supply,
Urban Development and Housing Divisions as well as the Private Sector Group helped in
guiding the direction of the study and in reviewing the outputs. In December 1998, a workshop,
hosted by ADB as an integral component of the study, provided a forum for the exchange of
ideas and experiences. Participation and contributions of delegates from many developing
member countries and representatives from the private sector in the workshop were very much
appreciated by ADB.

The publication of the five reports is especially timely as it coincides with the introduction
of a new strategy for private sector development by ADB.

Vladimir Bohun
Director
Infrastructure, Energy and Financial
Sectors Department (East)
 ii

ABBREVIATIONS

AAT – Airports Authority of Thailand

AATCo – Airport Authority of Thailand Company Ltd
ACCC – Australian Competition and Consumer Council
ADB – Asian Development Bank
ADC – Asian Dragons Consortium
AERIA – Aeroport International d’Abidjan
Aerocivil – Aeronautica Civil
ANAC – Agence Nationale pour I’Aviation Civile (National Civil Aviation Agency)
ANAM – Agence Nationale de I’Aviation Civile et de Ia Météorologie, (National
Agency for Civil Aviation and Meteorology)
ATC – Air Traffic Control
BAA – British Airport Authority
BOOT – build-own-operate-transfer
BOT – build-own-transfer
CAA – Civil Aviation Authority
CODAD – Compania de Desarollo Aeropuerto El Dorado SA (El Dorado Airport
Development Company, Limited)
CTRL – Channel Tunnel Rail Link
DMC – developing member country
DOTC – Department of Transportation and Communication
EU – European Union
GDP – Gross Domestic Product
ICAO – International Civil Aviation Organization
OECD – Organization for Economic Cooperation and Development
LGU – local government unit
MCCI – Marseilles Chamber of Commerce and Industry
MIAA – Manila International Airport Authority
MMC – Monopolies and Mergers Commission
mppa – million passengers per annum
NAIA – Ninoy Aquino International Airport
NATS – National Air Traffic Services
NAV CANADA – Canadian Civil Air Navigation Service Provider
NBIAC – New Bangkok International Airport Corporation
NERA – National Economic Research Associates
PIATCO – Philippine Intemational Air Terminal Company Inc.
PRC – People’s Republic of China
PSP – private sector participation
SEGAP – Société d’Exploitation et de Gestion Aèroportuaire
SODEXAM – Service Météorologique National de Ia Côte d’lvoire (National
Meteorological Service of the Ivory Coast)
UK – United Kingdom
US – United States
 TABLE OF CONTENTS

FOREWORD ................................................................................................................................. i

ABBREVIATIONS .......................................................................................................................... ii

LIST OF TABLES .......................................................................................................................... v

LIST OF FIGURES ........................................................................................................................ v

LIST OF BOXES ........................................................................................................................... v

EXECUTIVE SUMMARY ............................................................................................................ vi

A. Key Aspects of Good Practice ..................................................................................... vi
B. The Role of ADB .......................................................................................................... vii

PART ONE : STUDY OVERVIEW

I. INTRODUCTION ................................................................................................................... 1

II. THE GROWTH OF PRIVATE SECTOR PARTICIPATION .................................................... 2

A. Expansion and Contraction of Private Sector Investment ............................................. 2
B. The Challenge for Private Sector Infrastructure Investment .......................................... 3

Ill. CROSS-SECTORAL ISSUES FOR PRIVATE SECTOR PARTICIPATION .......................... 4

A. The Need for Reform and Role of Government ............................................................. 5
B. Institutional Reform ........................................................................................................ 6
C. Strategic Planning ......................................................................................................... 6
D. Legal and Regulatory Framework .................................................................................. 7
E. Unbundling and Introducing Competition ....................................................................... 7
F. Sources of Financing .................................................................................................... 8
G. Risk and Risk Mitigation ................................................................................................. 9

IV. SUMMARY OF SECTORAL BEST PRACTICES .................................................................. 9

A. Power .......................................................................................................................... 10
B. Water ........................................................................................................................... 11
C. Roads .......................................................................................................................... 13
D. Ports ............................................................................................................................ 14
E. Airports ........................................................................................................................ 15

V. THE ROLE OF THE ASIAN DEVELOPMENT BANK .......................................................... 17

PART TWO: AIRPORTS AND AIR TRAFFIC CONTROL REPORT

I. INTRODUCTION ................................................................................................................. 18

II. THE BUSINESS OF AIRPORTS AND AIR TRAFFIC CONTROL ....................................... 18

A. Introduction .................................................................................................................. 18
B. Airport Activities and the Role of Airport Operator ....................................................... 18
C. Airport Operators’ Revenues ....................................................................................... 19
D. Demand and Cost Conditions ..................................................................................... 23
E. The Conditions of Competition and the Need for Economic Regulation ...................... 24
F. Air Traffic Control Services .......................................................................................... 25

Ill. ALTERNATIVE MODELS OF PRIVATE SECTOR PARTICIPATION .................................... 26

A. Introduction ................................................................................................................. 26
B. Alternative Models of PSP ........................................................................................... 26
C. Full and Partial Privatization Compared ...................................................................... 32
D. Air Traffic Control ......................................................................................................... 33

IV. MEASURES TO ENCOURAGE SUCCESSFUL PRIVATE SECTOR PARTICIPATION ...... 34

A. Introduction .................................................................................................................. 34
B. Public Policy ................................................................................................................ 34
C. Industry Restructuring ................................................................................................. 36
D. Regulatory Institutions and Practice ............................................................................ 37
E. Contract Design and Risk Management ...................................................................... 41
F. Securing Efficient Tendering Outcomes ...................................................................... 46
G. PSP in Unprofitable Airports ........................................................................................ 48
H. Conclusions: The Main Lessons from Experience of Airport Privatization to Date ...... 48

V. THE ROLE OF ADB ............................................................................................................ 49

A. Introduction .................................................................................................................. 49
B. Disseminating Good Practice ...................................................................................... 50
C. Role in Training ............................................................................................................ 50
D. Improving Financing Terms ......................................................................................... 50

APPENDIXES
1. Examples of Private Sector Participation at Commercial Airports Worldwide
2. Specific Country Case Studies
3. Extracts from the Philippine Republic BOT Law
 LIST OF TABLES

Table 1: Airport Charges Index (ranking of 40 airports world wide), 1995

Table 2: The Currency of Airport Charges

Table 3: 1995 Traffic Levels at Principal DMC Airports

— million passenger miles per annum (mppa)

Table 4: Alternative Models of PSP

Table 5: Sources of Risk

Table Al .1: Examples of Full Privatization Schemes Worldwide

Table Al .2: Examples of BOT/Concession Schemes Worldwide

Table Al .3: Examples of Strategic Partner/Partial Divestiture Schemes Worldwide

Table Al .4: Examples of Management Contract Schemes Worldwide

LIST OF FIGURES

Figure 1: Average Land-side Revenue per Passenger by Ownership Structure

Figure 2: Key Lessons

Figure Al .1: Contractual Structure

Figure Al .2: Thai Airport Restructuring Proposals

LIST OF BOXES
Box 1: Past Project Finance and Future Infrastructure Demand — East Asia
 EXECUTIVE SUMMARY

This report is one of a series commissioned by the Asian Development Bank (ADB),
identifying best practice for promoting the role of the private sector in financing and operating
infrastructure in its developing member countries (DMCs). The report focuses on the airports
and air traffic control (ATC) sectors; other reports cover the power, water supply, roads and
ports sectors.

The background to the report is that, in common with other international agencies and
commentators, ADB believes there are significant advantages to expanding the role of the
private sector in financing and implementing transport infrastructure and related services in
the DMCs, principally for two reasons:

• First, private sector participation (PSP) may help to overcome constraints on public
sector borrowing, and, equally or possibly even more important, on the public
sector’s capacity to implement efficiently and cost-effectively large-scale
infrastructure programs.

• Second, the active participation of the private sector in all phases of the project life
cycle may secure better value-for-money in the project than the traditional ‘design-
build’ model, where the private sector’s role was limited to the project construction
phase.

The main findings of the report are summarized below.

A. Key Aspects of Good Practice

1. Public Policy — Airport privatization will be encouraged by the existence of

legislation in the form of a Build-Own-Transfer (BOT) law or similar, signaling the
state’s recognition of the need for private participation in infrastructure provision. It
is also important to ensure that government is able to demonstrate that any projects
offered to the private sector are economically viable.

2. Industry Structure — In the absence of any significant scale benefits from multi-
airport operation, there are advantages from using the privatization process as an
opportunity for reducing high levels of industry concentration. Equally, the existence
of unprofitable airports does not justify the maintenance of a highly concentrated
industry structure to facilitate cross-subsidy.

3. Regulatory Framework and Mechanisms — The process of PSP will be

encouraged by the existence of good regulatory governance structures, based on
independent regulatory agencies, operating within well defined public interest criteria,
and with well articulated appeal or arbitration mechanisms. Such a framework is
equally relevant under both full and partial (concession based) privatization
scenarios. The price cap approach to constraining airport charges is likely to
encourage better performance outcomes than one based on rate of return regulation.

4. Risk allocation — Some sharing of revenue or market risks between a

concessionaire and government may offer a better deal for the purchaser than full
transfer of such risks to the concessionaire. Denominating some or all airport charges
in US dollars may be a useful device for encouraging PSP in the wake of the recent
currency crisis. There would be advantages in the widespread adoption of
 vii

mechanisms already present in some concession contracts for compensating

concessionaires for “stranded assets” if the concession is terminated or transferred
to another party when it is rebid.

5. Tendering Procedures — Competition for the market will be encouraged by a

clear and transparent tendering process, based on equal treatment of bidders and
full disclosure of information to enable bidders to make as informed an assessment
as possible of the business opportunity.

B. The Role of ADB

ADB has a potentially important role to play in encouraging the successful

implementation of policies to extend PSP in airports and ATC services.

1. Disseminating Good Practice — First, through the medium of the regional technical
assistance umbrella, ADB can provide decision-makers in DMCs with authoritative
and impartial assessment of the rapidly evolving experience worldwide of PSP in
infrastructure development.

2. Training — Second, ADB can play a role in assisting public sector agencies in the
region to develop the capabilities required in contract negotiation and in regulation
to make BOT and concession agreements work successfully. It can do so directly
by providing training courses; and through measures encouraging development of
indigenous training capabilities, such as assisting in the establishment of training
centers and facilitating secondment of staff in DMC agencies to agencies in other
countries with greater experience of privatized infrastructure provision.

3. Improving Financing Terms — Finally, ADB can work to improve the financial
terms on which the private sector is willing to provide airport and ATC infrastructure
services in DMCs:

• by encouraging the corporatization of existing state owned providers of airport

and ATC services, so that potential private sector investors are better able to
understand the business proposition in which they might be engaged;

• by promoting good practice in regulatory governance and public procurement;

• by promoting good practice in contract design, to encourage the efficient

allocation of risk between the DMC government agencies and private sector
contractors;

• by assisting DMC agencies to bring forward economically viable projects which

attract widespread support from relevant constituencies within the DMC;

• by investing in or sponsoring a project, ADB can signal to private investors its

views on the soundness of the project, which, in turn, should improve the terms
on which external investors are willing to participate.
 PART ONE

STUDY OVERVIEW
 I. INTRODUCTION

An Asian Development Bank (ADB) regional technical assistance was approved with the
aim of developing sector specific best practices for promoting private sector participation (PSP)
in key infrastructure sectors in ADB’s developing member countries (DMCs). The sectors studied
included power, water supply, roads, ports and airports and the best practices covered: (i) sector
policy issues relating to pricing and competition; (ii) conducive legal and regulatory frameworks;
(iii) the unbundling, mitigating, and management of risks; and (iv) mechanisms to reduce
transaction costs. Five individual experts were engaged to undertake the study, one for each
sector. A two-day regional workshop was held at ADB on 9-10 December 1998 for the experts
to present their findings and validate them with an invited group of experienced senior
government and private sector individuals, together with ADB staff. These volumes represent
the final outputs of the study.

A summary of the expressed views in these volumes in relation to preferred forms of PSP
in infrastructure, informed by the currency crisis, is that it is “best practice” to have a customer
focus and a well structured regulatory environment around infrastructure projects, in part since
this can allow domestic financing. In other words, it is financially and economically sensible to
utilize the essential and often monopoly status of efficient infrastructure services in creating, in
effect, a customer finance model of PSP. Under this customer-focused concession or franchise
model, government provides the regulatory and legal framework that can satisfy customer and
investor alike, with the securitization of customer accounts (say via an escrow account) or insurance
techniques underpinning financing arrangements. Investors will always seek to mitigate
uncertainties, but many of the privatization models to date have done so by way of government
guarantees which have undermined the process in the longer run.

Regulation by entities appointed by the government is still required in the new model, given
that monopoly provision of key network assets is often the only efficient option. For example there
is a need to regulate access charges for connection to network assets such as pipelines, high
voltage wires and port channels. But where competition can be achieved in the product market,
as with electricity generation selling into a power pool, then this competition is generally the best
mechanism to achieve good outcomes for customers. Realistically, in much of Asia, there is little
experience with these new pro-competitive models of regulation and thus there is an expectation,
on the part of the experts, of a substantial phase-in to this regulatory element of best practice in
the future.

The challenge as we enter 2000 with its information-rich possibilities, is to learn from the
1990s infrastructure experience on investor-to-government build-operate-transfer (BOT) deals
and concession transactions so that DMCs can benefit from the adoption of best practices in the
various infrastructure sectors.

The following presents an overview of the study, including a discussion on the growth of
private sector infrastructure investment in Asia, a review of the cross-sectoral issues, a
summary of the sectoral best practices for each sector and suggestions on the role of ADB in
supporting private sector investment in infrastructure. Part 2 comprises the specific sectoral
report.
 2

II. THE GROWTH OF PRIVATE SECTOR PARTICIPATION

A. Expansion and Contraction of Private Sector Investment

The last decade, and notably the period to 1996, saw both the rapid expansion of private
investment in public infrastructure and a sharp increase in private management of the services
associated with this infrastructure. The investment was fuelled by the development of new
forms of PSP including varying forms of public/private partnerships: BOT, build-own-operate, build-
own-operate-transfer (BOOT), and concessions.

New financial instruments, especially project finance, and the globalization of private
investment funds, played a major role in the expansion of the infrastructure sectors in most
countries. PSP in infrastructure, and in particular power generation, was supported
enthusiastically by the multilateral development banks and bilateral development agencies; as
well as by the international financial community. But fewer transactions were completed in the
more complex and customer-focused areas such as water, electricity distribution and transport
infrastructure. Early successes involved financial transactions without major organizational
restructuring; later transactions focused on major infrastructure in mega-cities such as Manila,
Jakarta and Shanghai. For example, water treatment plants, bulk water supply, individual power
generation units, container terminals, passenger terminals, and airport toll roads.

In the first half of the 1990s, investment requirements for infrastructure in Asia were
seen to be on a scale that dwarfed earlier projections and experience. Asian tiger economies
were growing rapidly, and demanding massive investments in power, roads,
telecommunications and other infrastructure. In most Asian economies, there was also a sense
that development was being hindered by bottlenecks in power (e.g., the Philippines), transport
(e.g., Thailand), water (most of Asia) and telecommunications. Since government infrastructure
spending, international aid, and official sector lending could not be on a scale sufficient to
meet requirements, the private sector was the focus of attention.

The new infrastructure investment requirements were estimated by ADB to be of the

order of US$1,000 billion for the 1990s for East Asia. Subsequently, they were estimated by
the World Bank to be of the order of US$1,500 billion for the decade 1995 to 2004. Such
projections were useful as a means of highlighting the scale and structure of the huge
infrastructure requirements of a growing and increasingly prosperous and urbanized Asia.
They helped make clear the need for a major shift of focus towards PSP in infrastructure, to
some extent motivated by efficiency considerations, but mainly reflecting the view that public
sector financing for this scale of infrastructure requirements was neither feasible nor desirable.

There had also been a shift in views as to the comparative advantages of governments
and the private sector in performing the various roles related to the provision of quality
infrastructure services. Increasingly, an expanded regulatory and restructuring role was seen
for governments, with investment, construction, financing, and management viewed as best
opened to competitive PSP. Risks should, under this approach, be assigned to the parties
best able to mitigate them, and this meant a greatly expanded role for the private sector.

There was recognition that while many private sector investments of the BOT type
were being completed, the assignment of risks in many of these projects left much to be desired.
Government guarantees of bulk take-or-pay contracts (between utilities and investors), often
 3

indexed to exchange rates, had created huge contingent financial obligations of the utilities and
their governments.

As with many investment trends, optimism, a proliferation of Memoranda of Understanding

and glossy investment announcements gradually replaced careful evaluation. Some early
successes, under special circumstances, led to the assumption that this BOT approach could be
universally applied. The expression BOT had become a shorthand for PSP in many countries by
the mid 1990s; but by 1999 BOTs and often the associated power purchasing agreements had
also become a shorthand for unacceptable government risk exposure, and of project isolation from
customer and market pressures.

This optimism ended with the Asian financial crisis; itself brought on by a lack of sound
investment policies, in particular, in relation to government guaranteed power purchasing agreements.
The power purchasing agreements had inadvertently converted a shortage of power supply into an
oversupply, secured by take-or-pay guarantees. The result of the crisis has been a sharp contraction
in private sector investment and a significant exposure of government and private sector investors
to contingent liabilities. This contraction not only limits the capacity of governments to stimulate
economic growth but also has led to the deterioration or stagnation of many partially completed and
privately financed public infrastructure projects. The rise and fall of private sector finance is clearly
shown in the private finance data presented in Box 1.

The currency crisis has caused some dramatic revisions both to economic growth forecasts
and to infrastructure investment programs. However, as the analysis in Box 1 shows that while
forecasts for infrastructure are lower due to lower growth and the expected move to best practice,
the magnitude of investment is still huge and efficient PSP will be required.

B. The Challenge for Private Sector Infrastructure Investment

As this difficult period unwinds, it is important to re-consider the comparative advantages of

the public and private sectors and the critical role of improved regulation and governance —including
transparency, enforcement of contracts, and the adoption of viable commercial tariff structures.
There is a need to review, sector by sector, the strengths and weaknesses of the process that has
been used to implement these investments. The opportunities and risks of new approaches need
to be addressed — e.g., the case for expanding the emphasis on customer focused and privately
managed concessions. There is a need to develop bankable versions of these models, ideally
backed by the security of customer accounts rather than government guarantees or public sector
assurances. This series of volumes addresses these and other sectoral best practice concerns.

There are major challenges for governments and investors alike, emerging from this shift
to a new model for infrastructure development. The new best practice model does not mean a
total retreat by governments; on the contrary, moving to best or better practice involves a shift to
good governance, and requires an upgrade of regulatory, restructuring, and monitoring roles.
Without greatly improved governance, the shift to increased PSP could just mean monopoly
powers being shifted to the well connected in the private sector. Moreover, without improved
governance, PSP would eventually flounder and the demands for infrastructure will not be
met, as risks would become unacceptable.
 4

Box 1: Past Project Finance and Future Infrastructure Demand — East Asia

Asia Project Finance — Opportunity and Volatility

Figure 1:
Figure 1 draws on a Euromoney (CapitalDATA)
database and highlights the dramatic growth, and Project Finance
subsequent decline of infrastructure funded through 45
project finance in selected East Asian countries. The pre- 40 Energy
crisis level of nearly US$41 billion for 1996, contrasts Telecoms
sharply with the estimated level at the end of the 1980s, 35
Transport
when the total market for funding projects was less than 30 Water

US$ billion
US$5 billion per annum, as well as with the crisis figure
25
of US$12 billion for 1998. Clearly, in the 1990s and well
prior to the crisis, the importance of the private sector in 20
infrastructure development was rapidly increasing. As a 15
result of the crisis, the telecommunications sector has
10
shown the most dramatic decline, reflecting the fact that
such projects are typically purely privately funded, and 5
bear demand risk in a newly open environment. The 0
energy projects, on the other hand, appear more resilient, 1995 1996 1997 1998
but mainly because they have had some form of
government support, in the form of guarantees in relation YEAR
to bulk sales through PPAs.

