By Frédéric Blanc-Brude and Majid HasanEDHEC-Risk Institute

A guide to the valuation of privately held infrastructure equity and debt

INFRASTRUCTURE VALUATION

Published in March 2015 byPEI6th Floor140 London WallLondon EC2Y 5DNUnited Kingdom

Telephone: +44 (0)20 7566 5444www.privateequityinternational.com/bookstore

© 2015 PEI

ISBN 978-1-908783-77-6

This publication is not included in the CLA Licence so you must not copy any portion ofit without the permission of the publisher.

All rights reserved. No parts of this publication may be reproduced, stored in a retrievalsystem or transmitted in any form or by any means including electronic, mechanical, pho-tocopy, recording or otherwise, without written permission of the publisher.

Disclaimer: This publication contains general information only and the contributors arenot, by means of this publication, rendering accounting, business, financial, investment,legal, tax, or other professional advice or services. This publication is not a substitute forsuch professional advice or services, nor should it be used as a basis for any decision oraction that may affect your business. Before making any decision or taking any action thatmay affect your business, you should consult a qualified professional adviser. Neither thecontributors, their firms, its affiliates, nor related entities shall be responsible for any losssustained by any person who relies on this publication.

The views and opinions expressed in the book are solely those of the authors and neednot reflect those of their employing institutions.

Although every reasonable effort has been made to ensure the accuracy of this publica-tion, the publisher accepts no responsibility for any errors or omissions within this pub-lication or for any expense or other loss alleged to have arisen in any way in connectionwith a reader’s use of this publication.

PEI editors: Wanching Leong and Helen LewerProduction editor: Julie Foster

Printed in the UK by: Hobbs the Printers (www.hobbs.uk.com)

Contents

Figures and tables vii

About the authors xi

1 Introduction 11.1 Why infrastructure? 11.2 The demand for benchmarks 21.3 A roadmap 31.4 This book 6

PART I: UNDERSTANDING PRIVATELY HELD INFRASTRUCTURE INVESTMENTS

2 The nature of investable infrastructure 112.1 Existence of investable infrastructure 112.2 The role of long-term contracts 112.3 The role of commitment 132.4 The contractual nature of underlying infrastructure investments 132.5 Creating and regulating investable infrastructure 152.6 Monopoly privatisation and regulation 162.7 Project finance and corporate governance 182.8 Conclusion: Infrastructure assets are not real assets 19

3 Infrastructure project financing 213.1 Investment delegation and project financing 213.2 Definition 223.3 Agency and monitoring in project finance 273.4 A response to the long-term investment problem 323.5 Project financing as risk optimisation 333.6 Conclusion: Infrastructure project finance as a reference asset 35

4 Risk and the cost of capital in infrastructure projects 374.1 Construction risk 384.2 Leverage 404.3 Revenue risk 424.4 Contract renegotiation 434.5 Political risk 454.6 Conclusion: Systematic sources of risk in project finance 47

iii

PART II: ISSUES WITH LONG-TERM INVESTMENT PERFORMANCE MEASURES

5 Benchmarking infrastructure investments: Objectives and challenges 51

5.1 Why benchmark infrastructure investments? 515.2 Two challenges 58

6 Addressing the challenges of long-term investment benchmarking 63

6.1 Bayesian inference for long-term investment benchmarking 636.2 Towards intersubjective pricing measures 646.3 Families of generic infrastructure projects 65

PART III: MEASURING THE PERFORMANCE OF INFRASTRUCTURE PROJECT DEBT

7 Objectives and characteristics of infrastructure project debt 737.1 Objectives 747.2 Defining infrastructure debt 747.3 Observable asset values 757.4 Covenants and embedded options 777.5 Identifying default triggers 787.6 Reorganisations 807.7 Illiquidity and lumpiness 807.8 Documented performance 817.9 Conclusion 82