Future Demand for Infrastructure Investment Figure 2:

1996-2005 Infrastructure Investment Projections
New infrastructure projections for selected East
Asian countries: the People’s Republic of China (PRC); Scenario US$ trillion %
Indonesia; Republic of Korea; Malaysia; Philippines; and
Thailand for the period 1996-2005, adjusted to allow for Baseline 1.78 100
both the phase-in of private sector market discipline/best Case 1 1.53 86
practices and reduced economic growth. The revised Case 2 1.37 77
projections are 23 percent below the pre-crisis (baseline)
projections. They are based on establishing the value of
the capital stock of infrastructure in each country and
projecting infrastructure investments with varying gross
domestic product (GDP) growth assumptions and varying
infrastructure-to-output ratios. A summary is given in
Figure 2. The pre-crisis projections are based on the 1996
GDP growth forecasts. Case 1 is based on the current
GDP growth forecasts while Case 2 adds the impact of a
transition to a lower infrastructure-to-output ratio and
assumes a gradual 25 percent increase in efficiency in
each sector in each country. An important factor to note in
the projections for this region is that the PRC is assumed
to maintain its relatively high GDP growth rate, which
accounts for about two thirds of the infrastructure
spending in the region. The results for Case 1 indicate a
fall of 14 percent from the pre-crisis projections. If the
PRC is excluded, the reduction is 33 percent. Case 2,
which assumes a transition to best practices, with a
resulting change in the underlying infrastructure-to-output
and efficiency parameters, indicates further reductions in
the level of needed investments. The analysis clearly
shows the relative impact of lower growth and the potential
benefits of moving to best practice models of infrastructure
development. It also highlights the magnitude of
investment requirements, in excess of $120 billion per
year, and the need for PSP.
Source: “Private Sector Participation and Infrastructure Investment in Asia”, Asian Development Bank paper prepared for the
Finance Ministers Meeting, Asia Pacific Economic Cooperation, May, 1999; prepared by M.G Porter and C. McKinlay (Macquarie
Bank and Tasman Asia Pacific).
 5

III. CROSS-SECTORAL ISSUES FOR PRIVATE SECTOR PARTICIPATION

There are a number of cross-sectoral issues relating to promoting private sector investment
in infrastructure that were identified during the study. The review of best practices in each of the five
sectors highlighted the importance of competition, transparent tendering, and effective regulation.
There was broad agreement that:

• Government should specialize in planning, structuring, and regulation while the private
sector should specialize in management, investment, construction, and financing;

• The transfer of responsibility to the private sector should be accomplished through

deregulation and open competition or well-established contractual arrangements including
management contracts, capital leases, concessions, sale of assets and rights to operate;

• Economic regulation should be applied where there is insufficient competition but it

should be transparent and predictable while still accommodating the concerns of the
affected parties;

• Long-term domestic financing sources must be developed; and

• Commercial risks should be assigned to the private sector but other risks should be
assigned according to which party is able to mitigate the risks.

The cross-sectoral issues are discussed in more detail below.

A. The Need for Reform and Role of Government

PSP in infrastructure development still requires the government to play a key role in planning,
policy, and regulation. The reason that infrastructure industries have remained so long in the public
sector is that they have components that are natural monopolies; e.g., the costs are lower with only
one provider and the services are often essential (water, power and transport). These infrastructure
monopolies also typically have a relatively high proportion of capital costs, have long-lived assets
with low unit variable costs, and exhibit significant economies of scale. It had been a common
judgement that state ownership of such monopolies, rather than state regulation of privately owned
assets, was likely to deliver the best outcomes.

Existing service providers in these infrastructure areas have also had a considerable
competitive advantage over potential new entrants, because of the relatively long time required
to construct expensive new networks and to build up the market for their services. The scarcity
of land, rights-of-way and airspace suitable for development of the network also act as an
additional barrier to competition. Sites for airports and seaports, dams, power plants, and
rights-of-way for roads, rail lines and transmission systems had become increasingly difficult
to acquire. Another common argument for retaining these industries within the public sector
was that they must provide common (or universal) access to their services and that subsidies
are required.
 6

It turns out that public ownership and management is neither necessary, nor the best way
to ensure universal access. Subsidies can easily be a requirement of a competitive tender or can
be directly financed by government. A key advantage of having the private sector provide public
services is that it allows public administrators to concentrate on planning, policy and regulation.
The private sector, in turn, is empowered to do what it does best (i) invest capital; (ii) manage the
businesses; (iii) manage and create appropriate incentives for staff and management; (iv) deal
with customers; and (v) improve the efficiency and quality of service; more recently, under the spur
of benchmark competition - competition by comparison.

Governments should allow the private sector to provide infrastructure services to the
maximum extent possible, with governments concentrating on planning, policy and regulation, and
with the private sector on efficiently investing capital and improving the efficiency and quality of
such services.

B. Institutional Reform

The organization of the infrastructure sectors (i.e., ministries, regulatory agencies, and
utilities) has remained largely unchanged with the introduction of PSP. With financial transactions
being the primary mechanism for transferring infrastructure services to the private sector, insufficient
attention has been given to the broader issue of institutional reforms. It has been implicitly assumed
that the introduction of private management into the ownership or operation of specific assets
would obviate the need for such reforms. Instead, the weaknesses of existing institutional structures
have limited the effectiveness of the private sector initiatives. In most countries, the piecemeal
transfer of infrastructure components has proceeded slowly and the controlling bureaucracies that
add overhead costs and often limit improvements in infrastructure performance, have remained
relatively unaffected. The currency crisis has emphasized the importance of institutional reforms
but government bureaucracies rarely reform themselves. Governments should carefully review
the structure, size and responsibilities of state-owned utilities and other entities in the infrastructure
sectors and establish special reform units reporting directly to top level ministers to spearhead the
necessary reforms.

C. Strategic Planning

Governments’ acceptance of private sector investment in infrastructure has been due, in

part, to their failure to anticipate future bottlenecks and make timely strategic investments to
prevent shortages in capacity. The increased role of the private sector in developing
infrastructure has caused many governments to neglect their responsibility for sector planning.
Instead, governments have offered assets and public services to the private sector in an ad
hoc manner, often failing to ensure that individual investments were complementary. In certain
circumstances, unsolicited proposals have been used as a surrogate for planning. For its
part, the private sector has selected projects that had already been identified in government
plans, giving preference to those which offered the highest rate of return, the lowest risk or the
greatest short-term benefit. The private sector has had neither the interest nor the capacity to
consider the network implications of its proposals. Governments have failed to subject these
proposals to rigorous financial analysis to determine their sustainability in the absence of major
increases in user charges or government guarantees. Governments have also often overlooked
the complementary investment required from the public sector to make the private investments
successful. The results have been unsolicited proposals that involved little commercial risk
(government guarantees, wrap-around provisions, transfer of existing assets, granting select rights
of way) or politically generated proposals. Governments should maintain and strengthen
 7

their role in strategic planning of the infrastructure sectors and in the process identify where PSP
should be encouraged and the level of complementary support that should be provided.

D. Legal and Regulatory Framework

The effectiveness of PSP has suffered from the lack of adequate regulatory structures to
control both technical and economic performance. Regulation of tariffs and other economic factors
is particularly undeveloped. The basic objectives of autonomy, accountability, transparency and
predictability have been difficult to achieve. More importantly, the mechanism for consultation between
the public and private sector and for dispute resolution between the providers and users of the
network has not been fully developed. A further problem has been the failure to separate regulation
from administration in order to avoid conflicts of interest. Most countries have been slow to establish
autonomous regulatory agencies with independent funding and professional staff.

Unbundling the network into competitive and monopolistic components can significantly
reduce the need for regulation. The competitive components can be transferred to the private
sector in a way that promotes competition and allows deregulation. The monopolistic components
can then be transferred to the private sector once an effective regulatory framework has been
established. This regulation should create a situation where the businesses derive their profits
from increased efficiency and the attraction of additional demand.

Effective economic regulation covers also deterrence of anti-competitive practices. Most

of the developing countries lack laws or agencies for dealing with anti-competitive practices.
Economic activity continues to be concentrated in large conglomerates. The currency crisis has
provided new impetus for breaking up the monopolies and introducing anti-monopoly laws.

The lack of established legal and regulatory procedures applies to contract law as well. The
means for enforcement of contracts and the resolution of disputes are not well established. Political
interference in the award of contracts has also been a problem.

PSP without a well-developed legal and regulatory framework increases the level of risk to
investors. It also encourages investors to rely on special situations and political relationships rather
than their merits as a means for securing and implementing contracts. The transfer of infrastructure
services to the private sector should not lead to privileged deals or profits secured by government
guarantees. They should be businesses with regulated income streams which derive their profits
from increased efficiency and the attraction of additional demand. These income streams should
be capable of securing substantial private sector funding, both because their semi-regulated nature
makes them much like a government bond, and because the essential and often monopoly nature
of the service lowers demand risk. Such assets are also long-lived and thus attractive to pension
and similar long-term funds.

E. Unbundling and Introducing Competition

Experience in a number of countries has shown that unbundled infrastructure sectors with
individual components managed separately can perform better than centrally-controlled networks.
The additional costs of unbundled networks due to increased communications and transactions
among components have been reduced by improvements in technology. At the same time, the
unbundled management has been able to better focus on the capacity and productivity of the
individual components and their interface with other components.
 8

The unbundling of the infrastructure sectors is an important technique for reducing their
natural monopoly and promoting competition. Many parts of the network can support competition.
Where it is not possible to create direct competition between suppliers of network services, it is
often possible to create competition among providers of complementary network services. For
example, in the power sector, many countries are separating the networks into generation,
transmission, distribution, and in some cases, a fourth segment responsible for retailing power to
customers, with different companies responsible for each segment.

Where competition cannot be created, it is often possible to establish contestable

environments e.g., a market for the business. One method is through effective competitive bidding
for the sale or lease of assets and licensing or franchising of services. Another is to reduce the
period of the contractual agreements or to provide for a periodic review of performance. A third is to
introduce performance targets related to the quality of the service, the range of services, the prices
charged for the services and overall market share. The ability of the private sector to achieve these
targets is then linked to penalties, or provisions that may lead to early termination of the agreement.
A fourth method is to require comparable performance vis-a-vis other networks. This may be in the
form of requirements for increasing market share relative to other providers of similar services, or
requiring a quality of service and price that is comparable to other networks serving similar markets.

Most infrastructure sectors are composed of profitable and unprofitable components. One
practical, but not ideal, strategy for transferring the components to the private sector is to bundle
profitable and unprofitable components to produce a combination that has an acceptable level of
profitability. Another is to tender the profitable components through techniques ranging from operating
agreements and franchising to sales of assets and to transfer the unprofitable components using
management contracts; in effect, bidding out the government support for that component. A third
strategy has been to transfer the profitable components to the private sector and to retain the
unprofitable components in the public sector, but under control of local government units rather
than the national government.

F. Sources of Financing

Private sector funding of infrastructure usually brings the risk of foreign currency mismatches
in the financing package; income is in local currency, but the need to resort to foreign debt and
equity markets means that debt service requires substantial foreign currency. The root problem is
inadequate depth in capital markets in most DMCs which prevents a tailoring of local currency debt
to long-lived assets. The need to resort to foreign debt (and equity) creates substantial risks, which
have been exposed in the recent crisis. Few infrastructure consortia can withstand an exchange
rate depreciation of 40 to 50 percent, let alone the 80 percent decline experienced in Indonesia
when their product is sold for local currency. Hence the priority on programs to deepen the domestic
capital market.

In principle, currency matching requires that the bulk of debt funding of infrastructure
services such as transport, water supply, electricity and other urban services should be in
local currency. In the absence of the necessary capital market reforms, it is hard to see how
private sector provision of infrastructure can proceed on the scale required to meet future
demand. A priority, therefore, given the recent experiences, is that international development
agencies such as ADB expand their role both in facilitating political risk insurance and in fostering
the development of domestic capital markets in Asia, particularly bond markets.
 9

Direct foreign investment will remain an important source of funds for the development of
the infrastructure sectors. However, it will take time to restore investor confidence and, given the
experience of Indonesia, Pakistan, Philippines, Republic of Korea and others, governments will
naturally seek to limit their exposure to these funds in preference to local sources of capital, if
possible. The development of domestic long-term capital markets will be critical for private sector
investment in infrastructure, but these markets must have much better regulation as well.

G. Risk and Risk Mitigation

In order to reach financial closure, governments have often accepted commercial risks
that should have been assigned to the private sector. This includes not only the foreign exchange
risk but also demand/traffic (volume) risk. The most obvious example has been the take-or-pay
provisions in power purchase agreements. These guarantees have had three negative impacts.
First, they have isolated the private sponsors from the influences of the market. Second, they have
created a large amount of contingent liabilities for governments that now add to their fiscal problems.
Third, they have encouraged price rigidity leading to distortions in the market and reducing the
potential of the private sector to improve efficiencies in investment and operations. Other examples
are build-lease-transfer agreements and volume guarantees for toll roads, airports and seaports.

Because governments have had limited contract-related knowledge or experience, the private
parties have been frequently able to convince them to assume some of the commercial risks. Also,
because governments have often not been able to engage suitable legal, technical and financial
experts to assist during negotiations, they have been at a disadvantage in arguing with foreign
proponents concerning international practices such as take-or-pay contracts, or with international
lenders concerning guarantees to protect their loans. Bureaucrats who have gone through a long,
often contentious bidding process have been willing to accept some commercial risks during
negotiations rather than to face rebidding. Alternatively, private parties frustrated with drawn out
negotiations and the continuing renegotiating of clauses have accepted risks that should have
been borne by the government.

Governments should build up capacity to negotiate and deal with the private sector.
Commercial risks should be assigned to the private sector and other risks should be assigned to
the party best able to mitigate them.

IV. SUMMARY OF SECTORAL BEST PRACTICES

The challenge for governments is to encourage an appropriate form of private sector

investment in infrastructure. The study has identified significant differences among the
infrastructure sectors concerning the appropriate balance between private and public
participation in ownership of assets and provision of services. Only some of the sectors have
well defined models for PSP. Other best practices are still evolving and the menu will continue
to develop as experience grows. The decisions on which infrastructure components should
be transferred to the private sector are of a strategic nature. They depend not only on the
characteristics of the sector and the market it serves but also on government objectives. There
was consensus among the experts that the primary objective should be to benefit consumers.
However there were a number of additional objectives which governments should consider: (i)
reduction in national debt; (ii) stimulation of domestic capital markets; (iii) reduction in capital and
operating subsidies; (iv) investment in new infrastructure or rehabilitation of existing
 10

infrastructure; (v) improvements in the quality of service; (vi) increased range of services;
(vii) reduced prices for services; (viii) client-oriented operations; and (ix) more effective
marketing.

Governments have at their disposal a number of means for effecting the transfer of
infrastructure components to the private sector. The pace and sequence of such a transfer depends
on the: (i) size and complexity of the infrastructure sector; (ii) rate of growth in demand and the
competitiveness of the market; (iii) options for unbundling by function or geography; (iv) legal regime
regarding ownership of land and other critical assets; and (v) capacity for economic regulation. The
established mechanisms, which range from management contracts to unregulated competition,
are not new and have proven effective. The key is to have a vision of where the sector is going, and
to carry through the reforms as quickly as possible so as not to allow the interim change to become
the final state of affairs. The findings of the sectoral experts for each sector are summarized below.

A. Power

In the electricity sector, IPPs provided a quick solution (in the Philippines, for example) by
offering generation capacity needed for rapid economic growth. However, the costs were often
high because the new capacity was not consistent with the least-cost expansion path and the
private sector required high rates of return. However, these costs have been decreasing as the IPP
market has matured. The focus on production rather than efficient distribution put the public sector
in the position of retaining that activity in which it was least effective and restricting the private
sector from performing the customer focused activities (distribution and supply) where it had real
expertise. At the same time, it isolated the private sector from the market through a combination of
regulated pricing and guarantees against commercial risks.

The power sector expert advocates restructuring to achieve a competitive market model
with wholesale and retail competition. Such reform will encourage sustainable PSP and maximize
the benefits to consumers. The expert suggests five major steps in implementing this approach,
and their order of precedence. To some extent, these steps may proceed in parallel, but they
should be considered sequential actions that will lead to the implementation of a competitive power
market:

1. Getting the investment framework right.

2. Deciding on the goals of restructuring and the ideal industry structure.
3. Preparing the players to participate in a competitive market.
4. Privatizing existing and new assets.
5. Ensuring that the competitive market is implemented properly.
 11

Best practices for power sector restructuring would include the following:

• Create an enabling legal and regulatory environment to support competitive markets in

electricity.

• Unbundle the power sector into separate generation, transmission, distribution, and
possibly retailing sectors to achieve the maximum benefits for customers.

• Privatization should include the sale of power distribution utilities as well as generation,
and should include existing assets as well as new projects, using a transparent process.

• Open access to transmission and distribution wires, and the ability to trade power
between buyers and sellers in an open market, are critical to achieve a competitive
framework.

• Operate the generation and retailing markets competitively, with a large number of
generators selling into a wholesale electricity market at prices which balance demand
and supply throughout the day.

• Operate the transmission network as a concession on the basis of competitive bidding,

or privatize it within a tight regulatory framework, controlling rates of return, prices or
gross revenue.

• The independent regulator should mainly oversee prices and incentives for transmission
and distribution operations.

• Restructuring should proceed at a pace consistent with the development of a competitive

and unbundled system.

B. Water

The water sector has moved more slowly towards private sector investment, relative to
electricity and telecommunications for example, not least because of the jurisdictional, environmental
and sensitive social concerns about water supply, and its affordability. While major private sector
involvement has now been achieved in distribution (Manila and Jakarta), the bulk of transactions
were BOT models with take-or-pay clauses guaranteed by governments. Adding to these difficulties
was the lack of knowledge about the location and condition of the (underground) networks and
aquifers in many countries.

The volume on the water supply sector addresses the question of why, given the alternatives,
the private sector should seek to invest in a sector with so many uncertainties, natural, governmental
and financial. Water, unevenly supplied as rainfall, is often wrongly deemed a free public good,
despite the costs of treatment and retail supply. Thus, there is often an ill-informed community
constraint against private sector involvement in water supply, which in most countries has prevented
the sorts of best practice referred to in this report.

The water expert makes the point that when it comes to best practice in the case of
water supply, most of the messages are for government — to install sound and independent
 12

regulatory regimes, catchment management policies and enforceable laws on tariff setting and
collections. Once in place, best practices such as water supply concessions can be implemented.
If not in place, then best feasible practice may simply relate to contracting out some services under
government guarantee, or BOOT bulk supply to public sector water supply companies. It follows
from this that since the particular features of the water supply situation and regulatory and privatization
policies differ greatly across countries, so, too, will the feasible best practice.

One misunderstanding regarding the scope for bringing commercial practices to water
supply is the issue of affordability. The report notes that the poor often pay more for water than the
cost from efficient commercial piped supplies. Experience has shown that low-income families will
pay for quality water supply — and are not averse to PSP — if it delivers.

The key points recommended were:

• The benefits of PSP in the water sector must be explained to win public acceptance.

• The starting point in any reform process for water supply is to form a high-level reform
unit to drive and manage the process. It would be responsible for coordinating and
facilitating the entire reform and PSP process. The reform unit may be a crosssectoral
unit.

• While not essential to commence reform, the introduction of tradable water rights leads
to efficient use of water, particularly when it is scarce and has alternative uses.

• The water sector should be unbundled to the extent possible. The private sector
concession model is most likely to achieve the greatest benefits to the community and
the economy as a whole. The government continues to own the network while the
private operators lease the long-term right to use the assets and collect revenue from
service delivery. The benefits accrue due to strong financial incentives to reduce water
losses and expand service.

• If politically difficult, then the next best strategy is to use BOT, BOOT, and rehabilitate-
operate-transfer arrangements to bring expertise and finance to urgently required water
supply projects. The bidding procedure should be carefully managed to ensure reasonable
cost and the contractual arrangements should not constrain subsequent progression
to more competitive models.

• Commercialization/corporatization of water supply utilities together with tariff reform is

advantageous as an interim step if the introduction of PSP is to be phased.

• Tariff reform to achieve full cost recovery is essential for PSP. Cross-subsidies for the
poor can still be considered in a transparent manner.

• Critical to the success of PSP in the water supply sector is for the government to
create sound and independent regulatory regimes, catchment management policies,
and enforceable laws on tariff setting and collection.
 13

• Risks are likely to vary between countries and even between different water utilities in a
country. They should be managed by the party best able to minimize and manage each
risk most effectively. Where no party has a clear comparative advantage to manage the
risk, it should be shared.

C. Roads

In Asia’s roads sector, PSP has been equated with major BOT toll roads. These have been
targeted where traffic is greatest — in and near the capital city and sometimes along major inter-
city corridors. This private investment has produced some successes but also many failures. After
more than a decade of concerted effort, implementation experience has not matched expectations.
Indeed, surprisingly little has been implemented outside the PRC.