8 Approaching the valuation of infrastructure debt 858.1 Asset-pricing models 868.2 Corporate debt valuation models 888.3 Proposed approach 918.4 Conclusion 93

9 Valuation framework 959.1 Intuition 959.2 Cash flow dynamics 969.3 Structural framework 97

10 Debt restructuring model 10310.1 Restructuring upon a technical default 10310.2 Restructuring upon a hard default 104

11 Model implementation 11111.1 Cash flow model 11111.2 Risk-neutral measure 11311.3 Debt rescheduling upon default 117

iv

Contents

11.4 Algorithm 11911.5 Results 122

PART IV: VALUING EQUITY INVESTMENTS IN INFRASTRUCTURE PROJECTS

12 Objectives and approaches 13512.1 Objectives 13512.2 Existing approaches to private equity valuation 13512.3 Proposed approach 139

13 Expected dividend model 14313.1 A Bayesian framework 14313.2 Cash flow metrics and risk measures 14413.3 Dividend state transitions 14613.4 (Conditional) dividend payment distribution 15413.5 Prior elicitation 156

14 Valuation framework 16114.1 Modern asset-pricing theory 16214.2 A state-space framework 165

15 Observation (pricing) equation 16915.1 Pricing equation 16915.2 Matrix formulation 17115.3 Observation equation 172

16 State (discounting) equation 17516.1 Term structure model 17516.2 Matrix formulation 17816.3 State equation 179

17 Filtering of the term structure 181State-space model 18117.1 Filtering and updating 18217.2 Parameter estimation 18417.3 Model outputs 18617.4 Conclusion 186

18 Model implementation 18918.1 Prior elicitation 18918.2 Equity cash flow projections 19218.3 Results 19718.4 Return measures 20218.5 Risk measures 20218.6 Market dynamics 20418.7 Sensitivity to conditional dividend volatility estimates 205

Contents

v

Contents

PART V: CONCLUSIONS

19 Conclusions 20919.1 Key points 20919.2 Data collection and standardisation 21119.3 Next steps 215

PART VI: TECHNICAL APPENDICES

A Debt valuation 219A.1 Risk-neutral measure 219A.2 Risk measures 224A.3 Return measures 226

B Equity valuation 227B.1 Cash-flow metrics 227B.2 Bayesian estimates 229B.3 Pricing equation 234

Bibliography 243

About PEI 256

vi

Figures and tables

Figure 3.1: Simplified project contractual claims structure 24

Figure 3.2: Example of base case cash flows and debt service cover ratio for a 35-year old project company with contracted revenues 26

Figure 4.1: Difference between ex ante and ex post construction costs in traditional infrastructure procurement and project finance 39

Figure 6.1: Box plots of the average DSCR in projects with contracted, partly contracted and fully merchant revenues 68

Figure 6.2: Box plots of the tail size in projects with contracted, partly contracted and fully merchant revenues 68

Figure 7.1: Regional shares of the cumulative project finance deal flow and project bond issuance,1994–2013 75

Figure 7.2: Marginal probability of default in a sample of project finance loans 82

Figure 9.1: Black-Cox decomposition 101

Figure 10.1: Renegotiation and exit values immediately following a hard default 107

Figure 10.2: Outcome of renegotiation post hard default 109

Figure 11.1: DSCR models for merchant and contracted infrastructure 112

Figure 11.2: Risk-neutral distance to default for rising and flat DSCR 116

Figure 11.3: Risk-neutralised DSCR distributions for merchant and contractedinfrastructure projects 116

Figure 11.4: Debt rescheduling upon a technical default 118

Figure 11.5: Debt rescheduling upon a hard default 119

Figure 11.6: Flow chart of the determination of cash flows to debt holders 120

Figure 11.7: Comparison of mean debt payments and CFADS 123

Figures

vii

Figure 11.8: Comparison of the probability of technical or hard default, hard default only and probability of death for rising and flat DSCR 123