The road sector expert has advanced three reasons for modest progress in roads. First,
governments have not defined their policy, often leaving the private sector to identify projects.
Secondly, almost everyone involved has expected such toll roads to be profitable without government
support, but this has only rarely proved to be the case (outside the dense PRC market, which is
deemed a ‘special case’). Thirdly, it has proved difficult to introduce promised tariffs and tariff
increases in a sector where roads have become to be regarded as free.

What is clear is that private construction and maintenance of public roads produced better
results where there was adequate competition and effective methods for enforcing contracts. Efforts
to substitute private sector management for public sector officials in the management of the public
network are in their early stages, even in the developed economies, but the preliminary results are
encouraging.

Worldwide experience identifies a broad range of PSP modalities, in which BOT is close to
being the most difficult to implement. Other modalities include maintenance management contracts,
turnkey, operate, and maintain or rehabilitate-operate-transfer concessions. Many of these modalities
target improved maintenance, and rehabilitation of the network (rather than solely network capacity
expansion). They have potentially much greater application than BOT projects. Looking ahead, the
requirements are to both improve the BOT process, and to extend the modalities that are applied.
The key points to emerge are:

• Governments must prepare the PSP environment. Institutions may need to be

restructured with the objectives of controlling the PSP process in the public interest,
and creating a regulatory body, separate from vested interests. A sound legal framework
and a predictable regulatory regime are essential.

• Governments must identify priority PSP projects. This will almost always require an
independent feasibility study, which focuses on traffic and tariff policy, project staging,
network integration issues, risk allocation, finance and implementation issues.

• The best prospects for BOT projects are in middle-income countries (where the
willingness-to-pay tolls exist) along existing congested corridors, or where there
are missing links (e.g., estuarial/river crossings). A regulated income stream from
a tolled public toll road is capable of securing project financing of an appropriate kind
(i.e., suitable to pension funds and other long-term investor groups).
 14

• Private sector modalities other than BOT exist, e.g., concessions, and should be applied
more widely, as they can address many of the sector problems, and in the process
create a new high growth industry for transport management companies.

• Traffic risk is the major risk and may be shared. The core risk being taken by the private
sector, with government taking a share of the upside benefit and providing a downside
guarantee in the event of low traffic.

• Transparency and competition are essential in the procurement process.

• Government support should be defined upfront as a maximum so that the private sector
can prepare realistic bids.

D. Ports

In the port sector, the transfer of cargo-handling activities to the private sector has been, in
most cases, extremely successful in replacing inefficient government bureaucracy with
commercially-oriented management. Improvements in productivity and maintenance has increased
the quality of service. However, where there was no competition, these arrangements were less
likely to sustain these improvements. Private investment in port infrastructure has generally been
limited to new and existing cargo terminals. Trans-shipment terminals were the most successful,
since they were less dependent on local markets and land transport. Greenfield ports were slower
to develop because they were further from their markets and the transport access was less
developed. Basic infrastructure offered few opportunities for full cost recovery.

The ports sector expert, noted that the private sector has always been actively involved in
port affairs. The land and water transport services that use the port are almost entirely private
sector. Nearly all of the cargo shipped through ports is privately owned. The private sector provides
an array of complementary trade facilitation and logistics services for this cargo. Within the confines
of the public port, cargo owners, forwarders, and ship agents actively participate in decisions
concerning the handling and storage of cargo. The public sector’s role is to own, develop, and
manage basic port infrastructure and common-user facilities.

The process of port privatization has rarely involved pure privatization, since land and
infrastructure are rarely sold. Instead, the process involves PSP in operations and investment in
equipment and facilities. The process is not a monolithic effort because of the diversity and complexity
of ports and the services they provide. It can be divided into three components: (i) institutional
reform, (ii) divestiture of existing services and assets, and (iii) investment in new facilities and
services. These can be implemented individually or in combination. For each port component,
there are many possible public-private partnerships. The main points regarding moves to best
practice were:

• The bidding process should encourage unbundling not only of the network but also
for the services within the ports. Where ports are not financially viable, they should
not be bundled with profitable ports, but treated as stand-alone facilities that are
turned over to local government or put under management contract using a
competitive tender.
 15

• The landlord model is the best structure for promoting PSP because it accommodates
different forms of public-private partnership while recognizing that the only fixed
responsibility of the public port is the ownership of the site.

• The most effective and efficient procedure for promoting PSP in the port sector is to
lease existing facilities with relatively short-term agreements that allow for reorganization
and improvement in productivity. Subsequently, concession agreements can be used
to encourage private investment in additional capacity. Where this capacity is required
immediately, or labor problems make it difficult to lease out existing facilities, then
concessions might precede lease agreements.

• Continued public investment will be required, as it is difficult to recover the costs for
basic infrastructure in a time period reasonable to the private sector. Public investment
may also be required to reduce the barriers to entry. This is important where a new
entrant would otherwise have to make a large investment before competing with existing
service providers.

• The best form of tariff regulation is market regulation; the second best is through the
terms of the contract that identify the non-competitive services requiring regulation,
state the maximum rates, the formulae for escalating these rates over time, and the
arbitration procedures for discriminatory behavior in excess of that justified by commercial
pricing. The third best is the establishment of a regulatory agency outside of the port
which would apply a pricing formula related to cost recovery. All of these are preferable
to a vague procedure for negotiating future changes in tariffs.

• The private sector should assume all commercial risks. Other risks should be negotiated,
based on which party has the capability to mitigate the risk.

• The critical element in any effort to promote PSP is competition, or at least the potential
for competition. This can be provided through direct competition between private sector
service providers, between public and private service providers or between bidders in
the case of an activity that does not allow competition.

E. Airports

For the airport sector, PSP in terminal operations produced significant improvements in
financial performance and the quality of service. Private sector investments have increased
substantially over the last five years. During the previous twenty years, there was little capital
investment in airports, despite a five-fold increase in traffic. The airports coped with the higher
levels of traffic through a combination of larger aircraft, better air traffic control, improved runway
design, and the addition of second runways and additional terminal space. This period has now
ended and most countries need to invest in new airports. These are proving to be costly, complex
and often controversial investments.

The key policy questions concern how best to structure airports and groups of airports
to obtain maximum customer benefits. The discussion in the volume on airports and air traffic
controls indicates that there is little evidence of significant scale benefits flowing from multiple
airport operation; equally, however, there is little evidence of significant scale diseconomies.
The case for significantly reducing the concentration of airport ownership at privatization
 16

therefore depends on the trade-off between the up-front and visible costs of re-structuring, and the
possibly less tangible benefits of increased competition resulting from break-up. The competition
benefits in this industry are not clear-cut, primarily because major airports mainly serve distinct
regional markets.

In the United Kingdom, the authorities took the view that any potential competition gains
from breaking up the British Airport Authority prior to privatization would have been offset by
restructuring costs. In Australia, in contrast, the Government has preferred to restructure and reduce
industry concentration radically, emphasizing the public policy benefits of inter-airport competition
for long haul international traffic. The benefits of fragmented ownership also include those that flow
from yardstick competition, enabling regulatory agencies to assess individual operator performance
more effectively; and from introducing a limited element of competition by emulation between
operators. The airport expert found the benefits from the Australian model to be greater. Key
recommendations for the airport sub-sector are as follows:

• Airport privatization will be encouraged by the existence of legislation in the form of a

BOT law or similar, signaling the government’s recognition of the need for PSP in
infrastructure provision, It is also important to ensure that the government is able to
demonstrate that any projects offered to the private sector are economically viable.

• Regarding the optimum approach, full privatization based on asset transfer or acquisition
through long-term leases is preferable to more restricted forms of PSP (but is also
more demanding in terms of legal and regulatory frameworks).

• As to airport industry restructuring, there is no evidence of significant economies of

scale in airport operation other than those associated with increased traffic density at a
particular location. Hence, PSP can be based on individual airports (although facilities
may need to be bundled to assist financing of major new developments or extensions to
capacity).

• The existence of unprofitable airports does not justify the maintenance of a highly
concentrated industry structure to facilitate cross-subsidies.

• Limited sharing of traffic and revenue risk (between the private sector partner and
government) is justifiable in airport BOT or concession contracts.

• Denomination of some, or all, airport charges in US dollars is an effective way of hedging

against currency risk and may significantly reduce the risk premium required by private
investors;

• The benefits of PSP in airports are likely to be maximized by regulatory frameworks

that incorporate good regulatory governance practice. The price-cap approach to
constraining airport charges is likely to encourage better performance outcomes than
one based on rate of return regulation.

• Competition for the market, whether through sale or leases, or BOT/concessioning,

will be maximized by transparent bidding/sale processes.
 17

V. THE ROLE OF THE ASIAN DEVELOPMENT BANK

The crisis has focused on the urgent need for institutional strengthening and governance
reforms in both the financial and infrastructure sectors, areas where ADB can play a major
role. There are a number of ways identified in the study in which ADB can assist in the reforms
associated with increased PSP in infrastructure. The most obvious is to provide technical
assistance to define policy objectives, develop network master plans, identify and evaluate
projects, define the role of new regulatory institutions, and train regulators to handle their new
responsibilities, prepare contracts and negotiate with the private sector. ADB’s efforts to promote
financial sector reform and develop long term capital markets will also be important. This
would include efforts to improve the bankruptcy laws, and the regulation of domestic debt and
equity markets.

In order for ADB to have a significant role in promoting PSP, it should link this promotion
with on-going project lending. ADB can provide support for private sector investment directly
through its private sector window and through its guarantee operations. More importantly,
ADB should provide sovereign loans to complement but not compete with private sector
investment in the form of public-private partnerships. Public sector project lending should
also be used to finance basic infrastructure that cannot be packaged into financially viable
investments for the private sector but provides significant economic benefits and improves sector
efficiency. Program lending is another key modality to promote the necessary reforms where ADB
provides financing for the adjustment costs in stages, upon the satisfactory achievement or fulfillment
of government actions that will promote PSP and sector restructuring. This modality allows ADB
to exercise some leverage on government decisions and actions to support reform. Country
strategies should address which areas of development are to be financed by government using
sovereign loans, general revenues and government bonds and which are to be financed by private
investment and should ensure a coordinated approach to all forms of ADB assistance.
 PART TWO

AIRPORTS
AND
AIR TRAFFIC CONTROL
REPORT
 I. INTRODUCTION

This report is one of a series commissioned by the Asian Development Bank (ADB), identifying
best practice for promoting the role of the private sector in financing and operating infrastructure in
its developing member countries (DMCs). The report focuses on the airports and air traffic control
(ATC) sectors; other reports cover the power, water supply, roads and ports sectors.

The background to the report is that, in common with other international agencies and
commentators, ADB believes there are significant advantages in expanding the role of the private
sector in financing and implementing transport infrastructure and related services in the DMCs,
principally for two reasons:

• First, private sector participation (PSP) may help to overcome constraints on public
sector borrowing, and, equally or possibly even more important, on the public sector’s
capacity to implement efficiently and cost-effectively large-scale infrastructure programs.

• Second, the active participation of the private sector in all phases of the project life cycle
may secure better value-for-money in the project than the traditional design-build model,
where the private sector’s role was limited to the project construction phase.

The remainder of the report gives an overview of the economics of airports and of ATC
services, drawing attention to the diverse range of activities carried out at airports, the existing role
of the private sector in providing airport services, the conditions of competition, and the need for
economic regulation. The report then describes the alternative models for PSP in core airport
activities, and evaluates the strengths and weaknesses of each approach. Although full privatization
of airport assets is proposed as the appropriate target model for the sector, it is recognized that this
approach is relatively demanding in terms of the legislative and institutional infrastructure required
for implementation. More restricted forms of PSP, such as concessions or strategic partnerships,
may be appropriate interim vehicles for PSP in the absence of the necessary legislative framework
to enable privatization. Drawing on the set of case studies of different approaches to PSP in airports,
as well as experience of PSP in other types of transport infrastructure, the report then sets out the
contractual and other conditions likely to encourage successful PSP projects. Finally, there is a
discussion on the potential role of ADB and other development agencies in facilitating PSP ventures
in the airport sector.

II. THE BUSINESS OF AIRPORTS AND AIR TRAFFIC CONTROL

A. Introduction

This section discusses airports as businesses, drawing attention to the wide range of
activities carried out at airports, and the role of the airport operator. It describes the sources of
revenue for airport operators, covers demand and cost conditions, discusses the conditions of
competition applying to different types of services and the implications for economic regulation,
and describes the provision of ATC services.

B. Airport Activities and the Role of Airport Operator

An airport can be defined as one or more runways and complementary facilities for
aircraft (taxiways, apron areas) together with associated terminals and facilities for handling
 19

passengers and freight. Within the airport framework, the airport operator is typically directly
responsible for the provision and maintenance of airport infrastructure, and the provision of essential
services, including passenger search and perimeter security, fire fighting, and cleaning and
maintenance of passenger terminal areas (the latter often provided by sub-contractors). These
services are referred to as core airport activities. The airport operator also allocates space and
resources, both between airlines (for example, check-in desks, passenger departure lounges) and
between commercial concessionaires (such as retailers or caterers).1

Other airport services are typically provided by airlines or their handling agents (including
check-in processing, baggage handling, and aircraft maintenance) and control authorities (including
customs, immigration, policing and ATC). In addition, a wide range of customer services, including
retailing, catering, banking, and car hire, are provided by concessionaires appointed by the airport
operator.2

Whilst it owns the large majority of the airport capital stock, the airport operator may employ
directly only a modest proportion of airport employees, since many activities carried out by other
agencies, such as baggage handling and retailing, are relatively labor intensive. An extreme example
of this is London Heathrow airport, where the airport commercial activities have been exceptionally
intensively developed by the airport’s owner, the British Airport Authority (BAA) plc. As a consequence,
BAA directly employs less than 10 percent of all full time airport employees.3

C. Airport Operators’ Revenues

Airport operators typically derive revenues from airport charges, levied on airlines, to cover
the provision of core airport services. Airport operators also obtain revenues from on-airport
commercial activities, in the form of concession fees and rents from concessionaires, property
rents and charges for services such as airport car parking.

1. Airport Charges

The airport charges paid by airlines usually include:

• Take-off or landing charges, based on aircraft weight.

• Apron or parking charges, again based on aircraft size or weight, and the duration of
stay.

• Passenger handling charges, expressed as a rate per departing (or arriving) passenger,
and differentiated by domestic and international flights. Such charges are distinct from
the departure taxes, imposed by governments in the United Kingdom and elsewhere,
which accrue directly to the government concerned as general taxation and not to the
airport operator.

Broad principles covering both the costs to be recovered through airport charges and
also the structure of charges are set out in guidelines published by the International Civil

________________________________________________________
1
In most countries, passenger terminals are provided as common user facilities by the airport operator. However, in some countries,
such as the United States (US) and Australia, terminals may be owned and operated by airlines, either as exclusive or as common
user facilities.
2
Some airport operators continue to provide duty free facilities for international passengers.
3
Around 4,000 out of the total airport employment of 55,000 in 1996.
 20

Aviation Organization (lOAD),4 although such guidelines are not mandatory.5 Airport charges are
more directly constrained in certain respects by international legislation, notably the Chicago
Convention, paragraph 15 of which imposes non-discrimination conditions with respect to airport
access and airport charges on signatories. Further constraints on airport access and charging
may also be specified in bilateral Air Service Agreements governing the provision of scheduled air
services.

Table 1 presents an index of airport charges at 40 international airports throughout the

world in 1995, based on published airport charges for eight different types of aircraft. Although the
data do not take account of factors such as the availability of accounts or the extent to which there
may be variations in the range of services covered by the charges, it is clear that there are very
wide variations in the level of airport charges in different locations. Average charges in the upper
quartile of airports are over three times as high as those in the lower quartile. The average charges
show little systematic variation with factors such as airport size, traffic composition or ownership
characteristics.

Table 1: Airport Charges Index (ranking of 40 airports world wide), 1995

Airport Index Airport Index

Tokyo 100 Orlando 27
Frankfurt 53 Miami 26
Vienna 52 Singapore 25
Chicago 51 Washington 25
New Jersey 48 Madrid 23
Berlin 44 Heathrow 23
Munich 43 Milan-Linate 22
New York JFK 43 Los Angeles 22
Amsterdam 42 Budapest 22
Brussels 39 Vancouver 22
Paris Charles De Gaulle 39 Bombay 21
Athens 38 Honolulu 21
Oslo 37 Cairo 20
Helsinki 36 Gatwick 19
Moscow 35 Dallas/Fort Worth 18
Copenhagen 33 Hong Kong 16
Stockholm 32 Sydney 12
Lisbon 32 Jeddah 11
Dublin 28 Algiers 10
Johannesburg 27 Dubai 5
Source: Monopolies and Mergers Commission (1997, BAA plc).
Note: Based on an average of peak and off-peak charges for eight aircraft types (international
arrivals and departures).

The lCAO guidelines also recommend (paragraph 16(i)) that under normal
circumstances, charges should be expressed and payable in the local currency of the state
concerned. However, it is recognized that under special circumstances, “for example where
economic conditions are not stable”, airport operators are entitled to denominate user charges in
another currency.
________________________________________________________

4
5
ICAO, Statement from the Council on Airport Charges.
Thus, the ICAO guidelines recommend that the passenger handling charge is levied on airlines and therefore reflected in airline
fares, but many airport authorities continue to levy a charge directly on international passengers at the point of departure.
 21

Table 2 summarizes current practice with respect to the currency used in setting airport
charges in a group of twelve DMCs. It shows that in half of the cases (India; Malaysia; Singapore;
Thailand; South Korea; and Hong Kong, China) all types of airport charges are denominated in the
local currencies concerned. In three cases (Pakistan, Cambodia, and the Philippines) charges are
US dollar denominated, and in the three remaining areas (Indonesia, Viet Nam, and Laos),
international traffic charges are US dollar denominated, whilst domestic traffic charges are wholly
or partly local currency denominated.

As noted in the following section, denominating some or all airport charges in US dollars is
a potentially useful instrument for mitigating the currency risk faced by prospective overseas investors
in airports, some or all of whose financing costs may be denominated in US dollars.

Table 2: The Currency of Airport Charges

Currency of Currency of
Country International Landing Fees Domestic Landing Fees Passenger Charges

India India rupees India rupees India rupees

Pakistan US dollar US dollar US dollar
Malaysia Malaysian ringgit Malaysian ringgit Malaysian ringgit
Singapore Singapore dollar Singapore dollar Singapore dollar
Philippines US dollar US dollar US dollar
Thailand Thai baht Thai baht Thai baht

Cambodia US dollar US dollar US dollar

Indonesia US dollar Indonesian rupiah Indonesian rupiah
Viet Nam US dollar US dollar International: US dollar
Domestic: Vietnamese dong
Lao PDR US dollar US dollar Domestic: Laos kip
South Korea Korean won Korean won Korean won
Hong Kong, China Hong Kong dollar Hong Kong dollar Hong Kong dollar

2. Commercial Revenues at Airports

A major potential source of commercial revenues at airports with a significant volume of

international traffic are tax and duty-free sales to international travelers. At airports where the traffic
is wholly or mainly domestic, the scope for developing retailing activities is more limited, and focuses
on services such as catering, car parking, and car hire.

Revenues from commercial activities vary widely between airport operators, partly as a
function of exogenous factors, principally the role of the airport operator, and whether it operates
passenger terminals, and the mix of domestic and international traffic. However, there is also very
wide variation in the revenue mix between airports where passenger terminals are owned and
operated by the airport owner, and with a similar mix of international and domestic traffic.
Kapur 6 suggests that variations in the level of revenue generated from commercial activities
are linked to the ownership status of the airport operator. He shows that land-side (commercial)

________________________________________________________

6
Kapur, Anil (1995), Airport Infrastructure. The Emerging Role of the Private Sector, World Bank Technical Paper
313.
 22

revenue per passenger in the early 1990s varied by a factor of almost three between government
department owned airports and privately owned airports (Figure 1).

Figure 1: Average Land-side Revenue per Passenger by Ownership Structure

12 11.14

10
8.61
7.64
8 7.23
S/pax

6 4.88

0
Govt. Dept. Public Corp. Regional Public Private
Govt. Private

Source: Kapur (1995)

Kapur cites the example of privatized BAA plc as an airport operator which has been especially
effective in developing on-airport commercial activities, both in its own airports and, increasingly,
as a contracted manager of passenger terminals and commercial facilities in other countries,
including Hong Kong, China; Malaysia; and US.

Given the framework of economic regulation in UK, characterized by the application of

the so-called “single till” approach, 7 and price regulation of airport charges, BAA’s success in
developing its on-airport commercial activities has translated directly into strong downward
pressure on the level of airport charges. Average revenue per passenger from airport charges
at BAA’s London Airports fell by approximately 20 percent in real terms over the period 1992/93 to
1996/97.8

________________________________________________________

7
Under the “single till” approach, airport charges must be set taking into account the profits from other on-airport commercial
activities accruing to the airport operator. The general principles governing the level of airport charges are set out in the ICAO
guidelines in the following terms:
“In determining the cost basis for airport charges the following principles shall be applied: the cost to be shared is the full cost of
providing the airport and its essential ancillary services ... but allowing for revenues both aeronautical or non-aeronautical
accruing from the operation of the airport.”
8
See, Monopolies and Mergers Commission (1997), BAA plc.
 23

D. Demand and Cost Conditions

1. Demand Conditions

As with other types of transport infrastructure, the demand for airport capacity is subject to
daily and seasonal fluctuations, the nature of which depends on airport location; operating conditions
(24-hour or restricted opening); and traffic mix (domestic - international; short haul long haul; business
- tourist/recreation). Demand for air travel is relatively income-elastic and has been growing strongly
for many years (around five percent per annum world wide for passengers, and slightly above five
percent per annum for freight). The rate of growth of demand has varied between regions (Asia-
Pacific, Europe, US, etc.), mainly reflecting inter-regional variations in GDP growth rates. Until the
recent recession, demand had been growing particularly rapidly in the Asia-Pacific region, where
9
passenger volumes at major airports had been expanding at 10-13 percent per annum in the 1990.
Because of increases in aircraft size, the demand for airside capacity has been increasing slightly
less rapidly in recent years than the growth in passenger demand, and this trend is forecast to
continue.

Although there is no firm statistical evidence available, demand for airport capacity from
airlines at a particular airport location, given the prevailing level of airport charges world wide, is
believed to be very highly price inelastic for two main reasons:

• First, airport charges represent only around five percent of total airline operating costs
for the International Air Transport Association affiliated airlines (the proportion is
somewhat less for long haul operations, but may be as high as 15 percent for short haul
services). This means that, on average, even a 100 percent increase in airport charges
would only increase airline costs by about five percent;

• Second, there are usually no close substitute airport facilities available, especially for
major regional hub airports, such as Singapore, Hong Kong, Bangkok, Manila or Kuala
Lumpur.

2. Cost Conditions

The lack of close substitutes on the demand side reflects the fact that the supply of airport
services is subject to economies of traffic density, extending at least as far as the traffic
throughput of a single runway capable of handling the largest passenger jet aircraft (currently
the B747 series). The theoretical capacity of such a runway is determined by the maximum
number of incoming and outgoing flights that can be handled, which depends upon the quality
and availability of complementary ATC and taxiway facilities. The practical capacity will reflect
demand as well as supply side factors, including traffic mix, aircraft size, the airport load
configuration, the capacity of passenger terminals and parking areas, and any unsocial hours
restrictions on take-offs and landings. To illustrate the traffic volumes that can be accommodated
within the basic single runway configuration, Gatwick, one of BAA plc’s London system airports,

________________________________________________________

9
During the period 1991-1995, passenger volumes increased annually at major airports as follows:-
Hong Kong: 9.3%
Singapore: 9.7%
Manila: 10.7%
Bangkok: 11%
Jakarta: 12.4%
Seoul: 13.1%
Source: ICAO, Airport Traffic, 1995.
 24

currently handles around 24 million passengers per annum (mppa), and projected capacity for the
year 2000 is around 30 mppa.

Network scale economies in the provision of services by airlines probably mean that the
minimum efficient scale of airport operation exceeds the capacity of a single runway facility. A two-
runway airport handling, say, 50 mppa will offer airline passengers a wider range of interconnecting
services to passengers than two single-runway airports each handling 25 mppa. As a consequence,
the airlines’ willingness to pay for access to the 50 mppa facility will be worth more than twice their
willingness to pay for two 25 mppa facilities.

A traffic density of 50 mppa exceeds the demand for air travel at all airports except for a
handful of major conurbations throughout the world, such as London, New York, Paris and Tokyo.
Table 3 shows 1995 traffic levels at major airports in the DMCs covered in the present study.

Table 3: 1995 Traffic Levels at Principal DMC Airports

— million passenger miles per annum (mppa)

Country mppa

Republic of Korea (Seoul) 30.7

Hong Kong, China 27.4
Singapore (Changi) 21.7
Thailand (Bangkok) 20.9
Malaysia (Kuala Lumpur) 14.2
Indonesia (Jakarta) 12.8
Philippines (Manila) 10.9
India (Bombay) 10.7
Pakistan (Karachi) 4.9
Source: ICAO, Airport Traffic, 1995

Although there are strong traffic density economies in airport operations serving a single
conurbation or region, there is no evidence that scale effects extend to the operation of more than
one airport in different locations. Thus the decision to transfer all seven of the BAA’s airports (three
in London and four in Scotland) to privatized BAA plc owed more to the UK Government’s desire to
avoid the complexities and costs of restructuring, and to maximize flotation proceeds at an early
stage of the privatization process, than to a considered view on the existence of multi-airport scale
effects. In this respect, the more recent policy of the Australian Government in disposing of Federal
Airport Corporation airports individually, better reflects the economics of the business, and will
improve the effectiveness of economic regulation in the long run, by introducing an element of
yardstick or comparator competition into the process.

E. The Conditions of Competition and the Need for Economic Regulation

As well as having market power over the prices of essential airport services used by airlines,
airport operators (and/or their agents) at major hub airports also enjoy market power in respect of
a wide range of airport-related services, including rentals for on-airport facilities used by airlines or
other air transport businesses, and services to passengers which are highly localized in time and
space, such as air-side restaurants and airport parking.
 25

The position of many airport concessionaires, such as those operating retailing outlets
and land-side restaurants, is somewhat different. Such businesses compete with other local
off-airport retail businesses, and in the case of duty-free shops, with duty-free facilities at overseas
airports or on aircraft.

The monopoly power enjoyed by airport operators in the markets for air transport services
and airport-related ancillary services could be used by profit-seeking businesses to earn profits in
excess of those required to attract new investment into the business (so-called “supernormal”
profits). Such a policy might manifest itself not only in terms of high prices for services supplied by
the airport, but also in under investment in airport facilities, such that capacity expansion programs
were unreasonably delayed, leading to poor quality of service to airlines and passengers. Market
power can also be abused if the monopolist is able to earn sufficiently high profits given super-
competitive cost levels. The ability to abuse monopoly power in these ways would need to be
constrained in order to protect the interests of all types of airport user, including airlines, passengers,
and freight shippers.

F. Air Traffic Control Services

The provision of ATC services covers two distinct but complementary activities:

• Airport ATC, involving the control of take-off and landing at airports, and of surface
aircraft movements within the area of the airport, where the ATC activity is usually
carried out from an airport control tower.

• En route ATC, where a single-control center, often quite separate from an individual
airport ATC facility, may regulate traffic over a wide area.

The two systems must be closely integrated for operational purposes; as aircraft approach
or take-off from the airport, so control passes between the airport and en route ATC centers. Both
activities are subject to strongly increasing returns to traffic density and, in the case of en route
ATC, to scale, in the sense of the size of area controlled.

An important aspect of ATC provision is the need for interface between civil and military use
of the airspace. In some countries, such as Argentina, this interface is secured by having all ATC
services provided by the military authorities. Elsewhere, civil ATC operates within designated
controlled airspace, and the military authorities provide ATC services to both military and civil aircraft
in so-called uncontrolled airspace. The existence of a strong defense interest in the provision of
ATC services, which can affect many aspects of the specification and operation, has sometimes
been held to preclude full privatization of ATC assets, other than those supplying airport ATC.

ATC costs are recovered through air navigation charges; as with airport charges, the
structure and level of ATC charges are subject to international agreements and conventions.
For example, in Europe, ATC charges are subject to the Eurocontrol Multilateral Agreement, under
which charges are adjusted annually, so as to recover the full costs incurred by each member
state.
 26

Ill. ALTERNATIVE MODELS OF PRIVATE SECTOR PARTICIPATION

A. Introduction

Airport services do not exhibit the classic public good characteristics of non-rivalry, non-
excludability and asymmetric information between suppliers and purchasers which make provision
of the services by profit-seeking private sector businesses problematic. They are, however,
characterized by pervasive external effects, in the form of noise, visual intrusion, and air pollution,
and by spill-over effects and complementarities with other surface transport infrastructures, which
mean that public authorities will necessarily continue to play a significant role in project initiation
and planning. As noted, core airport services are also natural monopoly activities, so that a framework
of economic regulation is required to limit the abuse of a dominant position, irrespective of whether
the facility is publicly or privately owned.

However, there is now ample experience world wide of natural monopoly utility assets being
owned and operated by private sector businesses within a framework of economic regulation.
There is increasing evidence that privatized provision, combined with incentive compatible forms
of regulation, based on the price cap approach, offers superior performance outcomes (lower
prices, and improved service quality as well as improved profitability) compared to service provision
by state-owned enterprises.10

In this section, the report discusses alternative models of PSP in airport and ATC activities,
ranging from complete or almost complete privatization, through flotation or trade sale, to more
restricted forms of management contracting. Complete privatization is proposed as an appropriate
target model, but it is recognized that this approach works best within a relatively sophisticated
regulatory framework involving not only an appropriate legislative framework but also a wider set of
conditions covering the conduct of economic regulation. A recent National Economic Research
Associates (NERA) report to ADB11 indicates that current regulatory arrangements in most Asian
DMCs fall far short of international best practice. In this situation, more restricted forms of PSP
may provide appropriate means of extending the role of the private sector in airport activities. The
sector also covers the privatization of ATC facilities.

B. Alternative Models of PSP

Table 4 summarizes the key features of alternative models of PSP:

• Full privatization.

• Partial privatization.

- Concessions.
- Strategic partnership(s).
- Management contract(s).

________________________________________________________

10
See, for example, NERA (1996), The Performance of Privatised Industries: Prices and Service Quality, a report for the
Centre for Policy Studies.
11
NERA (1998), Governance and Regulatory Regimes for Private Sector Infrastructure Development: Final Report, ADB
RETA 5758-REG.
 27

1. Full Privatization

Full privatization involves the transfer of ownership of airport assets from a public
corporation to private investors through a flotation or through a trade sale. Following
privatization, the privatized entity is fully responsible for operating the airport facilities, directly
or through agents/concessionaires, and for financing investments in airport assets internally, from
retained earnings, or externally through the issue of new equity or debt.

The first example of airport privatization occurred in UK, with the flotation of BAA plc in
1986. BAA plc took control of the seven airports previously owned by British Airports Authority, a
public corporation. Since 1986, several other publicly owned airports have been privatized, including
Belfast International, East Midlands, Southampton and Bristol.

Outside UK the most far-reaching privatization program has taken place in Australia, where
long term leases (50 years with an option to extend for a further 49 years) were offered for sale over
18 of the 22 airports operated by the Australian FAC in a two phase sales program. The first phase
covered the sale of leases on three major international gateway airports, at Brisbane, Melbourne
and Perth, and the second phase the sale of leases on a further fifteen airports.

2. Partial Privatization

The evidence on the current extent of PSP in airports, summarized in Appendix 1, shows
that there have been relatively few instances of full privatization. There are more examples where
private sector financing and operation of airport assets have been introduced via partial privatization
measures, although the large majority of cases listed in Appendix 1 refer to situations where some
form of partial privatization is being considered or planned but has not yet been realized. The
comparative absence of full privatization based on two interrelated factors:

• First, governments’ reluctance to cede control over what, at least in the case of major
capital city airports, is widely regarded as a vital national asset.

• Second, the lack of an appropriate framework of economic regulation and regulatory

governance to balance the interests, short and long term, of consumers and investors,
given the highly immobile and long-lived character of airport assets.

These two factors combine to encourage forms of PSP based on long-term contracting or
public-private partnerships, in which the contract or partnership mechanism acts, first, to secure
investment in airport infrastructure sought by the government partner, and, second, to protect the
private investor against arbitrary or opportunistic behavior by government.

Three principal variants of this partial privatization model can be identified:

• Concession or build-operate-transfer (BOT) projects.

• Strategic partnerships.
• Management contracts.
 Table 4: Alternative Models of PSP

Partial Privatization
Strategic Management
Privatization Concessions Partnerships Contract

1. Roles

Ownership Private State State State

Investment Private Private/mixed Mixed State

Operation Private Private/mixed Private/Mixed Private/Mixed

2. Regulation Independent Regulator Contract, ownership or Ownership Ownership

independent regulator

3. Examples UK Colombia — Bogota12 Thailand12 US

28
- British Airports Authority plc Philippines — Manila12 South Africa – Indianapolis
(BAA)
Cambodia — Phnom Penh – Pittsburgh
- Regional airports Australia12
Argentina12 Italy
- Federal Airport Corporation
(FAC) airports Côte d’lvoire — Abidjan12 – Naples

Malaysia

– Kuala Lumpur

________________________________________________________

12
Covered in case studies in Appendix 2.
 29

a. Concessions

As demonstrated in Appendix 1, the large majority of actual or planned cases of private

sector privatization in airports involve concession contracts, under which control of some or all
core airport assets is transferred to a private investor, who is responsible for financing investment
and operating the airport for the term of the concession, typically 25-30 years. The precise extent of
the private sector operator’s responsibilities varies. In some cases, the concession agreement
may cover financing and operation of all airport assets; in others, the concessionaire may only be
responsible for financing and operating a particular facility, such as a passenger terminal or runways,
alongside the incumbent public sector airport operator.

The case studies in Appendix 2 include three airport concession projects in developing
countries:

• Abidjan Hophôuet-Boigny Airport, Côte d’lvoire.

• Ninoy Aquino International Airport, Philippines.

• Bogota El Dorado International Airport, Colombia.

Responsibility for the operation and development of Abidjan Hophôuet-Boigny Airport (AERIA)
was transferred through a 15-year concession agreement signed in July 1996 between the
Government of Côte d’lvoire and AERIA, a special purpose company controlled by Société
d’Exploitation et de Gestion Aéroportuaire (SEGAP), itself a jointly owned subsidiary of the Marseilles
Chamber of Commerce and Industry (MCCI) and Groupe Sofrevia, a French aviation services
company. MCCI operates Marseilles airport, and SEGAP also operates Libreville (Gabon) airport
under a 30-year concession agreement signed in 1998.

Under the concession agreement, AERIA committed to a four year investment program,
covering a major expansion of the international passenger terminal and associated parking apron
and taxiway areas together with runway reinforcement and extension, aimed at expanding the
airport capacity form around one to 1.7 mppa by 2001. AERIA will finance the investments from
airport user charges and other airport revenues; it will also pay concession fees to the Ivorian
authorities, amounting to approximately 20 percent of turnover, largely to finance the operation of
unprofitable interior airports in Côte d’lvoire.

At Ninoy Aquino International Airport (NAIA), a concessionaire, Philippine International

Air Terminal Company Inc. (PIATCO), has been selected to construct and operate a new
international passenger terminal over a 25-year concession period. The Manila International
Airport Authority (MIAA), a public corporation, will continue to own and operate other airport
assets, including the two existing passenger terminals and runways. Under the concession
contract, PIATCO will be able to increase international passenger charges (levied on airlines)
from the initial level determined by the Ministry of Transport and Communications in line with a
pre-determined formula. PIATCO will also pay a two-part concession fee to MIAA, consisting
of a sum fixed in real terms, and a variable fee, expressed as a proportion of total revenues
from international passenger charges and other commercial activities. This two-part structure
introduces a degree of revenue risk sharing into the agreement.13

________________________________________________________

13
There is further discussion of this and other contract design issues in Section IV below.
 30

In the Bogota El Dorado International Airport project, the concessionaire, Compania de

Desarollo Aeropuerto El Dorado SA (CODAD) will construct and maintain a new runway and taxiway
facilities over the 20-year concession period. However, it will also be responsible for maintaining
the existing runway, taxiways and parking areas, and will be remunerated by revenues from airport
landing charges. The airport passenger terminal facilities will continue to be operated by Aeronautica
Civil (Aerocivil), the government agency responsible for operating other airports and ATC facilities
in Colombia.

The concession agreement provides for the concessionaire to double landing charges on
completion of the new facility. Thereafter, charges will be increased in line with domestic inflation.
Unlike the Manila concession, the agreement does not require the concessionaire to make any
payments to the Colombian Government. Recently, CODAD has successfully issued 15-year US
dollar denominated revenue bonds, on the basis of a minimum revenue guarantee by the Colombian
Government. Revenue bonds, linked to tax incentives, have been widely used as a vehicle for
airport infrastructure financing by public sector authorities in the US, including the Port Authority of
New York and New Jersey.

The majority of concession projects identified in Appendix 1 cover specific investment

projects, similar to those implemented at Manila and Bogota. Francophone countries, in particular,
such as Gabon and Cambodia, offer examples of more wide-ranging concession agreements,
covering the upgrading and operation of an entire airport. On a much larger scale still is the recently
concluded concession agreement covering the transfer of 33 of Argentina’s 59 civil airports to a
private sector-led consortium.14 In this case, the 30-year concession agreement requires the
concessionaire to finance and undertake a US$2 billion upgrading and modernization program
covering the principal international and regional airports, and to pay the government an annual
concession fee of US$171 million throughout the concession period. The airports included in the
concession include a mix of profitable and unprofitable facilities; investment and operating losses
at the unprofitable airports will be financed by cross-subsidy from the profitable airports. It is expected
that the investment program will be financed by a combination of free cash flow from existing
airport operations and some limited project finance.

The make-up of a concession consortium depends upon factors such as the nature of
the project and the extent of any constraints imposed by governments on stake-holdings (many
governments require a significant, or majority, ownership stake to be held by indigenous
businesses). For projects involving construction and operation of passenger terminals, the
consortium would normally include an airport operator, such as BAA plc, alongside a
construction company (often local) and an investment bank. Shareholders usually take only a
very limited equity stake (5-10 percent of funding) with the bulk of the financing taking the form
of externally held debt, sometimes including an element of development bank funding.

________________________________________________________

14
See Appendix 2 for further details.
 31

b. Strategic Partnerships

Under the strategic partner model, a private sector firm or consortium acquires a stake
(typically a minority shareholding) in a state owned airports operator. Proposals for re-structuring
the Airports Authority of Thailand (AAT) via a strategic partnership are currently being considered
15
by the Thai Government, and the South African authorities have offered a minority stake in the
Southern African Airports Company Limited to a strategic investor, as a possible step towards full
privatization.

In Thailand, it is expected that, apart from taking an equity stake in the re-structured airports
operating company, the strategic partner will also be awarded a management contract covering
the operation of a range of airport ancillary activities, such as internal telephone systems, car
parking and aircraft maintenance facilities.

Strategic partnerships provide a vehicle for introducing private sector finance and operational
expertise in order to directly relieve public financing constraints and to improve operational and
financial performance. Another form of partial privatization occurs when a minority of shares in a
state-owned airports operator are sold to private investors, usually through a flotation. This approach
has been followed in Vienna Airport (Austria) and in Copenhagen Airport (Denmark). The incumbent
management, and through its power to appoint top management, the state, retains control of the
business, but the conduct of the business is exposed to a measure of external capital market
discipline. As with strategic partnerships, this may be the precursor to eventual privatization.

c. Management Contracts

Under the management contract model, a private sector contractor is retained to manage
airport assets, usually passenger terminal facilities or retailing activities within passenger terminals.
Other airport operational activities, such as maintenance and operation of runways and ATC facilities,
continue to be undertaken by the airport owner or other state sector agencies. This model enables
the private sector contractor to transfer best practice across a range of airport activities, thereby
reducing costs and enhancing revenues and improving standards of services. Responsibility for
funding investment in airport assets is retained by the airport owner, but the prospects for more
wide-ranging types of privatization may be greatly improved by the increased profitability of the
business under the management contract.

The concessionaire would either receive a management fee, linked to revenues generated
in the activities for which it was responsible, or it would receive a share of airport revenues, but
would pay a lease or rental charge to the airport owner. With responsibility for financing major
investments remaining with the airport owner, the length of a management contract would tend to
be significantly shorter than the term of a BOT contract.

To date, management contracts have been almost exclusively applied by public airport
authorities in developed OECD countries as a means of improving service quality and the
financial performance of the airport. In developing countries, the stimulus to engage the private
sector is more frequently related to securing additional funding for investment projects and for
gaining the benefit of private sector skills in project management.