Figure 11.9: Comparison of EL, VaR and cVar for rising and flat DSCR 124

Figure 11.10: Exit value of lenders and total firm and debt values 126

Figure 11.11: Probability of lender and equity haircuts (of any size) 127

Figure 11.12: Loss given default as a percentage of the value of debt 128

Figure 11.13: Loss given default in absolute terms for a $1,000 investment 128

Figure 11.14: Recovery rates 129

Figure 11.15: Time evolution of duration 129

Figure 11.16: Trade-off between credit and interest rate risk 130

Figure 11.17: Comparison of yield and z-spread for rising and flat DSCR 131

Figure 11.18: Range of yields for rising and flat DSCR 132

Figure 13.1: Prior and posterior densities of the beta distribution of π11over 12 iterations using 50 data points 152

Figure 13.2: Prior and posterior densities of the beta distribution of π01over 12 iterations using 50 data points 153

Figure 13.3: Prior and posterior estimates of the probability of observing a positive dividend at time t 154

Figure 13.4: Prior and posterior densities of the log-normal distribution of ESCRt over 12 iterations using 50 data points 157

Figure 13.5: Prior and posterior densities of the log-normal distribution of ESCRt over 12 iterations using 50 data points 158

Figure 13.6: Prior and posterior densities of the m and p parameters of the log-normal distribution of ESCRt 159

Figure 15.1: Investment and cash flow timeline 170

Figure 17.1: State-space predicting and updating procedure with Kalman filtering 184

Figures and tables

viii

Figure 18.1: How prior knowledge of infrastructure project finance can be combined with observable data to derive posterior parameters of the distribution of dividends 190

Figure 18.2: Assumed DSCR dynamics for merchant and contracted infrastructure 193

Figure 18.3: Base case and expected free cash flow (CFADS) and dividend payments 194

Figure 18.4: Expected dividends as a multiple of the first non-zero dividend 194

Figure 18.5: Probability of default, lock-up and bankruptcy 195

Figure 18.6: Probability of non-zero dividend payment 195

Figure 18.7: Mean and volatility of ESCRt 196

Figure 18.8: Unconditional density of ESCRt at different times in the project’s life cycle 196

Figure 18.9: Range of forward period risk premia of expected cash flows at t = 0 198

Figure 18.10: Range of term structures of discount rates of expected cash flows at t = 0 198

Figure 18.11: Range of stochastic discount factors of expected cash flows at t = 0 199

Figure 18.12: Range of prices during project’s life cycle at t = 0 199

Figure 18.13: Forward curves of period equity risk premiums over the project’s life 201

Figure 18.14: Term structure of discount rates over the project’s life 202

Figure 18.15: Cash yield or dividend/price ratio in each period, conditional on the base case being realised in the previous periods 203

Figure 18.16: Evolution of equity IRR across the project life cycle 203

Figure 18.17: Expected and extreme loss measures at t0 204

Figure 18.18: Evolution of effective duration 204

Figure 18.19: Filtered estimates of the IRR and transaction prices 205

Figure 18.20: Filtered prices and term structure with varying expectations of dividend volatility 206

Figure 19.1: Building blocks of infrastructure project finance debt cash flow dynamics 213

Figures and tables

ix

Table 4.1: Differences between ex ante and ex post construction costs in traditional infrastructure procurement and project finance 40

Table 10.1: Descriptions and symbols for the variables used in the renegotiation model 106

Table 11.1: Merchant and contracted infrastructure characteristics 112

Table 11.2: DSCR models for merchant and contracted infrastructure 112

Table 13.1: True, prior and posterior values of example transition probabilities at time t 151

Table 13.2: True, prior and posterior values of example dividend distributionparameters at time t 151

Table 18.1: Merchant and contracted infrastructure project financing characteristics 193

Table 18.2: DSCR models for merchant and contracted infrastructure 193

Table 19.1: Data collection requirements 214

Table B.1: Posterior mean and variance of example transition probabilities at time t in each observation round 231