________________________________________________________

15
See Appendix 2 for further details.
 32

C. Full and Partial Privatization Compared

It was argued above that given the current political climate and regulatory practice in DMCs,
full privatization of airports as in UK or Australia, is probably not feasible in these countries. However,
if the climate was to change and regulatory arrangements were to move closer to international best
practice, so that full privatization was feasible, what are the pros and cons of the two approaches?

The first point to make is that any improvement in regulatory practice will generally be
expected to improve the terms on which the private sector is willing to invest in DMCs. Although
contractual mechanisms might be regarded as a substitute for economic regulation, concession
contracts will almost certainly need to be revised over the term of the concession in light of
unexpected market developments. The presence of an effective regulatory governance framework
both increases competition for the market and improves performance under contracts, by giving
investors better assurance that contract terms would be revised in a manner which respected their
interests as well as those of consumers.

The potential advantages and disadvantages of full and partial privatization reflect the
economic characteristics of the two approaches. By comparison with full privatization, in which all
operational assets are owned and operated by the private sector, partial privatization arrangements,
are certainly time limited, and are often scope limited. Experience to date indicates that BOT projects
and management contracts, in particular, are often embedded within a wider set of airport activities
which continue to be provided by a public sector agency. Finally, partial privatization through strategic
partnership requires control to be shared between the public and private sector partners.

Compared to full privatization, these characteristics of partial privatization may have certain
potentially adverse effects on the performance of the service provider:

• Weaker incentives to invest and to innovate, especially during the later stages of the
concession, because its time-limited nature restricts the scope of benefit capture.

• Higher transaction costs associated with incomplete contracting.

• More generally, a partial privatization framework based on formal contractual

mechanisms may restrict the private sector partner’s ability to respond flexibly to
unexpected market developments, etc.

• Finally, concessionaire’s costs may be inflated for one of two (mutually exclusive)
reasons. First, suppose that the concession agreement does not contain explicit
provisions for compensating the concessionaire for the residual value of all assets
acquired during the concession period,16 but whose economic life extends beyond the
end of the concession terminal date. The concessionaire will need, prudently, to amortize
the cost of the assets over the remaining term of the concession, since there is no
guarantee of the concession being extended. This would mean that the level of
depreciation to be recovered through airport charges will be higher than if the airport
assets had been privatized. Alternatively, if the agreement does contain
compensation provisions, there may still be significant uncertainty as to how they will
be applied in practice, especially in first generation concessions and in the absence

________________________________________________________

16
As several existing concession agreements do not.
 33

of mature regulatory institutions and precedents. This uncertainty will either be reflected
in a higher cost of capital (conceptually similar to the so-called “regulatory risk premium”)
or in conservative estimates of the terminal cash flow element, leading to less favorable
17
financial bids.

However, performance outcomes will also depend upon the nature and extent of competitive
pressures on the incumbent, and in this respect, a time-limited concession approach offers some
advantage over full privatization, since it enables periodic competition for the market, albeit across
a more restricted set of activities. This benefit is attenuated in practice because of the long term
nature of airport concessions, which is required in order to provide the concessionaire with adequate
incentives to invest in very long-lived assets, in the absence of robust arrangement for compensation
based on residual asset values. The extent of any advantage is also uncertain because a privatized
business will itself be subject to competitive pressures from the capital market.

On balance, its advantages over partial privatization solutions means that full privatization
of airport operations should be regarded as an appropriate target model for DMCs. Improvements
in regulatory practice should therefore not only improve outcomes under existing partial privatization
initiatives, but should bring wider benefits, by improving prospects for full privatization of airports.

D. Air Traffic Control

There is clear evidence of rapidly expanding interest in and experience of private sector
financing and operation of airports. However, there are few signs of any corresponding developments
in ATC provision, despite widespread concern regarding the difficulties of financing investment to
upgrade en-route ATC, given constraints on public expenditure, and the perceived inefficiency of
many state-owned ATC providers.

Apart from full privatization of the Civil Aviation Authority’s (CAA) ATC division National Air
Traffic Services (NATS), which was proposed by the UK Government in 1994, but not subsequently
implemented, two other types of structural reform of ATC provision have been implemented in
recent years:

• Provision of ATC services by a non-profit making trust. This approach has been adopted
in Canada, where the Canadian Air Navigation System was sold to NAV CANADA in
1996. NAV CANADA is constituted as a non-profit corporation, with a Board comprising
representatives of airlines, government and the air traffic controllers’ union. It is allowed
to set user charges to recover costs, but any surpluses must either be used to retire
debt or enhance traffic services.

• Corporatization of the ATC activity, either by transforming an existing authority into a

public corporation (as in New Zealand) or by transferring ATC assets from an authority
to an existing corporation (as in Malaysia, where ATC assets are now owned by the
state airports corporation).

Following the abandonment of full privatization in 1995, the UK Government has explored
a partial privatization approach to ATC, within the so-called Private Finance Initiative

________________________________________________________

17
It should, however, be noted that problems of this kind may arise under full privatization, for example, if regulators change the rules
of the game with respect to regulatory asset values, or apply discretion in deciding which assets should be included in the
regulated firm’s regulatory asset base.
 34

framework. This would involve a BOT contract for construction and operation of a new en-route
control center in Scotland.

This approach has been strongly criticized by the existing NATS management, who claim
that it would be significantly more costly than traditional public sector financing for three reasons:

• A higher private sector cost of capital.

• Additional bidding and transaction costs (which NATS claimed would add around two
percent to project costs).

• Loss of potential competition for subsequent upgrading contracts.

Both NATS and the CAA have argued that it would be preferable to privatize NATS as a
regulated utility subject to a price cap.

IV. MEASURES TO ENCOURAGE SUCCESSFUL

PRIVATE SECTOR PARTICIPATION

A. Introduction

This section reports on the existing experience of PSP in airports and in other types of
transport infrastructure, both in DMCs and elsewhere, in order to identify measures likely to encourage
successful privatization initiatives in the airports sector. The discussion is organized as follows:

• the importance of establishing a favorable public policy environment to encourage

PSP.

• the case for restructuring the airports industry in order to accommodate PSP.

• a discussion covering aspects of the legal and regulatory framework necessary to

secure the interests of key stakeholders following privatization.

• the allocation of risk between the state and privatized airports operators.

• measures to encourage efficient tendering outcomes where privatization is implemented

either through trade sale of leases or concessions.

• a review of options for engaging the private sector if the airport is unprofitable.

• the main lessons, both positive and negative to be learned from experience to date.

B. Public Policy

All other things being equal, the role of the private sector in financing and operating all types
of public utility infrastructure will develop most rapidly when the stance of public policy is openly and
consistently supportive, and recognizes that privatized provision of utility services should be regarded
as the rule rather than the exception.

How precisely this environment is achieved will depend upon the legal and administrative
traditions of the countries concerned. In UK, for example, the process of utility privatization
 35

developed in a relatively ad hoc fashion. The major privatizations of the mid 1980s, involving British
Telecommunications, British Gas, and BAA were each justified in terms of the specific circumstances
of the nationalized industries concerned, and an increasing recognition of the benefits of privatization
to the public finances. It was only in the mid-late 1980s that a coherent philosophy emerged regarding
privatized ownership of the utilities as the norm.

Other countries, including some in Eastern Europe and some DMCs, have taken a different
approach, and have created legislative frameworks to facilitate privatization across the utilities
sector. The Philippines, for example, has passed the so-called BOT law authorizing the financing,
construction, operation, and maintenance of infrastructure projects by the private sector. As amended
in 1993,18 the law begins with the following declaration of policy:

“It is the declared policy of the State to recognize the indispensable role of the
private sector as the main engine of national growth and development and to provide
the most appropriate incentives to mobilize private resources for the purpose of
financing the construction, operation and maintenance of infrastructure and
development projects normally financed and undertaken by the government. Such
incentives, aside from financial incentives as provided by law, shall include providing
a climate of minimum government regulations and procedures and specific
government undertakings in support of the private sector.”

As well as identifying the specific types of PSP covered by the legislation, other sections of
the law make provisions covering, inter alia:

• The duty of government agencies to identify candidate PSP projects and to give wide
publicity to the projects thus identified.

• Procedures for tendering and project award.

• Authorization for concessionaires to derive revenues from user charges or rentals

stemming from the possession and use of infrastructure assets.

• Arrangements for terminating or extending concessions.

• Arrangements for establishing Regulatory Boards, to supervise pricing and performance

of the private sector.

Similar provisions are contained in legislation on concessions in some Eastern Europe

19
countries such as Hungary, where the legislation enables concessions in public transport
(including airports), water services, telecommunication networks, energy/supply and postal
services. The Hungarian legislation has provided a framework for negotiating BOT motorway
projects, but so far has not been applied to the airports sector.

________________________________________________________

18
Republic Act 7718.
19
Act XVI, 1991 on concessions.
 36

C. Industry Restructuring

In most countries, the existing structure of the airports industry falls into one of the
following categories:

• A highly concentrated industry structure in which a national airports or civil aviation

authority or corporation owns and operates all major civil airports.

• A more fragmented industry structure, characterized either by local or regional authority

ownership of airports, or by a combination of state and local authority ownership.

If airport ownership is highly concentrated, the issue arises of whether, and, if so, how, the
industry should be restructured in order to accommodate PSP.

The case studies in Appendix 2 illustrate different approaches to this issue. In UK, for example,
privatization has taken place without any significant industry restructuring. The BAA, a state
corporation, which owned and operated a total of seven airports, three in south-east England and
four in Scotland, was privatized as a single entity, BAA plc. Although ownership of the non-BAA
airports is relatively fragmented, the industry remains highly concentrated by virtue of the
predominance of BAA’s south-east airports as international gateways.

At the other extreme, the airport sector in Australia has been radically re-structured as a
result of the privatization process. Prior to privatization, the FAC, a state corporation, owned and
operated an extensive network of international and regional airports. Following the privatization
process, which involved the sale of very long term leases on 17 of the principal airports, ownership
is now divided between 10 different consortia, with no single consortium controlling more than one
of the major international gateway airports so far offered for privatization.

In between these two extremes, the case studies of Thailand, Côte d’lvoire and Argentina
offer examples where privatization has been, or will be, accompanied by some degree of industry
re-structuring. The simplest case is in Côte d’Ivoire, where, prior to privatization, a national civil
aviation authority - Agence Nationale de l’Aviation Civile et de Ia Metéorologie -owned and operated
airports and ATC facilities throughout the country. Following the concessioning of the major airport
at Abidjan in 1996, the remaining airport and ATC activities were transferred to a newly created
state corporation, le Service Météorologique National de Ia Côte d’lvoire (SODEXAM), which also
acts as the conceding authority in relation to the Abidjan airport concession.

In Argentina, airports and ATC facilities were operated by the Argentina Air Force. Thirty
three of the principal airports, including the major international gateway at Buenos Aires were
transferred to a single private sector concessionaire in 1998; the remainder, many heavily loss-
making, were transferred to the regional governments concerned.

The most complex re-structuring is envisaged in Thailand, where major airports are
currently owned and operated by a state corporation, the AAT. The Thai Government plans to
offer a minority stake in a newly created joint venture company, the Airport Authority of Thailand
Co. Ltd., to a strategic partner. The Airport Authority of Thailand Co. Ltd., in turn, will be either
the sole or majority shareholder in two successor companies, the first responsible for operating
the existing international airport at Bangkok, and the second for developing and operating a
new international airport at Bangkok. AAT’s major regional airport assets will be transferred to
a newly created regional airport company, in which the majority shareholder will be a private
 37

sector partner, possibly the same as the strategic partner participating in the development of the
Bangkok system.

An important factor in this complex set of proposals is the need to preserve a state sector
majority shareholding in the entity charged with developing the new airport, in order to benefit from
a large soft loan facility, offered by Overseas Economic Cooperation Fund, to finance the new
airport construction, which is only available on a government-to-government basis.

The discussion in Section II indicates that there is little evidence of significant scale benefits
flowing from multiple airport operation; equally, however, there is little evidence of significant scale
diseconomies. The case for significantly reducing the concentration of airport ownership at
privatization therefore depends on the trade-off between the up-front and visible costs of re-structuring,
and possibly less tangible benefits of increased competition resulting from break-up. The competition
benefits in this industry are not clear-cut, primarily because major airports mainly serve distinct
regional markets. Even in UK, where BAA’s south-eastern airports were clearly serving broadly the
same regional market, competitive pressures at the time of privatization were weakened by traffic
distribution rules (although these have since been abandoned), and, more fundamentally, by the
dominant position of Heathrow within the system (although this has possibly weakened somewhat
as a result of traffic growth at the other airports).

In UK, the authorities took the view that any potential competition gains from breaking up
BAA prior to privatization would have been offset by restructuring costs. In Australia, by contrast,
the government has preferred to restructure and reduce industry concentration radically, emphasizing
the following public policy benefits of the approach:

• The possibility of some limited inter-airport competition for long haul international traffic.

• The benefits of fragmented ownership in generating yardstick evidence, enabling

regulatory agencies to assess individual operator performance more effectively; and in
introducing a limited element of competition by emulation between operators.

D. Regulatory Institutions and Practice

Irrespective of whether PSP in airports and ATC involves full or partial privatization,
arrangements are needed to safeguard the interests of airport users and to balance the long-term
interests of users and investors.

1. Regulating Fully Privatized Airports Operators

The economic regulation of fully privatized airports involves the imposition of constraints,
either on the maximum prices charged, or on airport profitability, as measured by the rate of return
on capital. Both UK and Australia have preferred to regulate prices rather than profitability, in the
belief that this form of regulation will encourage better performance outcomes, in particular, in
respect of cost efficiency.

In UK, the primary legislation, the Airports Act, 1986, identifies a set of licensed airports,
where the license conditions impose obligations on the airport authority relating to the safe
operation of the facilities, and delegate powers to the authority, such as the power to set local
by-laws needed to fulfil these obligations. Within the set of licensed airport operators, the
Airports Act also specifies a minimum threshold scale of operations, currently £1 million of
 38

turnover,20 above which the airport operator is subject to economic regulation, in the sense that it
must apply to the regulator for permission to levy charges. Within the airports subject to economic
regulation in this way, secondary legislation identifies a further subset of four airports, London
Heathrow, Gatwick and Stansted, all owned by BAA, and Manchester International Airport, where
the airport license includes conditions relating to the maximum level of charges.

The rationale for only designating four airports as subject to price regulation offered when
the legislation was passed was that charges at other airports would be constrained by competition
from other airports, including those subject to price regulation, the provisions of existing UK and
European Union (EU) competition policy legislation, and the threat of designation.

A broadly similar approach has been followed in Australia, where price cap regimes
have been applied at major international gateway airports and at a number of larger regional
airports.21 At smaller airports, the regulatory agency is able to determine charges if the airport
operator is unable to agree charges with airline customers.

Another common feature of economic regulation in UK and Australia is the establishment

of an independent regulatory agency, separate from the executive departments of government.

In UK, responsibility for the economic regulation of airports has been vested in the CAA, a
specialized sectoral regulator, which also owns and operates ATC facilities, and exercises safety
and economic regulatory responsibilities in relation to UK registered airlines. In Australia, the task
of setting and reviewing airport price caps and of monitoring service quality performance and
compliance with other contractual obligations has been given to the Australian Competition and
Consumer Council, which has wider responsibilities as a competition policy agency.

In both UK and Australia, the decisions of the regulatory agency in respect of the maximum
prices to be allowed, are taken in the light of certain public interest criteria, specified in primary
legislation. For example, Section 39 (2) of UK Airports Act, 1986, states that the CAA shall perform
its functions so as to:

• Further the reasonable interests of users of airports within the UK.

• Promote the efficient, economic and profitable operation of such airports.

• Encourage investment in new facilities in time to satisfy anticipated demands by the

users of such airports.

• Impose the minimum restrictions that are consistent with the performance by the CAA
of its functions.

The criteria in Section 39 thus recognize that whilst the regulatory agency must protect
the interests of airport users, it must also balance the interests of consumers and investors, to
encourage continuing investment in new facilities.

________________________________________________________

20
To be altered to 1 million passengers per annum under current government proposals.
21
The following 12 airports are subject to price regulation:
Darwin, Townsville, Brisbane, Coolangatta, Sydney, Canberra, Melbourne, Adelaide, Alice Springs, Perth, Hobart, and
Launceston.
 39

The Airports Act also determined the procedures for carrying out periodic reviews of airport
charges at airports subject to price cap regulation. Under these provisions, the Monopolies and
Mergers Commission (MMC) was given responsibility for reviewing performance under the existing
price cap and on any other aspects of airport operator conduct referred to it by the CAA. The MMC
review process involved extensive interchange of information and opinion with the regulator and
other parties, such as airlines. On the basis of the results and expectations about future market
developments, the MMC then recommended, in a published report, a new quinquennial price formula.
The CAA was given final responsibility for determining the price formula in light of the MMC
recommendations, although it was not bound to accept these recommendations. Once again, the
GAAs decision and its justification was published.

This set of institutions, conduct rules and processes scores highly in terms of five of the
six criteria of good practice in regulatory governance identified in a recent NERA report to
ADB. 22 Specifically it ensures:

• Clarity in the role and objectives of the regulating agency.

• Regulatory autonomy.

• Effective participation by regulatees and other interested parties, such as airlines, in the
regulatory process.

• Regulatory transparency.

• Regulatory accountability.

It did not ensure regulatory predictability, since the MMC was not bound by its own previous
decisions. However, it can be argued that other characteristics of the review process, notably
clarity of objectives and regulatory transparency, at least help to reduce the likelihood of inconsistency
of approach between reviews. In practice, in contrast to the position in some other UK regulated
industries, it appears that the regulatory authorities have so far taken a consistent approach at
successive periodic reviews over matters such as regulatory asset valuation.

The roles assigned to the MMC and the sector regulator (the CAA) under the 1986 Act
were, in fact, distinctive in relation to practice in other UK regulated industries, where the regulatory
agency, such as the Office of Telecommunications, was responsible for conducting the periodic
review and making proposals for resetting the price cap. The MMC would act as an appeals body in
the event that the regulated company and the regulator could not reach agreement on a revised
price cap formula.

Following a recent review of utilities regulation, the Government has announced

proposals for bringing the conduct of future periodic reviews in the airports sector into line
with practice in the other industries. The CAA, in common with other such regulators, will be
responsible for reviewing outcomes and making proposals for resetting the price cap, with the
MMC acting as an arbitrator if the parties cannot reach agreement. In the event of an MMC
referral, the MMC’s report will be published.

________________________________________________________

22
NERA (1998), op cit.
 40

2. Economic Regulation under Partial Privatization

Under partial privatization involving a concession, the government (or its agent) acts as a
purchaser of the services provided by the private sector contractor, following a process of competition
for the market, and the resulting contract terms act as the primary instrument of regulation. The
case studies of airport concessions in Appendix 2 include examples of different approaches to
regulating concessionaire conduct.

In the Côte d’lvoire, the terms of the concession agreement require the concessionaire,
AERIA, to consult with the conceding authority (SODEXAM) before adjusting airport charges.
However, the conceding authority is obliged under the concession agreement to sanction any
increases in charges necessary to maintain the concessionaire’s financial equilibrium, defined in
the agreement as the full recovery of operating and financing costs. If approval is withheld, then the
government must compensate the concessionaire directly. This framework therefore corresponds
closely to a rate of return or cost of service form of regulation, which is generally believed to weaken
the cost efficiency incentives of the concessionaire. The concession agreement appears to address
the resulting incentive problem by envisaging a high degree of interaction between the concessionaire
and the conceding authority, with annual negotiations on tariff adjustments and other operational
aspects of the concession agreement, similar to those that might occur between a state-owned
enterprise and its sponsoring ministry.

In the cases of Argentina, Colombia, and the Philippines the concession contract contains
a price cap mechanism to restrict the maximum permitted rate of increase in airport charges. In
contrast to cost of service regulation, the price cap approach offers strong incentives to the
concessionaire to be cost-efficient, reducing the need for the intrusive regulation that appears to
characterize arrangements in the Côte d’Ivoire. However, airport concessions are typically very
long lived, and arrangements must be made to monitor regulatee performance under the contract,
especially over matters such as the timely completion of investment projects, and service quality,
to review and revise financial and other contract terms in the light of outturn experience (which may
differ significantly from assumptions on which the contract was based) and, finally, to discourage
opportunistic behavior by government.