Table B.2: Posterior mean and variance of example ESCR distribution parameters (log-scale) at time t in each observation round 233

Table B.3: Comparison of filtered prices for different choices of initial values of and ø 241(γt+i)Ti=1

Tables

Figures and tables

x

About the authors

xi

Frédéric Blanc-Brude is research director at EDHEC-Risk Institute and heads its the-matic research programme on infrastructure investment. Prior to joining EDHEC, heworked for ten years in the infrastructure finance sector and was actively involved intransactions representing a cumulative value of more than $6 billion in Europe, Asia andthe Middle East. Frédéric also teaches the infrastructure investment seminar of the YaleSchool of Management, EDHEC-Risk Institute Certificate in Risk and InvestmentManagement. Frédéric holds a PhD in Finance from King’s College London.

Email: frederic.blanc-brude@edhec.edu

Majid Hasan is a research assistant at EDHEC-Risk Institute and a PhD in Finance can-didate at EDHEC Business School. His research interests include asset pricing in imper-fect markets and designing solutions to channel private capital towards publicinfrastructure. Majid holds a Masters in Applied Mathematics from Western University,Canada, and a Bachelors in Physics from the University of Punjab, Pakistan. Before join-ing EDHEC, Majid worked as a research and teaching assistant at Western University,and the University of BritishColumbia, Canada.

The authors would like to thank the following sponsors for their support of thisresearch: Meridiam Infrastructure, Campbell Lutyens, Natixis and the MonetaryAuthority of Singapore. n

Introduction

Matching the huge demand for capital in infrastructure projects around the world withthe available supply of long-term funds by institutional investors – whether pensionfunds, insurers or sovereign wealth funds – has never been so high on the internationalpolicy agenda.

This momentum, illustrated by the G20’s focus on long-term investment in infrastructure,coincides with the steadily growing investment appetite from investors for unlisted andilliquid assets. However, full-fledged investment solutions demonstrating the benefits oflong infrastructure investment for institutional investors remain elusive, and documentingthe investment characteristics of long-term investment in infrastructure has become apressing issue.

Long-term investment in infrastructure is part of a broader trend among institutionalinvestors to improve portfolio diversification or seek higher returns through alternativeinvestments, to invest increasingly outside of public capital markets, to find sufficientlylong-dated instruments with a more attractive performance than government bonds, andto invest in inflation-linked securities other than low-yielding Treasury Inflation-ProtectedSecurities (TIPS). One of the salient features of these emerging investment choices is thedecision to buy assets that are infrequently traded and hold them until maturity.

The growing interest of investors for infrastructure investment is motivated by what wecall the ‘infrastructure investment narrative’ (see Blanc-Brude, 2013c). This is the notionthat infrastructure investments uniquely combine the following characteristics:

• The provision of ‘essential services’, hence low correlation with the business cycle.• Low price elasticity of demand for service and a degree of monopoly power, hence

pricing power that provides an inflation hedge.• Predictable and substantial free cash flow. • Attractive risk-adjusted cash flows, available over long periods of time. • Access to unlisted, illiquid financial assets.

Therefore, investing in infrastructure implies:

• Improved diversification.• Better liability hedging, including inflation protection.• Less volatility than capital market instruments.

Today, the infrastructure investment narrative is the only available ‘benchmark’. However,this narrative is defined a priori and has not been subject to a rigorous and comprehensive

1.1 Whyinfrastructure?

1

1

performance measurement exercise. Investors considering investing in infrastructure needto form return and risk expectations to make allocation decisions. In this book, we developasset valuation methodologies and data collection templates that aim to contribute to theeffective benchmarking of privately held infrastructure project debt and equity.

The need for benchmarks of privately held investments in infrastructure may seem incon-gruous at first. After all, infrastructure projects are lumpy and highly idiosyncratic endeav-ours. If every infrastructure project is different, what can investors learn from a benchmark?

In modern finance, asset allocation is not about picking individual investments, butinstead focuses on investing in groups of reasonably homogenous assets giving accessto remunerated risk factors. The performance of each of these groups can be evidencedby a benchmark.

In effect, the narrative described above is also a model that describes the expectedcharacteristics of the average investment in infrastructure. In turn, individual investmentsin specific jurisdictions, relying on one form or another of contractual or regulatoryarrangement, may only have some or none of these characteristics.

However, assessing the opportunity to increase allocations to privately held infrastruc-ture requires that investors understand the risks and performance to expect over timeand in different economic environments, and that regulators understand the risksinvestors are taking.

Benchmarking the expected behaviour of privately held infrastructure investments hasbecome necessary to allow investors to fully integrate such financial assets into theirasset-liability management exercises, as well as to allow regulators to calibrate the risk-based regulatory frameworks that make these investments possible (or not) for largeinstitutional investors.

The objective of building long-term investment benchmarks is simply to create the per-formance measures that investors and regulators need. These measures include theexpected return that reflects the dynamic risk profile of infrastructure projects and ade-quate risk measures for portfolio management (such as conditional value at risk), as wellas correlation measures between infrastructure investments and between baskets ofinfrastructure investments and other types of assets.

The information created by such benchmarks will be instrumental in matching the supplyand demand of long-term capital.

None of these measures are currently available to long-term investors in infrastructure.Indeed, creating relevant performance measures is not without difficulties. Issuesinclude the absence of readily available and comparable data, and the limitations ofusing listed proxies to benchmark lumpy and privately held financial assets like infra-structure equity or debt.

1.2 The demandfor benchmarks

Introduction

2

1.3 A roadmap

Documentingunderlying

instruments

To achieve this objective, we have designed a roadmap describing a series of appliedresearch and data collection steps to arrive at the desired measures of infrastructureinvestment performance.

Benchmarking long-term investments in infrastructure requires a two-level approach,starting with underlying instruments and then documenting the behaviour of differentportfolios built with such instruments (see Blanc-Brude, 2014a, for a discussion).

At the underlying level, five steps are necessary to clarify and document the performanceof infrastructure financing instruments, both equity and debt:

1. Define the relevant financial assets Improving the benchmarking and regulation of any type of investment first requires

well-defined underlying instruments. As infrastructure investment is currently ill-defined, the first step of our roadmap is the creation of unambiguous definitions offinancial instruments for long-term investment in infrastructure.

Indeed, as we show in Chapter 2 infrastructure assets are not real assets but finan-cial contracts. From an asset allocation perspective, industrial classifications such as‘roads’ or ‘power’ are close to useless. A first solution to the absence of a widelyagreed on definition of ‘infrastructure’, which we develop in Chapter 3, is to focus onproject finance debt and equity as defined in the Basel II Accord. Other approachesto infrastructure investment at the underlying level must also be developed, as longas they refer to well-identified financial instruments (for example, the equity capital ofcertain types of regulated network operators).

2. Design adequate valuation and risk measurement methodologies Once a clear and broadly accepted definition of underlying instruments is in place, ade-

quate valuation and risk measurement methodologies that take into account the infre-quent trading of most underlying infrastructure equity and debt can be developed.

By ‘adequate’ we mean that such methodologies should rely on the rigorous use ofasset pricing theory and statistical techniques to derive the necessary input data,while aiming for parsimony and realism in data collection. The proposed methodolo-gies should lead to the definition of the minimum data requirement (MDR), which isnecessary to derive robust return and risk estimates.

3. Determine the data collection requirements While ensuring theoretical robustness is paramount to the reliability of performance

measurement, a trade-off exists with the requirement to collect real world data frommarket participants. In particular, proposed methodologies should aim to minimisethe number of inputs in order to limit the number of parameter estimation errors.

Adequate models should also focus on using data points that are known to exist andare already collected/monitored, or could be collected reasonably easily. In all cases,

Introduction

3

data requirements should be derived from the theoretical framework, not the otherway around.