As shown in the section below, contractual mechanisms that allocate risks efficiently between
the parties can significantly reduce the need for formal contract review mechanisms. It can also be
argued that the issues posed by uncertainty and opportunism are less important in the aviation
sector than in other types of transport infrastructure, such as roads or light rail transport systems.
Uncertainty over future demand is arguably less of a problem for airports (and ATC) because there
are few substitutes and demand is not highly price sensitive. Also, government has fewer incentives
to behave opportunistically over airport charges than over user charges in other sectors, whose
impact is more widely felt by domestic consumers (and voters).

However, governments may behave opportunistically over other terms of the concession
agreement than those relating to airport charges. An example of this has occurred recently in
Cambodia, where the concession agreement at Phnom Penh airport included a Government
commitment to designate Phnom Penh as the sole international gateway for Cambodia. 23
Subsequently, the Government has reneged on this commitment by permitting international
flights to land at an airport elsewhere in Cambodia, and it is not yet clear whether some

________________________________________________________

23
Similar provisions are included in the Ninoy Aquino International Airport BOT agreement. See Appendix 2.
 41

compensating adjustment to the terms of the concession has been negotiated between the
government and the concessionaire.

It seems certain that actions of this kind will adversely affect the private sector’s future
willingness to invest in Cambodia. Such effects might have been mitigated by the existence of an
independent arbitrator, armed with clear conduct rules, to whom the concessionaire could have
appealed in the event of a failure to agree an appropriate adjustment to contractual terms with the
Government.

Assessing current regulatory practice in six DMGs (Bangladesh, India, Indonesia, Malaysia,
24
Pakistan, and the Philippines) against criteria of good practice, NERA’s recent report to ADB
found that whilst there was evidence of a clear trend towards generally more effective regulatory
governance, there was as yet:

“no clear evidence of convergence to best-practice or any other common solution.

It remains to be demonstrated whether international best-practice regulatory
frameworks can and do work effectively in Asian DMCs.”

As recent experience in Cambodia, Côte d’Ivoire, Gabon, the Philippines, and in several
Latin American countries, illustrates, imperfections in regulatory governance do not prevent PSP in
airports, given certain favorable characteristics of the market environment (strong growth in demand,
strong market position of hub airport operators). However, they may well reduce the flow of private
sector investment in infrastructure projects, and worsen the terms on which the private sector is
willing to invest.

E. Contract Design and Risk Management

An important aspect of good practice in the concession model of PSP is the management
of risk through contract design. It is necessary at the outset to define what is meant by risk
management in this context. Mainstream finance theory distinguishes between systematic risk,
which refers to the relationship between the variation in project returns and variations in the average
return across a wide portfolio of assets (the market return), and the specific risk of a project, which
refers to the variability in the expected return to the project. Mainstream theory emphasizes the role
of systematic risk in determining the cost of equity capital, under efficient capital markets.

1. Sources of Risk

The term risk management as used in the present context is distinct from these more
familiar concepts of risk, and can perhaps best be understood by considering the set of variables
which determine the expected present value of an airport investment project to a potential
contractor. These are shown in Table 5, together with some of the factors affecting outcomes
in respect of each variable. With the exception of exchange rate risk, where outcomes are
entirely exogenous, outcomes under the determining variables reflect a combination of external
factors and contractor performance.

The objective of efficient contract design is to allocate risk to the party best able to
manage it. This means that the government may protect the concessionaire from certain types of
risk, such as risk arising from planning delays or changes in externally imposed safety or

________________________________________________________

24
NERA (1998), op cit.
 42

security regulations whilst leaving them exposed to factors reflecting their own performance,
especially those relating to costs.

Table 5: Sources of Risk

Variable Risk Factors

Construction cost/time Contractor performance
External - planning/legislative time scale
- regulatory factors
- safety/security regime
- purchaser behavior (change of specification).
- inflation
- exchange rate (for imported equipment)
Operating cost Contractor performance
External - change in security regime
- inflation
Demand/revenues Contractor performance
External - charges
- macroeconomic factors
- airline industry factors
- inflation
Financing cost Contractor performance
External - market perception of project risk
- investment bank policies
Exchange rates External

2. Managing Risks

The approach now widely adopted in concession projects in the airports sector, and in
other types of transport infrastructure, is to identify a range of external risk factors, such as those
shown in Table 5, and to specify compensating adjustments in the financial terms of the contract in
respect of pre-defined outcomes under each factor. For example:

• Contractors are not normally exposed to planning risk, which covers a variety of
contingencies that arise because large transport infrastructure projects, such as major
airport developments, are prone to suffer both delays and cost increases as a result of
planning enquiries or legislative processes. Thus, if planning processes lead to a
requirement for greater expenditure on environmental protection than was anticipated
when the contract was negotiated, the contractor would normally benefit from a
compensatory adjustment to contract terms.

• Similarly, contractors are also protected against general inflation risks, though price
indexation clauses in contracts (including price cap mechanisms).
 43

a. Exchange Rate Risks

The recent Asian economic downturn has highlighted the potential importance of exchange
rate variability and demand uncertainty as determinants of project viability. Exchange rates affect
viability because user charges, including airport charges, are set in local currencies, whereas a
proportion, possibly the majority, of the concessionaire’s liabilities may be hard currency
denominated.

To the extent that exchange rate fluctuations reflect, or cause, variations in the rate of
domestic inflation relative to global inflation, indexation of output prices in local currency offers a
measure of protection against exchange rate risk. A more extreme solution would be to have airport
charges denominated in a hard currency, such as US dollars. Whilst this would still leave the
operator exposed to currency risk in relation to services from commercial activities, such as retailing
and car parking, many of the costs arising in these activities would also be denominated in local
currency.

As noted above, setting charges in a non-local currency is admissible under ICAO

recommended procedures:

“in special circumstances, for example, where economic conditions are not
stable.”25

According to the evidence on current airport charging practice, several DMCs now set
many or all airport charges in US dollars. Moreover, the only two DMCs in the group shown in Table
2 to have implemented airport concessions to date, Cambodia and the Philippines, have each
adopted a policy of dollar denomination of charges.

b. Demand and Revenue Risk

The case studies and experience in other types of transport infrastructure project have
identified widely differing approaches to the treatment of demand and revenue risk in concession
projects:

• At one extreme, some concession contracts allocate demand and revenue risk entirely
to the project promoter.

• Elsewhere, a range of contractual mechanisms has been devised to share revenue

risk between the government purchaser and the project promoter.

Contractual mechanisms where revenue risk remains with the contractor take one of two
forms:

• The concessionaire pays a once-for-all or annual concession fee to the government

that is fixed in real terms, or if the project is unprofitable, the concessionaire receives
a fixed payment from government. Projects are awarded to the concessionaire
offering the highest (least negative) fee. Examples of this type of agreement include
the Argentine airports concession, where there is a constraint on the maximum level
of charges that can be set by the concessionaire, and the original concession

________________________________________________________

25
ICAO, Statement by the Council on Airport Charges, Section I C.
 44

agreement for the Channel Tunnel Rail Link (CTRL) in UK, where the market was
judged to be competitive, and so no constraint on charges was imposed.

• The concessionaire retains project revenue for a fixed term and the contract is awarded
to the bidder offering the lowest level of user charges (a Chadwick auction mechanism).

Fixed fee or revenue retention contracts of this kind are favored because they provide the
strongest possible incentives for the concessionaire to perform effectively. The potential difficulties
which they introduce are well illustrated by the recent experience with the CTRL concession. Fixed
price bids for the project were sought at a time when future demand was highly uncertain. In the
event, realized demand has been far lower than that projected by the bidder, and the agreement
has had to be abandoned, at considerable expense to both the concessionaire and to the UK
Government. The outcome exemplifies the “winner’s curse” problem characteristic of this type of
bidding process, since the winning bid was based on a highly optimistic view of the level of future
demand for the project. It can be argued that the level of demand uncertainty in airport development
projects is generally less significant than for projects, such as CTRL or roads, where there may be
far closer substitutes available, and where realized demand may be greatly affected by competitor
response.

A range of contractual devices are available for sharing demand and revenue risk. The
airports concession projects covered in the case studies illustrate three variants of this approach:

• The most straightforward is when, as in Côte d’lvoire, the concession agreement is

based on cost of service regulation. Given price inelastic demand for airport services,
this approach would enable the concessionaire to increase charges in real terms to
whatever extent was necessary to restore financial equilibrium, allowing market risks to
be fully transferred to airport users.

• A government minimum revenue guarantee (Bogota). This approach is conceptually

similar to devices such as minimum offtake agreements, observed in concession
contracts in the power sector. The Bogota project was awarded to the bidder seeking
the lowest guaranteed level of revenue and the lowest average landing fee.

• Concession fees variable with airport revenue (the Philippines). In this case, the
concession fee payable to MIAA contains both a fixed and a variable element, the latter
expressed as a proportion of total revenue accruing through passenger terminal
operations. In this case, the contract has been awarded to the bidder offering the highest
expected concession fee revenue, with the revenue being evaluated across a range of
demand scenarios.

Other risk sharing devices observed in transport sector concession projects include:

• Variable concession lengths; this approach has been applied in a number of tolled
motorways and bridges. If the present value of concession revenue reaches a certain
pre-determined level before the maximum term of the concession, which is
specified, the concession is terminated. This approach leaves some residual revenue
risk with the concessionaire, but the maximum concession length is usually based on
a pessimistic view of future traffic levels, so that this residual risk is very limited in
practice.
 45

• Shadow tolling; in UK design-build-finance-operate program, the concessionaire’s

revenue varies with the level of traffic. However, bids are expressed in terms of a declining
block tariff structure, such that the marginal rate of shadow toll payment falls with
successively higher levels of traffic. The variation in revenue is damped, and should
broadly match the variation in operating costs with traffic at the margin. Although this
approach has so far only been used to underpin concession projects involving non-
tolled facilities, it could equally well be applied to airport runway projects, such as the
Bogota El Dorado airport concession. In this case, the public sector airport operator
would retain landing charge revenues, and the concessionaire would be remunerated
through the shadow toll mechanism.

The impact of revenue risk sharing measures on the terms potential private sector contractors
are willing to offer (given expected market conditions), will depend upon the precise nature of the
measure. Two effects can, in principle, be identified.

First, any such action reduces the specific risk in a project, either by curtailing downside
risk, or by capping both downside and upside risk. Although orthodox finance theory would suggest
that changes in specific, i.e., diversifiable, risk should not affect the terms on which capital markets
are willing to supply capital to a project, it is not clear that the assumptions underpinning the theory
are completely satisfied given the infant industry character of airport concession projects. If not,
then a reduction in specific risk may reduce the required rate of return sought by potential
concessionaires.26

Second, measures such as minimum revenue guarantees, that underwrite downside risks
without capping upside risk, increase the expected private net present value of the project, although
the expected social net present value is unchanged. In itself, this would tend to improve the private
sector’s willingness-to-pay for the project, given expected market conditions, etc.

The potential downsides of contingent contracting and bidding mechanisms are, first, that
they may weaken the contractor’s incentives to perform effectively, by maximizing demand and
revenue and by completing projects to time, and, second, that the bid evaluation procedures may
be more complex than under fixed fee/term contracting, since in some cases, bids must be evaluated
across a range of demand scenarios.

The emerging consensus in favor of different forms of contingent approach suggests that
market participants expect the potential downside effects to be outweighed by the potential risk
sharing benefits of the contingent approach.

3. Other Aspects of Contract Design

A frequently cited problem with time-limited concessions in industries such as airports,

where the assets are specific and long-lived, is that the economic life of the assets acquired by the
concessionaire may be far longer than the concession period. Because there is no guarantee
that the concession will be renewed, the concessionaire must seek to amortize such assets
over the remaining term of the concession, and to recover these costs from user charges. As
a result, the cost base is inflated compared to the full privatization model; the cost inflation
impact is especially acute for assets acquired later in the concession period.

________________________________________________________

26
For further discussion, see Jones, I, Zamani, H, and Reehal, A (1996), Financing Models for New Transport Infrastructure,
Luxembourg, Office for Official Publications of the European Community.
 46

Provisions in the Abidjan airport concession agreement, which are similar to recent
developments in water industry concessions, offer the prospect of mitigating, if not altogether
removing, such an effect. The approach in these agreements is to make provision in the concession
contract for paying compensation to the concessionaire based on the residual value of assets not
fully depreciated under industry standard accounting rules, if the concession is terminated or
transferred to another party following a rebidding process at the end of the concession period. The
impossibility of fully mitigating the effects stems from the need for the conceding authority to protect
itself against underwriting unwise or opportunistic investment by the concessionaire. Thus, water
industry compensation terms may incorporate concepts such as “used and useful”, which introduce
some element of ex-ante uncertainty for the concessionaire as to what the conceding authority will
offer. Uncertainty of this kind may be limited by provisions offering arbitration if agreement cannot
be reached between the parties on a fair valuation of stranded assets, as in the Abidjan airport
concession agreement. The introduction of explicit provisions for dealing with stranded assets in
airport concession contracts involving more than the construction and operation of a particular
facility, such as a runway, thus offers scope for improved performance outcomes.

F. Securing Efficient Tendering Outcomes.

The net benefits of engaging the private sector through concessions and other forms of
contracting arrangement will generally be higher the more effective is competition for the market,
which, in turn, is linked to the effectiveness of public procurement procedures. Experience in the
EU, where public procurement throughout the Union is subject to a series of EU Directives,27
indicates that good practice covers the following aspects of procurement:

• The absence of restrictive (or preferential) conditions on bidder eligibility (in the case of
the EU directives, this takes the form of banning discrimination on grounds of nationality).

• Transparency in selecting potential contractors and awarding contracts, secured by

- the requirement to publicize any contract whose estimated value exceeds a specific
threshold (which will vary according to the nature of the goods or services being
purchased). In this way, potential contractors are kept well informed about possible
opportunities;

- the use of objective criteria which must be known beforehand in order to prevent a
contracting authority from selecting candidates and tenders on the basis of criteria
different from those initially stated;

• a precise indication of which of the permissible award procedures has been chosen:

- an open procedure, in which any firm may tender;

- a restricted procedure, in which only firms that have been invited to tender by the
contracting authority may do so;

________________________________________________________

27
Directive 93/36/EEC (public supplies); Directive 93/37/EEC (public works); and Directive 92/50/EEC (public services). A further
directive, 89/665/EEC, covers measures to ensure compliance with the primary directives.
 47

- a negotiated procedure, in which the contracting authority consults selected firms

and negotiates the contract terms with one or more of them; and

• compliance with explicitly stated technical requirements standards.

The case studies in Appendix 2 offer examples of widely differing levels of competition
either for concession contracts or, in the case of airport privatization in Australia, for airport leases.
In Côte d’lvoire, procedures for awarding the concession at Abidjan airport were informal, in the
sense that the initial call for expressions of interest was not tightly specified, and only limited financial
accounting data were available on the performance of the airport. This was followed by an extended
process of negotiation, initially with two, but finally with a single favored bidder, selected on the
basis of the overall quality of the expression of interest.

In Australia, the airport privatization program attracted a great deal of interest and the sales
proceeds were higher than had initially been expected. Several factors appear to have contributed
to the success of the privatization program:

• The establishment of a regulatory framework and process which set down clear
guidelines within which the newly privatized airports would operate.

• The provision of as full as possible disclosure of information on the current and expected
performance of the airport businesses.

• The use of transparent tendering procedures setting out clear and unambiguous rules
by which tenders would be assessed.
The first two factors meant that prospective bidders had as clear as possible a view of the
commercial potential of the businesses; the third factor ensured that bids were well-focussed and
observed the same rules of the game, enabling the authorities to make decisions that were defensible
in terms of the specified criteria.
The Philippine BOT law, referred to earlier, contains provisions covering procedures for
tendering and awarding BOT projects, relevant extracts from which are in Appendix 3. These
provisions score highly in requiring public authorities wishing to award BOT contracts to publicize
the opportunity widely, and in setting out admissible procedures and decision criteria leading to the
award of contracts.
As described in more detail in Appendix 2, the BOT project to construct and operate a
second international terminal at NAIA resulted from an unsolicited bid, admissible under Section
4-A of the BOT law, in respect of financially viable projects. This provision recognizes that
public sector agencies may well not be able to identify all potentially viable projects, and offers
scope for entrepreneurial initiatives from the private sector. However, it is also important to
ensure that the projects promoted through this process are economically viable, on the basis
of comprehensive and rigorous cost benefit and financial analyses. In the case of the new
international terminal at NAIA, doubts were initially expressed about the economic justification
for the project, first, because it appeared to conflict with existing plans to develop a new
international gateway airport for the Manila region at Clark Airforce base, and, second, because
the project will require the premature closure of another existing terminal. While subsequently
it was demonstrated that the project was economically viable, in other circumstances, issues
of this kind would tend to weaken the private sector’s willingness to bid for a project, because
they would increase the risk that the Government might not continue to support the project.
Conversely, the availability of Government-sponsored studies demonstrating a strong case for
 48

the project would signal to prospective private sector bidders that the Government would be
committed to the project.

The Philippine law requires the purchasing agency concerned to seek competitive bids for
any project for which an unsolicited bid has been received. In this case, a competitive bid was
submitted, offering very significantly better terms than those offered in the unsolicited bid, which
the original bidder felt unable to match. As a result, the BOT contract was awarded to the competitive
bidder.

G. PSP in Unprofitable Airports

There are several possible mechanisms for engaging the private sector in unprofitable
airport operations, defined as a situation where the expected present value of future airport revenue
streams is less than the expected present value of future operating costs, including both current
and capital expenditures.

First, the airport may be offered to the private sector with an element of government subsidy,
injected either through a negative concession fee mechanism, or by a grant towards the cost of a
capital project. This approach has been extensively applied elsewhere in the transport sector,
notably in the provision of unprofitable rail or bus services, but there is no evidence of its application
in the airport sector.

An alternative approach, which removes the need for any direct or explicit subsidy payment
from government, is to bundle unprofitable airport activities with profitable airports, or possibly with
property or commercial development opportunities, where the activities are closely complementary
to the airport. The Argentine airport privatization program provides an example of the bundling of
profitable and unprofitable airports within a single multi-airport concession, which preserved the
pattern of cross-subsidy within the existing state-owned entity. The authorities in Côte d’lvoire are
currently examining options for developing the airport at San Pedro, in the south-west of the country,
in conjunction with a major expansion of tourist facilities in the region; one possible approach would
be to offer a concession covering both the airport and hotel development, etc.

The Australian authorities also considered, but rejected, the idea of bundling groups of
airports together in reviewing structural options for the privatization program. In doing so, the
Government took the view that it would be more efficient to allow the ownership structure to be
determined by the market, rather than imposing a structure ex-ante. The availability of incentive
compatible mechanisms for allocating subsidy through a competitive bidding process, added to
the public policy benefits of increasing the transparency of subsidies, further reinforces the
advantages of the unbundled approach to engaging the private sector.

H. Conclusions: The Main Lessons from Experience of

Airport Privatization to Date

In drawing lessons from the existing experience of airport privatization, it should be

emphasized that the extent of this experience remains strictly limited. The flotation of BAA plc,
the first major airport privatization episode, occurred in 1986, and it is only in the past 2-3
years that PSP, in what we characterized earlier as core airport activities, has become at all
widespread, chiefly as a result of developments in Australia and Latin America.
 49

The key lessons to be learned from experience to date are summarized in Figure 2.

Figure 2: Key Lessons

Public Policy
• Airport privatization will be encouraged by the existence of legislation, in the form of a BOT law or similar,
signaling the state’s recognition of the need for private participation in infrastructure provision.
• It is also important to ensure that government is able to demonstrate that any projects offered to the private
sector are economically viable.

Industry Structure
• In the absence of any significant scale benefits from multi-airport operation, there are advantages from using
the privatization process as an opportunity for reducing high levels of industry concentration.
• Equally, the existence of unprofitable airports does not justify the maintenance of a highly concentrated
industry structure to facilitate cross-subsidy.

Regulatory Framework and Mechanisms

• The process of PSP will be encouraged by the existence of good regulatory governance structures, based on
independent regulatory agencies, operating within well defined public interest criteria, and with well-articulated
appeal or arbitration mechanisms.
• Such a framework is equally relevant under both full and partial (concession based) privatization scenarios.
• The price cap approach to constraining airport charges is likely to encourage better performance outcomes
than one based on rate of return regulation.

Risk Allocation
• Some sharing of revenue or market risks between a concessionaire and government may offer a better deal
for the purchaser than full transfer of such risks to the concessionaire.
• Denominating some or all airport charges in US dollars may be a useful device for encouraging PSP in
airports in the wake of the recent currency crisis.
• There would be advantages in the widespread adoption of mechanisms already present in some concession
agreements for compensating concessionaires for stranded assets in the event that the concession is
terminated or transferred to another party when it is rebid.