In reality, the amount of available data initially will be limited both in scope, since notall types of infrastructure projects exist in large numbers, and by time frame, becauseinfrastructure investments may have multi-decade lives and available records areunlikely to span such periods. Such data paucity can also be addressed if modelsare designed to allow for learning.

Whether the necessary data already exists or not, the determination of a parsimo-nious data set for asset pricing will also inform the standardisation of a new invest-ment data collection and reporting framework.

4. Standardise performance reporting Standardising infrastructure investment data collection allows for the emergence of

an industry-wide reporting standard, which can be recognised by investors and reg-ulators alike as best practice.

This reporting standard can increase transparency between investors and managers,who can be mandated to invest in a well-defined type of instrument and commit toreport the relevant data. Adequate reporting will also maximise industry participationand reduce the cost of compliance.

5. Create a database of infrastructure equity and debt cash flows Once the required data and a standardised reporting/data collection template is

identified, a database of infrastructure project cash flows can be built to apply themethodologies mentioned above.

Initially, historical data can allow for documenting the past performance of well-defined infrastructure debt and equity instruments. Later, the ongoing collection ofproject cash flows can permit the production of regular updates of the known per-formance of such instruments over time.

Once the adequate valuation and risk measurement methodologies have been deter-mined for a given type of financial instrument, and data collection and reporting havebeen standardised, the benchmarking of long-term infrastructure investments can effec-tively take place by focusing on the relevant performance measures at the portfolio level.

While a portfolio consisting of a representative basket of assets is the most intuitivebenchmark, this approach is virtually impossible for investment in unlisted infrastructuredebt and equity. Given currently available infrastructure investment vehicles, an investorcannot instantaneously buy a basket of assets that is representative of investable infra-structure projects in existence at that point in time.

It may be possible to invest in such a representative basket over time, but this may takeseveral years, by which time what constitutes a representative basket of infrastructureinvestments is likely to have changed with the evolution of public procurement policies.

Introduction

4

Building relevantportfolios

Therefore, the most useful long-term investment benchmarks are likely to be a combina-tion of well-documented ‘building blocks’ capturing systematic risk factors found inportfolios of infrastructure instruments. Such benchmarks can correspond to well-iden-tified investment strategies and combine the performance of the different groups offinancial instruments available in infrastructure finance to optimally achieve this explicittarget strategy.

At the portfolio level, three more steps are necessary to arrive at useful long-term invest-ment benchmarks in infrastructure:

1. Identify building blocks A number of risk factors can be expected to systematically explain investment per-

formance in infrastructure projects. For example, in Chapter 4, we show that the mostimportant of such factors are the contractual features of individual investment proj-ects, in particular to what extent they are exposed to commercial risks, as well as thedevelopment of the typical project life cycle, which, in infrastructure project finance,tends to correspond to the continuous deleveraging of the firm’s balance sheet.

Therefore, the risk/return profiles of most infrastructure project financing instru-ments can be grouped by revenue risk profile on the one hand and life cycle stageon the other.

In other words, at any given point in time, infrastructure debt or equity can corre-spond to either what is typically known as greenfield or newly built assets, or brown-field or already existing and operating ones. The former are typically riskier and yieldhigher returns, the inverse is often true for the latter.

The same debt and equity instruments can also correspond to, for example, tollroads and merchant power projects in the higher commercial risk/higher return cat-egory, and projects (also roads) that receive an income from the government basedon the availability of service or those that have a take-or-pay offtake contract (possi-bly a power plant) in which their future income is independent of market conditions.

Such projects exhibit less volatile cash flows and lower returns than those exposedto commercial risks. Therefore, any equity or debt instrument used to finance aninfrastructure project can be categorised according to such a two-dimensionalmatrix of revenue risk profile and life cycle stage.