Tendering Procedures
• Competition for the market will be encouraged by a clear and transparent tendering process, based on equal
treatment of bidders and full disclosure of information to enable bidders to make as informed an assessment
as possible of the business opportunity.

V. THE ROLE OF ADB

A. Introduction

This concluding section discusses ADB’s role in facilitating PSP in airports and ATC services
in DMCs. In this section the report examines how ADB can assist in disseminating good practice,
both in respect of the mechanics of contracting, and in the wider context of public policy and
regulatory structures and conduct. Further, ADBs role in providing training for officials in DMC
government agencies, to equip them for the tasks of negotiating contracts with private sector
organizations is considered. Finally, the discussion shows how ADB’s actions can improve
the terms on which the private sector is willing to participate in financing and operating
infrastructure facilities.
 50

B. Disseminating Good Practice

Through this medium and other projects undertaken under ADB’s regional technical
assistance umbrella, ADB can provide decision makers in DMCs with authoritative and impartial
assessment of the rapidly evolving experience worldwide of PSP in infrastructure development. In
the case of the airports sector, much of the experience currently available relates to the type of
arrangements that are feasible. There is as yet little substantial evidence on what particular
arrangements are likely to produce better performance outcomes in practice than others. However,
there are some aspects of public policy, industry structure, regulatory practice, contract design
and public procurement procedures, where the elements of “good practice” are now visible.

C. Role in Training

Transforming the basis for procuring infrastructure services from the traditional publicly
financed design and build model (with subsequent operation by the public sector agency) to a
private financing and operation model, requires significant changes in the procurement and project
management activities of public sector agencies. To make BOT arrangements work effectively, the
public sector must develop new capabilities in contract negotiation, especially in the area of risk
allocation and management, and in the regulation of contract performance in the operational phase,
which extends over many years, and may include contract review and renegotiation.

In the short term, shortage of necessary skills may be covered by hiring external advisers.
However, in the long run it will generally be more cost effective to develop internal agency skills,
especially if there is a steady flow of new projects to be negotiated. Moreover, parameters of the
procurement process should become increasingly standardized on whatever model emerges as
best practice in light of experience. The more routine the process, the more efficiently it can be
managed by bureaucratic rules and procedures within public sector agencies.

Experience with programs such as the Private Finance Initiative in UK suggests that
developing the necessary agency skills can be achieved by a combination of re-training of existing
agency staff and external recruitment, although external recruitment may be problematic because
of disparities between public and private sector remuneration.

Agencies such as ADB can act both as direct training providers, especially during the early
pump-priming stages of the process of transforming public procurement strategies, and through
initiatives to encourage indigenous training capabilities. These include supporting the development
of “model” training programs, facilitating secondment of staff from DMCs to agencies in other
countries with greater experience of private sector provision of infrastructure and assisting in the
establishment of training centers in the DMCs, such as the BOT Center in the Philippines.

D. Improving Financing Terms

ADB has a fivefold role in improving the financial terms on which the private sector is willing
to provide airport and ATC infrastructure services in DMCs:
 51

• First, by encouraging the corporatization of existing state-owned providers of airport

and ATC services, so that potential private sector investors are better able to understand
the business proposition in which they might be engaged.

• Second, by promoting good practice in regulatory governance and public procurement.

• Third, by promoting good practice in contract design, to encourage the efficient allocation
of risk between the DMC government agencies and private sector contractors

• Fourth, by assisting DMC agencies to bring forward economically viable projects which
attract widespread support from relevant constituencies within the DMC.

• Finally, ADB’s willingness to invest in or sponsor a project can act as a signaling device
to private investors about the soundness of the project, which, in turn, should improve
the terms on which external investors are willing to participate.
APPENDIXES
 Appendix 1, page 1

EXAMPLES OF PRIVATE SECTOR PARTICIPATION

AT COMMERCIAL AIRPORTS WORLDWIDE

A. Full Privatization

The private sector entity is responsible for the ownership, investment, management and
operation of airport infrastructure. Privatization is implemented mainly through flotation or trade
sales.

Table A1.1: Examples of Full Privatization Schemes Worldwide

Country Plans or actions for airport privatization

Australia 17 airports sold on long-term leases of 50 years, with an option for additional 49 years.
Bahamas Transferred ownership of Freeport International Airport to a private entity.
Dominican Republic Transferred ownership of Puta Cana International Airport to a private entity.
New Zealand Implementing trade sale of Auckland and Wellington International Airports.
Slovakia Plans to sell Bratislava Airport to a private entity.
United Kingdom Share offering in British Airport Authority responsible for the operation of seven airports.
Local government sold Belfast International Airport to a private company formed by
employees.
UK regional governments have sold East Midlands International airport to a private
entity and are planning to sell shares in Birmingham Airport.

B. Partial Privatization

1. Concession or BOT Schemes

Under a concession or build-own-transfer (BOT) contract, a private entity finances, builds

or modernizes a facility before operating the facility and gaining revenue from the operation. After a
certain period, ownership of the facility transfers to the government. There is often some overlap
between BOT schemes and long term management contracts as in both cases investment is
required from the private entity.

Table A1.2: Examples of BOT/Concession Schemes Worldwide

Country Plans or actions for airport privatization

Algeria Plans to construct and operate new international terminal a Houari Boumedienne
Airport near Algiers.
Argentina Long term concession agreement implemented for operation and development of 33
of the country’s airports.
Bolivia 25-year master concession offered for running of the major three airports in Bolivia.
Assets remain with the government throughout.
Bulgaria Plans a 30-year BOT scheme to upgrade Sofia International Airport.
Brazil Plans a contract with a private entity to rehabilitate the terminal at Guarapes International
Airport in Recife.
Cambodia 25-year BOT contract for upgrading of Pochetong Airport in Phnom Penh. Plans for a
15-year BOT contract for projects at Sihanoukville Airport on Naga Island.
Cameroon Plans for a long-term lease to build and operate a terminal at Yaound Airport.
 Appendix 1, page 2

Table A1.2: Examples of BOT/Concession Schemes World Wide (continued)

Country Plans or actions for airport privatization

Canada Private entity implemented to build and operate a third terminal at Pearson International
Airport in Toronto on a long-term lease basis.
Chile Plans a 15-year BOT contract with a private entity for a second terminal at Arturo Merino
Benitez International Airport in Santiago.
People’s Republic of Planning to contract with private entities to develop and operate eight airports.
China
Columbia Awarded a BOT contract to private entity to build second runway and operate both runways
at El Dorado International Airport, Bogota.
Côte dIvoire 15-year concession granted for Abidjan Airport.
Costa Rica Plans a BOT contract with private entity for new airport in San Jose.
Ecuador Plans to combine BOT contracts for two new airports in Quito and Guayaquil with operation
of existing two airports in same cities.
Egypt Plans BOT project for new airport near Cairo.
Gabon 30-year concession granted for Libreville Airport.
Germany Considering contracts with private entities to develop and lease airports, including a
major Berlin Airport.
Greece Implementing a 30-year BOT contract for a new airport near Athens.
India Considering contracting with private entity for construction and operation of new airport
in Bangalore.
Malaysia Implemented a BOT contract for a new terminal and a lease-develop-operate contract
for non-aeronautical portions of new international airport in Sepang.
Mexico Considering long term leasing of 58 airports to private entities, following national
legislature passing a bill to allow these leases.
Myanmar Plans BOT contract for the new Hanathawaddy Airport near Rangoon. Panama Plans a
10-year contract to expand and maintain passenger and cargo facilities at Tocumen
International airport near Panama City.
Pakistan Plans BOT scheme for a new terminal at Lahore International Airport.
Peru Implemented a lease to build and operate a terminal and runway at Jorge Chavez
International Airport in Lima.
Philippines Has agreed long-term BOT agreement with a private entity for a new terminal at Ninoy
Aquino International Airport in Manila.
Qatar Plans a BOT contract for new international airport in Doha.
Russia Implementation of private concessionaire to design, finance, construct, own and operate
a new international terminal at Pulkova Airport in St. Petersburg, with initial 49-year lease
period.
Trinidad and Tobago Implementing BOT contract for a new terminal at Piarco International Airport.
Turkey Implemented BOT contract with 3 year 9 month operational period for new terminal at
Ataturk International Airport near Istanbul.
Venezuela Plans long-term contract to build, operate and manage a new airport between Bolivar
City and Guayana City in Eastern Venezuela.
Viet Nam Plans BOT contract for new international passenger terminal at Tan Son Nhat International
Airport in Ho Chi Minh City.
Uruguay Plans a 20-year contract with a private entity to expand the terminal, build a new
runway and make other improvements at Laguna del Sauce International Airport
near Maldonado.
 Appendix 1, page 3

2. Strategic Partners/Partial Divestiture

A private entity or investors may have majority or minority holdings in an airports

company, act as a partner in a joint venture to provide extra investment capital for infrastructure
development or act as a strategic partner to increase the efficiency of existing operations.

Table A1.3: Examples of Strategic Partner/Partial Divestiture Schemes Worldwide

Country Plans or actions for airport privatization

Austria Shares sold in Vienna Airport; 47 percent of total shares are privately held.
People’s Republic of Implementing a joint agreement to build and operate a new airport in Haikou.
China
Denmark Shares sold in Copenhagen International Airport.
Hong Kong, China Implementing a joint development agreement with a private entity for the new Chek
Lap Kok Airport on Lantau Island.
Hungary Implementing a joint development agreement with a private entity for a new international
terminal at Ferighey Airport in Budapest.
Indonesia Plans a joint development agreement with a private entity for a new airport in Medan.
Italy National Government owned airlines are divesting their shares in Rome and Milan
Airports.
Jamaica Plans a joint partnership at Sangster International Airport in Montego Bay.
Macau, China Implemented a joint development agreement with a private entity to develop and
manage a new international airport.
South Africa A private company was sold a 20 percent stake in the South African Airports company
to become a strategic equity partner to improve the company’s performance prior to
privatization.
Switzerland Sold 50 percent of shares in Zurich International Airport to private holdings. A private
firm operates the airport.
Thailand A strategic partner is being sought to participate in the development and operation of
a second international airport at Bangkok and in a regional airports company.

3. Management Contracts

Ownership and investment is maintained by the state and management and operation is
carried out through a private sector body. This may be in the form of a management contract, a
service concession, a multiple concession or a contracting out strategy.

Table A1.4: Examples of Management Contract Schemes Worldwide

Country Plans or actions for airport privatization

Albania Contracted with a private entity to modernize and expand Tirana Airport.
Chile Implemented a contract for a private entity to operate the passenger terminal at Arturo
Merino Benitez International Airport in Santiago.
Italy Plans to contract with a private entity to manage Naples Airport.
Hong Kong, China Private company awarded management contract of Kai Tak Airport.
 Appendix 2, page 1

SPECIFIC COUNTRY CASE STUDIES

A. Australia1

1. Policy Background

Following a policy review in 1993, the Australian Federal Government decided to implement
a radical airports privatization program. Key features of the privatization policy were as follows:

• In order to ensure that airports continued to be operated as airports, it was decided to

sell very long term leases (50 years with an option to renew for a further 49 years) rather
than to sell the airport freeholds.

• Airports were offered for sale individually, with cross-ownership restrictions imposed on
a limited number of airports, between which it was believed there was some scope for
competition.

• In order to encourage competition in downstream (airline) markets, airline ownership of

airports was restricted and, in fact, none of the airports offered for sale is airline controlled.
Airport operations have also been made subject to economy wide access provisions in
order to encourage new airline entrants.

• Airport charges at core airports were regulated via a price cap mechanism. However,
non-aeronautical charges will not be regulated.

2. The Privatization Process

The privatization of 17 of the 22 airports took place in two phases in 1997 and 1998.

a. Phase One

The first phase involved the sale of the major international gateway airports at Melbourne,
Brisbane and Perth, which was initiated in April 1996, and completed with the granting of licenses
on 1 July 1997. The tender process took place within a relatively short time period — expressions
of interest were invited by 10 October 1996. After short listing, nine consortia submitted offers by 30
January 1997. Six consortia were then asked to present revised bids by 10 April 1997. The three
successful consortia were announced on 7 May 1997. The three successful bids totaled A$3.31
billion, far in excess of the A$2.2 billion estimated at the time the sale was announced.

b. Phase Two

Following the success of the first sale, a further 15 airports were offered for sale
individually on 1 October 1997. Sydney airport, the country’s largest, was excluded from the
process due to problems caused by environmental restrictions, such as a government-imposed
traffic cap and a night-time curfew. The second phase sales were completed by 30 June 1998,
although no satisfactory bids were received for one airport, which was withdrawn from sale.

________________________________________________________

1
This section has benefited greatly from detailed comments by Mr Robin Renwick, Senior Director, Office of Asset Sales
and IT Outsourcing, Australia, a discussant at an Asian Development Bank (ADB) workshop.
 Appendix 2, page 2

Twenty six consortia were on the original shortlist. The Phase Two airports were divided into two
groups — 10 Regular Public Transport airports and five General Aviation airports.

As a result of the Phase One and Two sales program, the industry structure was radically
transformed, with ownership divided between ten private sector consortia.

3. Investment Structure

As part of their bid, each lessee company provided an airport masterplan, in which they
committed to include major development plans as well as satisfying various building requirements.
For Melbourne and Brisbane, these involved new runways, focusing on freight and using airport
land to build new hotels and business facilities.

4. Regulation

The main features of the economic regulation framework applied to the core privatized
airports are as follows:

• Application of airport specific price caps to aeronautical charges for an initial five-year
period. The value of X varied between airports according to an assessment by the
regulatory agency, the Australian Competition and Consumer Council (ACCC) of the
scope for productivity improvements at individual airports. There is also scope for
individual price caps to be varied at the initiative of the operator and subject to the
consent of the ACCC and users to accommodate increases in costs as major new
investments are brought on stream.

• A quinquennial review of the existing framework, when airport operators will be given
the opportunity to come forward with their own proposals for future regulation.

• Monitoring of service quality against pre-determined performance indicators (service

quality standards are not mandated).

• Monitoring of lessee investment programs against airport masterplan commitments.

• An emphasis on transparency, with financial reporting requirements imposed on lessees

in order to facilitate inter-airport comparisons of performance.

• Airport operations to be made subject to economy wide conditions on access to essential

facilities.
2
B. Colombia

1. Introduction

Following deregulation of the Colombian air transport market in 1990, domestic

passenger traffic had doubled and international traffic had increased 2.5 times by 1996. As a
result, the single runway at El Dorado International Airport in Santa Fe de Bogota, responsible
for 65 percent of Colombia’s total air traffic, reached its technical capacity in 1993. In 1993,
the Government of Colombia also separated airport operations from air navigation activities through
the corporatization of the Civil Aviation Authority, Aerocivil. At the same time, it undertook the

________________________________________________________

2
We acknowledge the assistance of Justin Anstee (Deutsche Bank, London) in preparing this case study.
 Appendix 2, page 3

development of a second runway, urgently needed to cope with an annual growth of eight percent
in international traffic and 12 percent in domestic traffic, utilizing a BOT scheme with an opening
date of September 1998.

2. The Process

a. The Concessionaire

The consortium Compania de Desarollo Aeropuerto El Dorado S.A. (CODAD), owned by

Dragados y Construcciones S.A. of Spain (66 percent), Ogden Corporation of the US (19 percent)
and Concreto Ingenieros Civiles S.A. of Colombia (15 percent), won the 20-year concession following
a competitive bidding process initiated by Unidad Administrativa Especial de Aeronautica Civil
(Aerocivil), the Colombian Government agency responsible for the regulation and operation of
airports. Once bidders had fulfilled the technical requirements, bids were evaluated on the basis of
the net present value of the minimum landing fee revenue that the bidder would require throughout
the concession period and the weighted average landing fee in US dollars.

b. Project Details

The project has a BOT structure, under the terms of which CODAD will build a second
runway and maintain both runways from the time that operation commences for the 20-year duration
of the concession contract. The project involved building

• a runway 3,800m long by 45m wide;

• a parallel taxiway 30m wide;

• three 90° exits and one high speed exit;

• taxiways to connect the old and new runways;

• installation of lighting, visual guidance equipment, radio-navigation aids and

meteorological sensors;

• re-routing of 2,600m of the Bogota river.

3. Contractual Structure and Remuneration

Following construction, CODAD will be responsible for any investment that is needed in the
maintenance of the new runway, and, in return, will have the right to commercial aircraft landing
fees generated at El Dorado airport. Following the completion of the second runway, the
concessionaire may immediately increase its charges by double the 1997 tariffs for the original
Bogota runway (following a 20-30 percent increase in tariffs between 1996-1997). The scheme
has a high guaranteed minimum return on its investment such that if the landing fee structure or
traffic volume (or both) cannot support the required revenue stream the Government would
compensate the concessionaire from a trust fund equivalent to 30 percent of the annual landing
fee revenue. The concessionaire will receive a 20 percent equity stake in the facility.
 Appendix 2, page 4

Figure A1.1: Contractual Structure

Colombian
Security Trust Colombian
Compania de Desarollo Aeropuerto Eldorado
Agreement Trustee

Indenture Indenture
Noteholders
Trustee

Concession
Contract

Aerocivil
Minimum
Revenue
Guaranty Fund
Agreement
Dragados
Contractor
Construction Note
Contractor
Contract Purchase
Agreement
Conconcreto

Using private sector capital to relieve the capacity constraint at El Dorado airport is
expected to allow Aerocivil to focus investment on increasing the level of airside operations
and improve the safety of flight operations.

4. Financial Structure

Access to private capital has been provided through a US$116 million CODAD offering of
senior secured notes with a BBB/BBB rating and a due date of 2011 (15 year final maturity).
The initial financing offered a spread of 340 bps over a 10-year treasury bond with a coupon of
10.1 percent, while in secondary trading this tightened to 330 bps. The proceeds from this
issuance and sale of notes to the consortium will be used to fund the development, construction
and financing costs of the new runway estimated at US$100.3 million. The offering was
successful, being substantially oversubscribed and attracting a number of first time project
3
finance buyers.

The CODAD financing structure set a number of benchmarks for Colombia as it

represented the first investment grade rated cross-border transportation project, the longest
maturity cross-border transportation project and the first example of non-energy project financing
in the country.

________________________________________________________

3
Deutsche Bank, 1997
 Appendix 2, page 5

5. Regulatory Structure

Aerocivil acts as the regulatory body. Aerocivil is not an independent regulator, such as the
UK Civil Aviation Authority, but is a part of government, which allows the concessionaire more
freedom in their charging behavior. The initial generous ceiling on landing charges has been outlined
above.

C. Philippines

1. Introduction

Under the provisions of the Philippine BOT law, as amended in 1994, unsolicited bids can
be made to carry out financially viable public sector projects on a BOT basis. Proposals were put
forward in 1996 by the Asian Dragons Consortium (ADC), a group of Filipino businesses, to construct
and operate a third passenger terminal at Ninoy Aquino International Airport (NAIA) to handle
international traffic. The airport, a two-runway facility, currently handles around 12.3 million passengers
per annum (mppa) (7.8 mppa international, 4.6 mppa domestic) and is owned and operated by
Manila International Airport Authority (MIAA), a state-owned corporation created following the
separation of airport operations from the Philippine Air Transport Organization, an agency of the
Department of Transportation and Communications (DOTC). The Philippine Air Transport
Organization had previously owned and operated airport and air traffic control (ATC) assets
throughout the Philippines.

2. The BOT Project

The ADC proposal envisaged the construction and operation of the new international terminal
alongside the MIAA’s existing airport operations. The BOT entity would be entitled to receive passenger
charges, as agreed between MIAA and DOTC for passengers using the terminal. These charges
would increase in line with domestic inflation. It would also receive revenues from commercial
activities (retailing, car parking, office rentals) carried on in the terminal for a period of 25 years
from the commencement of the concession.

Use of the existing international terminal would be discontinued and all international traffic
will be transferred to the new terminal upon its completion. The concession agreement would also
commit the Government not to authorize any development of new facilities for international traffic in
the region (for example, at Clark Airbase), until international traffic at NAIA had exceeded 10 mppa
for three consecutive years. On current projections, made before the onset of the economic and
financial crisis, this would not occur until around 2005-2010.

The concessionaire would pay MIAA a two part (fixed plus variable) concession fee; the
variable element would be expressed as a proportion of total revenue generated by the terminal.
The concession contract committed the concessionaire to maintain certain service quality standards
and to provide specified peak passenger handling capacity through the terminal.

3. A Competitive Bid
4
Under Section 4-A of the BOT law, the government is obliged to seek competitive
proposals for unsolicited projects, once financial terms have been negotiated with the bidder.
In this case, a competitive bid was received from a consortium including Lufthansa as well as

________________________________________________________

4
See Appendix 3.
 Appendix 2, page 6

indigenous Filipino organizations. The competitive bid offered significantly better financial terms
than the ADC proposal, with a higher fixed fee and a larger proportion of revenue. Had ADC felt
able to match these terms, the project would have been awarded to them; however, they did not,
and the competitive bid was accepted.

ADC subsequently challenged the award of the contract in the Philippine courts.

D. Thailand5

1. Introduction

The major international gateway airport in Thailand (Don Muang) has experienced very
rapid traffic growth (11 percent per annum in the period 1990-95), and is likely to become increasingly
congested. The Thai Government has plans to construct a second international airport for Bangkok
at Nung Ngu Mao (the so-called Cobra Swamp). Proposals for the new airport, published in 1996,
envisaged that the first stage of the new airport would open around 2000. In September 1996, the
Japanese government agreed to provide a 25-year soft loan to Thailand, including US$280 million
towards the construction of the new airport.

The five major airports in Thailand - Bangkok, Ohiangmai, Ohiengrai, Phaket and Hadyai
— are currently owned and operated by the Airports Authority of Thailand (AAT). Twenty five smaller
airports are owned and operated by the Thai Department of Aviation. AAT was responsible for
developing plans for the new airport. However, the Thai Government has been examining options
for attracting private financing to the airports sector. To facilitate this, and to minimize the risk of
more bureaucratic delays to the project, it has created a new state corporation, the New Bangkok
International Airport Corporation (NBIAC), jointly owned by the AAT and the Finance Ministry, to be
responsible for the new airport project, if possible with some private sector funding. A Government
committee examining privatization options has also recommended that the four principal regional
airports at Phuket, Chiangmai, Ohiengrai and Hadyai, which are perceived to be potentially attractive
to private investors, as a single package.

2. Restructuring and Strategic Partnership Proposals

Options for restructuring AAT and privatization are constrained by conditions attached to
the Overseas Economic Cooperation Fund loan, a key element of the financing package for
the new airport. The loan is a government to government facility and can only be made available
to a state-controlled entity within Thailand. To satisfy this constraint, whilst at the same time
creating a role for private sector participation (PSP), the Thai Government’s financial advisers
have proposed a complex re-structuring package, involving two strategic partnerships with
the private sector, as shown in Figure A1.2.

________________________________________________________

5
The case study of Thailand has benefited from the constructive comments of Sirote Duangratana, of the New Bangkok International
Airport Corporation, a discussant at ADB workshop.
 Appendix 2, page 7

Figure A1.2: Thai Airport Restructuring Proposals

AATSE

70%
▼
30%
Ministry of Finance AAT Co., Ltd Strategic Partner (1)

▼
8%

▼ ▼ 92% ▼ 100% 25% ▼

75%
BIA Co. Regional Co. Strategic Partner (2)

▼
NBIA Co.

At the heart of the restructuring, is the creation of two joint venture companies, Airport
Authority of Thailand Co. Ltd. (AATCo), which would be 70 percent owned by a state holding company,
Airport Authority of Thailand State Enterprise (AATSE), and 30 percent by a private sector strategic
partner, and the Regional Airports Co. Ltd, which would be 25 percent owned by AATCo, and 75
percent owned by a strategic partner, which might be the same organization as the strategic partner
in AAT. AAT would wholly own BIA, the operator of the existing Bangkok International Airport, and
would own 92 percent of NBIAC, with a small residual holding being retained by the Finance Ministry.

This structure thus allows the injection of private finance through both the minority AAT
holding and through the majority stake in the regional airports company.

It is recognized that although this structure offers a private partner effective control over the
regional airports entity, the minority holding in AAT may be unattractive. Also, whilst operation of the
regional airports might be relatively profitable, initial returns on the investment in AAT may be modest,
given the need for major investment in the new airport. To remedy this, the strategic partner may be
offered a management contract and exclusive rights to commercial revenues in some of the more
lucrative lines of business at the two Bangkok airports.

The Thai Government is promoting a corporatization Bill, to facilitate the establishment of

the new AATCo entity and to enable AATCo to benefit from soft loan funding. The Bill is expected to
become law by the end of 1998. It is now expected that the selection of a strategic partner and full
implementation of the proposed new structure will be completed by early 2000.

3. Regulation of Charges

Increases in airport charges sought by AAT are subject to Government approval. No definite
proposals have so far been made regarding future arrangements for determining airport charges
under the new ownership structure, either at the Bangkok airports or at the regional airports.

E. Argentina

1. Introduction

The concessionaire for the large scale privatization of 33 of Argentina’s airports in a

single package was chosen in February 1998. Previously an arm of the Argentine Air Force
owned, administered, operated and regulated the Argentine airport infrastructure, in a relatively
unique combination which would have influenced the decision to attempt a large scale
 Appendix 2, page 8

privatization scheme. However, because of opposition from some shareholding provincial

governments, only 33 of the originally intended 59 airports in Argentina were involved in the
1997/98 process, although this proportion still gave the private operator a monopoly in airport
service.

2. The Process

a. The Concessionaire

In its invitations for bidding, the Government stipulated that the operator must have a
minimum 10 percent share within a bidding consortium, recorded passenger flows of 10 million,
and relevant construction experience of US$150 million over the past five years. The Government’s
insistence that the airport operator had to hold equity left many potential bidders out of the process,
although four major consortia prequalified.

In February 1998, the consortium Aeropuertas Argentina 2000 (AA2000) won the 30-year
contract to manage 33 of Argentina’s state-owned airports, with an option for 10 additional years.
Argentina 2000 is 30 percent owned by Milan airport operator (SEA), 28 percent owned by Ogden,
a US ground-handling company, and 35 percent owned by a local company, Corporacion America
Sudamericana. In May 1998 the consortium commenced management of Ezeiza International
Airport, the largest Argentine international airport. The two other main airports, Aeroparque de Buenos
Aires and Paja Blancas in Cordoba, were to be transferred to the consortium within 90 days of the
agreement, whilst the other 30 transfer within a year. The consortium will employ and supervise
airport personnel and subcontractors though it will not be responsible for the maintenance and
operation of ATO or landing systems and security, which will be handled by the Air force and the
government.

There are large variations in traffic volume between different airports with Ezeiza and
Aeroparque in Buenos Aires capturing two-thirds of total terminal volume, with annual turnover of
5.5 million and 6 million people respectively. Estimates that only a maximum of eight airports are
profitable indicate the need for cross subsidies to improve and develop inadequate existing facilities
at many airports. As a result, a sole-party concession was deemed necessary but has also led to
fears of higher charges at larger airports.

3. Investment Structure

The consortium offered the Argentine Government a US$2.2 billion capital investment
program over the 30 year concession period in addition to a US$171 million annual license
fee. No additional funding will be provided by the government for these investments. The bulk
of this total investment (US$1,346 million) will go to Buenos Aires airports. Approximately
US$859 million will be invested in the first five years of the concession (1998-2002). A large
amount of which would be invested in a proposed expansion of Ezeiza airport, which will be
merged with the operations of Buenos Aires’ domestic airport, Aeroparque, to create a single
domestic and international airport at the current location of Ezeiza by 2001. Following this
initial five year period there is a declining financial commitment for the remainder of the 30-
year concession, from a peak of US$180 million total throughout the system in 2000 to
approximately US$10 million in 2027.

An environmental study for each airport, identifying existing problems has provided the
operator with a list of remedial work to be carried out, the cost of which will be deducted from
 Appendix 2, page 9

their license payments due to the Government. Operator accountability will begin when problems
arise beyond this original study.

4. Remuneration

The consortium will collect and retain all airport revenues from air and land side activities,
except where existing contracts have not expired. Airport land may also be utilized for its commercial
value, making low passenger volume airports located in Industrial Free Zones more attractive. New
tariff levels will be announced to be effective from 22 June when the concessionaire takes over the
airports. It is expected that landing charges will stay constant and passenger charges will increase
by up to 50 percent, alongside increases in other charges, such as real estate rents. By year six
the consortium anticipates making an 18 percent return on capital, and profits are anticipated to
average seven percent of revenue. Having been given a free hand to exploit commercial opportunities,
the consortium expects revenues to approach US$20 billion over the life of the concession. The
US$2.2 billion for capital improvements is expected to be funded from system cash flows and non-
recourse project finance.

Duty free, ground handling and warehousing services are not included in the concession
where pre-privatization contracts will not expire until 2010, unless the operator and current franchisees
agree on other arrangements. The concessionaires will get some revenues from these contracts
up until that point, though these amounts are expected to be lower than international standards,
and may lead to an increase in charges on airlines or passengers.

5. Regulation

Post privatization, roles will be divided between the existing airport manager (ORA) and the
concessionaire. A regulatory body under the Ministry of Economy, the Organismo Regulador del
Sistema Nacional de Aeropuertas, will exist to resolve conflicts between the two parties, and if
conflicts involve the regulatory body itself, the Argentine judicial system will intervene.

F. Côte d’Ivoire

1. Background

Until July 1996 all airports in Côte d’lvoire and ATO services at airports other than the
major international airport at Abidjan6 were owned and operated by Agence Nationale de l’Aviation
civile et de la Meteorologie (National Agency for Civil Aviation and Meteorology (ANAM)), which
also acted as a national regulatory agency for air transport. Under legislation passed in 1996,
however, responsibility for the operation and development of Abidjan airport was transferred
through a 15-year concession agreement to AERIA, a special purpose company controlled
by Société d’Exploitation et de Gestion Aéroportuaire (SEGAP), a jointly owned subsidiary
of the French based Service Group and the Marseilles Chamber of Commerce and Industry
(MCCI). MCCI operates Marseilles airport, and SEGAP also operates Libreville (Gabon)
airport under a concession agreement signed in 1988. AERIA took over responsibility for
st
operating Abidjan airport on 1 July 1996. The concession agreement contains provisions for the
Ivorian state to hold up to 20 percent of AERIA’s share capital.

________________________________________________________

6
ATC services at Abidjan are owned and operated by ASECNA (Agence pour Ia Securité de Ia Navigation aeriénne en Afrique et
a Madagascar), a multinational agency jointly owned by a group of fourteen Francophone African States, charged with providing
en-route air navigation services on behalf of the member states.
 Appendix 2, page 10

2. The Award of the Concession

Following a decision in 1994 to seek PSP in upgrading and extending the facilities at Abidjan
airport, the Côte d’Ivoire authorities issued a call for expressions of interest containing outline terms
of reference for the project. Respondents were invited to put forward proposals covering:

• The form of PSP, and the relationship between the private sector party and the state.

• An investment program to upgrade and extend the airport infrastructure over the period
1996 - 2000 to the extent necessary to consolidate Abidjan’s position as the major hub
airport in Francophone, West Africa. The authorities also anticipated that the
concessionaire would undertake significant commercial development on the airport and
in areas adjacent to the airport.

• Indicative financial terms.

In response, the authority received outline proposals based on the concessions model of
PSP from SEGAP and from a consortium led by Aéroports de Paris, and entered negotiations with
both parties in 1995.

SEGAP was identified as the preferred bidder primarily on the basis of the quality and depth
of its proposals, in particular, concerning the investment program during the early years of the
concession period. Negotiations were concluded early in 1996, and a special purpose company,
AERIA was established as concessionaire.

3. Principal Features of the Concession Agreement

Under the terms of the concession agreement, AERIA gains title to aeronautical7 and other
commercial revenues (such as rents and car parking fees) arising from airport operations at Abidjan
which previously accrued to ANAM. From these revenues, AERIA is committed to operating and
maintaining airport assets over the 15 year concession and to undertaking and financing a specified
program of investments covering the first four years of the concession period. AERIA also commits
to pay around 20 percent of total turnover (net of VAT) in concession fees to the concessioning
authority. Approximately a half of the fee is described as a usage charge. The remainder is identified
as a subsidy to cover deficits at interior airports.

AEIRA must also set aeronautical tariffs on the basis of principles established by the
International Civil Aviation Organization (ICAO). Any proposed adjustments to aeronautical tariffs
(though not commercial charges) must be approved by the concessioning authority. In assessing
AERIA’s proposals for tariff adjustments, the concession agreement requires the concessioning
authority to ensure the “financial equilibrium” of the concessionaire. This is described simply in
terms of revenues from aeronautical charges and other sources covering the concessionaire’s
costs, including the concession fee paid to the government.

As an aspect of financial equilibrium, the concession agreement recognizes AERIA’s

right to recover the full costs of the contractually committed investments over the life of the
concession. However, the agreement also recognizes, and attempts to address, the issue of

________________________________________________________

7
Aeronautical charges are levied on airlines and include landing charges (based on aircraft weight), aircraft parking charges (also
aircraft weight based) and passenger handling and freight handling fees.
 Appendix 2, page 11

unanticipated investment undertaken later in the concession which would be “stranded” if the
concession was terminated after 15 years. It does so by providing for the concessionaire to receive
compensation from the conceding authority in respect of unamortised investments on the basis of
a valuation undertaken by an independent expert or experts. Although not explicitly stated, it is
expected that in the event of a failure to agree on such a valuation, the issue would be covered by
the general arbitration provisions in the concession agreement. The agreement also provides for
the modification of detailed aspects of the investment programs in the light of technical etc., changes
and the emerging needs of the airport business.

In order to enable the concessioning authority or its agent to carry out its responsibilities for
technical and financial oversight of the concessionaire, the agreement requires the concessionaire
to provide to the concessioning authority annual budgets and financial accounting data, and to
report on the progress of the agreed investment program, and on the state of inherited assets.

The concession agreement provides for the termination or suspension of the concession
in the event of serious failure to perform, for example, in respect of severe interruption to airport
services other than for reasons beyond the control of the concessionaire. Apart from relatively
extreme circumstances of this kind, there are no specific terms in the agreement covering AERIA’s
obligations in respect of service quality, nor are there provisions requiring the concessionaire to
provide specific types of information on service quality outcomes.

Finally, the concession agreement allows the concessionaire to propose an extension to

the concession period. However, any such proposal can only be considered within the last two
years of the concession, and renewal can only be formally approved by the concessioning authority
in the final year of the concession.

A 15-year concession term is significantly shorter than the norm for concession or BOT
contracts elsewhere in the airport sector; SEGAP’s response to the call for expressions of interest
suggested a 25-year concession period, in line with international experience.

4. Regulatory Agencies

Following the winding-up of ANAM, that agency’s regulatory and policy development
responsibilities for the aviation sector were transferred to the Agence Nationale pour l’Aviation
Civile (National Civil Aviation Agency (ANAC)). ANAC also retained certain of ANAM’s operational
responsibilities for providing airport terminal security (passenger and baggage screening) at Abidjan
H-B.

ANAC’s responsibilities have recently been reviewed by the Ivorian Government, partly to
achieve an even clearer separation of regulatory and operational responsibilities between ANAC
and le Service Méteorologique National de la Côte d’lvoire (National Meteorological Service of the
Ivory Coast (SODEXAM)).

Proposals for modifying the existing arrangements were made in a communication from a
government working group in October 1998, and these were subsequently ratified by the Council
of Ministers. The measures cover the re-definition of the responsibilities of ANAC, SODEXAM
and AERIA, and the creation of a new government committee to co-ordinate policy for the airport
sector.
 Appendix 2, page 12

ANAC’s role as the regulatory agency for the aviation sector is confirmed. However,
operational responsibility for airport security at Abidjan H-B is to be transferred to AERIA, under a
new concession agreement, with ANAC acting as the concessioning authority. The new structure
will presumably require provisions for sharing security charges levied at Abidjan between SODEXAM
and AERIA, although this is not specifically addressed in the government’s proposals. ANAC does
not appear to have any locus in “regulating” AERIA’s concession agreement. This function remains
with SODEXAM as the conceding authority.

SODEXAM’s existing operational responsibilities are confirmed, and it is identified as the

conceding authority for all airport infrastructure (including Abidjan). As such, SODEXAM (not ANAC)
must approve AERIA’s proposals for varying airport tariffs and SODEXAM also “vets” AERIA’s
investment program. There is no obligation on SODEXAM to publish its decisions on AERIAs tariffs
or other aspects of the concession contract. The government working group’s communication
expressed the hope that SODEXAM’s responsibilities for operating interior airports might eventually
be transferred to private sector agencies within a concession framework.

The inter-ministerial committee is given wide-ranging responsibilities for initiating and

coordinating policies affecting the development of the airports sector. In the short-term, the primary
focus of its work will be to implement the new arrangements for airport security and to review
current provisions for financing and staffing ANAC and SODEXAM.
 Appendix 3, page 1

EXTRACTS FROM THE PHILIPPINE REPUBLIC BOT LAW

The following extracts from the Philippine Republic Build-Operate-Transfer (BOT) law cover
key aspects of public procurement in respect of BOT projects.

A. Section 4-A: Unsolicited Proposals

Unsolicited proposals for projects may be accepted by any government agency or local
government unit on a negotiated basis: Provided, That, all of the following conditions are met:
(1) such projects involve a new concept or technology and/or are not part of the list of priority
projects, (2) no direct government guarantee, subsidy or equity is required, and (3) the
government agency or local government unit has invited by publication, for three weeks, in a
newspaper of general circulation, comparative or competitive proposals and no other proposal
is received for a period of sixty (60) working days: Provided, further, that in the event another
proponent submits a lower price proposal, the original proponent shall have the right to match
that price within thirty (30) working days.

B. Section 5: Public Bidding of Projects

Upon approval of the projects mentioned in Section 4 of this Act, the head of the infrastructure
agency or local government unit concerned shall forthwith cause to be published, once every week
for three consecutive weeks, in at least two newspapers of general circulation and in at least one
local newspaper which is circulated in the region, province, city or municipality in which the project
is to be constructed, a notice inviting all prospective infrastructure or development project proponents
to participate in a competitive public bidding for the projects so approved.

In the case of a build-operate-and-transfer arrangement, the contract shall be awarded to

the bidder who, having satisfied the minimum financial, technical, organizational and legal standards
required by this Act, has submitted the lowest bid and most favorable terms for the project, based
on the present value of its proposed tolls, fees, rentals and charges over a fixed term for the facility
to be constructed, rehabilitated, operated and maintained according to the prescribed minimum
design and performance standards, plans and specifications. For this purpose, the winning project
proponent shall be automatically granted by the appropriate agency the franchise to operate and
maintain the facility, including the collection of tolls, fees, rentals, and charges in accordance with
Section 5 hereof.

In the case of a build-and-transfer or build-lease-and-transfer arrangement, the contract

shall be awarded to the lowest complying bidder based on the present value of its proposed schedule
of amortization payments for the facility to be constructed according to the prescribed minimum
design and performance standards, plans and specifications: Provided, however, that a Filipino
contractor who submits an equally advantageous bid with exactly the same price and technical
specifications as those of a foreign contractor shall be given preference.

In all cases, a consortium that participates in a bid must present proof that the members of
the consortium have bound themselves jointly and severally to assume responsibility for any project.
The withdrawal of any member of the consortium prior to the implementation of the project could be
a ground for the cancellation of the contract.

The public bidding must be conducted under a two-envelope/two-stage system: the first
envelope to contain the technical proposal and the second envelope to contain the financial
 Appendix 3, page 2

proposal. The procedures for this system shall be outlined in the implementing rules and
regulations of this Act.

A copy of each contract involving a project entered into under this Act shall forthwith be
submitted to Congress for its information.

C. Section 5-A: Direct Negotiation of Contracts

Direct negotiation shall be resorted to when there is only one complying bidder left as defined
hereunder:

• If, after advertisement, more than one contractor applied for prequalification and it meets
the prequalification requirements, after which it is required to submit a bid/proposal
which is subsequently found by the agency/local government unit (LGU) to be complying.

• If, after advertisement, more than one contractor applied for prequalification but only
one meets the prequalification requirements, after which it submits a bid/proposal which
is found by the agency/LGU to be complying.

• If, after prequalification of more than one contractor, only one submits a bid which is
found by the agency/LGU to be complying.

• If, after prequalification, more than once contractor submit bids but only one is found by
the agency/LGU to be complying: Provided, That, any of the disqualification Bids
and Awards Committee within 15 working days to the head of the agency, in case
of national projects or to the Department of the Interior and Local Government, in
case of local projects from the date of the disqualification was made known to the
disqualified bidder. Provided, furthermore, That the implementing agency/LGUs
concerned should act on the appeal within 45 working days from receipt thereof.