While these are not the only systematic risk factors found in infrastructure projects,we review recent research in Chapter 4, which shows that such risk categoriesexplain most of the variance in credit spread levels in project finance loans.

Once the most homogenous subgroups of individual infrastructure finance equityand debt instruments have been identified, relevant investment strategies usingthese building blocks can be designed.

Introduction

5

The statistical validation of these insights is a key step in the roadmap towards build-ing an infrastructure investment benchmark, including ensuring that individual build-ing blocks exhibit low levels of correlation between themselves.

2. Define relevant investment strategies As they comprise long-term illiquid assets, a basket of infrastructure projects is not

easily or instantly investable. However, the building blocks discussed above can becombined to meet investors’ individual long-term objectives.

For example, Blanc-Brude and Ismail (2013b) show that combining a block of green-field debt with one of brownfield debt can create substantial diversification benefits(that is, increase returns and lower portfolio risk).

Therefore, along the greenfield/brownfield spectrum there is a continuum of efficientportfolios that can serve as a reference point for investors looking to build a portfolioof infrastructure project debt using available instruments over a given period of time.

For each strategy, and using the asset pricing and risk measurement methodologiesdiscussed above, various return measures (period return, yield to maturity, return inexcess of the investment base case), risk measures (expected loss, value at risk andexpected shortfall), and (effective) duration can be computed and inform both theasset allocation process and calibration of prudential regulatory frameworks or inter-nal risk models.

3. Investment benchmarks Finally, the strategies identified above can be used as long-term infrastructure invest-

ment benchmarks. Using historical data, the correlation of each strategy’s perform-ance with other investment opportunities, such as corporate debt or public equity,can be measured and estimated on an ongoing basis.

This last step answers the all-important question of the diversification potential ofindividual strategies using infrastructure instruments as underlying assets.

Implementing this roadmap requires an extensive data collection and modelling effort.Initially, historical data needs to be collected to calibrate valuation and risk models, andto provide a comparison with other asset types.

Such benchmarks can also be computed on an ongoing basis not only to continuouslyinform investors’ asset allocation choices, but also provide them with performanceassessment tools vis-à-vis infrastructure managers or their own direct investment pro-grammes. The adoption of standardised performance reporting by market participantswill be instrumental in this regard.

This book addresses the first three steps of our roadmap: it focuses on defining, pricingand documenting the behaviour of certain types of financial instruments widely used ininfrastructure finance – infrastructure project finance debt and equity.

1.4 This book Objectives

Introduction

6

Organisation

It also provides a comprehensive review of existing academic research on the nature andperformance of such instruments. It also develops valuation models and data collectiontemplates that can be applied by the financial industry and lead to the creation of globalinvestment benchmarks in privately held infrastructure.

Each valuation framework is designed with data availability and collection costs in mindwhile focusing on applying state-of-the-art valuation approaches that allow computingthe required investment metrics, (that is, not only expected returns but also risk meas-ures that are essential from an asset allocation, asset liability or prudential perspective).

This book is organised into five parts. Part I provides a comprehensive discussion of thetheoretical and empirical literature on infrastructure investments, and highlights the natureof underlying assets and the determinants of systematic risks found in such investments.

Part II reviews the rationale for building long-term investment benchmarks, outlines theempirical and theoretical challenges of valuing privately held infrastructure investments,and discusses several ways forward to building applicable yet theoretically robust val-uation frameworks.

Part III details the design and implementation of an infrastructure project debt valuationmodel using a structural credit risk approach that can easily be documented with exist-ing creditor data, as well as a debt renegotiation model that allows valuing the embed-ded options found in infrastructure project loans, in particular, the value of their ‘tail’.

Part IV proposes an application of the classic dividend discount model to the case ofinfrastructure project equity when complete time series of dividends are unavailable andmarket prices are not uniquely determined by an equivalent (replicating) portfolio of list-ed assets.

Finally, Part V summarises our findings and presents the data collection requirements toimplement the proposed valuation frameworks. n

Introduction

7