stochastic approach to risk assessment of project finance structures under public private partnerships

Trang 1

Stochastic approach to risk assessment of project finance structures under public private partner

Alves, Leandro F

ProQuest Dissertations and Theses; 2006; ProQuest Central pg n/a

Stochastic Approach to Risk Assessment of Project

Finance Structures under Public Private Partnerships The Thesis Presented

by

Leandro F Alves to

The School of Business

in partial fulfillment of the requirements for the degree of Doctor of Philosophy in the subject of Finance with a minor in International Business Advocate

Professor Theodore M Barnhill

The George Washington University

Trang 2

The quality of this reproduction is dependent upon the quality of the copy submitted Broken or indistinct print, colored or poor quality illustrations and photographs, print bleed-through, substandard margins, and improper alignment can adversely affect reproduction

In the unlikely event that the author did not send a complete manuscript and there are missing pages, these will be noted Also, if unauthorized copyright material had to be removed, a note will indicate the deletion ® UMI UMI Microform 3214152 Copyright 2006 by ProQuest Information and Learning Company

ProQuest Information and Learning Company 300 North Zeeb Road

P.O Box 1346

Trang 3

THE GEORGE WASHINGTON UNIVERSITY WASHINGTON DC 1 Si BZ a Zé PH.D PROGRAM “il aN

REPORT ON FINAL DOCTORAL DISSERTATION EXAMINATION

The undersigned Committee has examined Mr Leandro F Alves, a candidate for the

Doctor of Philosophy degree, on his dissertation entitled: "Stochastic Approach to Risk Assessment of Project Finance Structures under Public Private Partnerships.” The

Committee has found the candidate's work to be acceptable and recommends to the

Board of Trustees that he be granted the Doctor of Philosophy degree on

August 31, 2006

cử a) i? ự Professor of International

ERO Ủ £ Lo Banking and Finance

- Yoon Shik Park ¥ |

48 CẾC >——— Professor of International Business

and International Affairs Robert J Weiner aS me of Technology, Innovation and Entrepreneurship ias G Carayannis ——_-? ~~ Assistant Professor of Finance “Rébert Savickas Nedore MG bu ntld Advocate Theodore M Barnhill

hy LƠ Presiding for the Committee

Lg aw Kaw —_— on Doctoral Studies

Pradeep A Rau

May 22, 2006

SCHOOL OF BUSINESS

Trang 4

THE GEORGE WASHINGTON S| UNIVERSITY EBM) WASHINGTON DC PH.D PROGRAM

STOCHASTIC APPROACH TO RISK ASSESSMENT OF PROJECT FINANCE STRUCTURES UNDER PUBLIC PRIVATE PARTNERSHIPS by Leandro F Alves Bachelor of Arts University of Maryland, 1990 Master of Business Administration Marymount University, 1992 Master of Science in Information Management Marymount University, 1993

A Dissertation Submitted to the School of Business of the George Washington University

in Partial Fulfillment of the Requirements for the

Trang 5

This research is dedicated to my lovely wife, Valeria, who has

supported my academic pursuit through the years, and to my parents, Euro and Carmen, who served as perfect role models for my life

Stochastic Approach to Risk Assessment of Project Finance Structures under Public Private Partnerships

Leandro F Alves

Trang 6

Abstract

The dissertation provides a forward-looking methodology in order to quantify the risks of a portfolio of project finance structures As an application of such methodology, the thesis values the contingent government risk in Public Private Partnerships (PPP) for a set of project finance structures in a pre-determined time frame Since the early 1990s, there has been evidence, due to fiscal restrictions, that governments are utilizing PPPs to fulfill basic infrastructure requirements that were traditionally provided by the public sector The PPP project’s viability lies in the government’s ability to bear particular risks In order for investors and international financiers to be comfortable that the government fulfills its obligations under the PPP, a guarantee fund is envisioned Such guarantee fund is expected to be collateralized and directed to make payments under particular scenarios that future PPP projects encounter in a particular time frame The study provides a framework in estimating the asset requirements for the guarantee fund in a pre-defined time frame Traditional project finance structures rely on sensitivities test in assessing the robustness of the underlying project; such risk assessment has significant limitations Utilizing the Monte Carlo simulation to determine the level of contingent obligations the government may face over the life of a set of projects under the PPP program further expands the field of project finance in the area of risk assessment A Payment at Risk (PaR) methodology together with the Portfolio Simulation Approach (PSA), traditionally employed in the banking sector, may be used to determine the size of the guarantee fund to cover the government’s contingent obligations under their respective PPP programs for a pre-defined period

Stochastic Approach to Risk Assessment of Project Finance Structures under Public Private Partnerships Leandro F, Alves

Trang 7

Theodore M Barnhill

Professor of Finance

Department of Finance

Robert Weiner

Professor of Intl Business and Intl Affairs

Chairman, Department of Intl Business

Robert Savickas

Assistant Professor in Finance

Elias G Carayannis Professor of Management Science

Director of Research, Science, Tech Innovation and EURC

Yoon S Park

Professor of International Banking and Finance Department of Intl Business

Ph.D Dissertation Committee Members

The George Washington University Washington, DC 20052

barnhill@gwu.edu

Department of International Business The George Washington University Washington DC 20052 rweiner@gwu.edu Department of Finance The George Washington University Washington DC 20052 savickas(@gwu.edu

Department of Management Science The George Washington University Washington DC 20052

caraye@gwu.edu

Trang 8

ABBREVIATIONS

A$ Australian Dollars

ABN ABN Amro Bank

ARCH Autoregressive Conditional Heteroskedasticity ARMA Autoregressive Moving Average

BBW Best case/base case/ worst case (scenario risk analysis)

BC Build Contract

BLT Build, Lease, Transfer

BMdF Ministerio da Fazenda (Brazil’s Ministry of Economics)

BNDES Banco Nacional de Desenvolvimento Economico e Social (Brazilian Development Bank)

BOO Built-Own-Operate

BOOS Build, Own, Operate, and Sell BOOT Build, Own, Operate and Transfer

BOT Built-Operate-Transfer

bps Basis points

BTO Build, Transfer, Operate C$ Canadian Dollars

CAFDS Cash Available for Debt Service CAPM Capital Asset Pricing Model

CDI Certificado Depositorio Indice (local short-term index)

CFE Compania Federal de Electricidad, the Mexican Federal Utility Company

COFINS Contribuigdo social para Financiamento da Seguridade social (Federal tax for social security)

CPI Consumer Price Index

CPP Companhia Paulista de Parceria, a State of SGo Paulo owned corporation

DBFM Design, Build, Finance and Maintain DBFO Design, Build, Finance and Operate

DERs Departamento Estadual de Rodovias (the state highway departments), DNER Departamento Nacional de Estradas e Rodovias (the National Highway

Department)

DOT Department of Transportation

DtD Door-to-Door

GNSS Galileo Global Navigation Satellite System GOB Government of Brazil

GOCR Government of Costa Rica GPS Global Positioning Satellite

€ European Euro

EBRD European Bank of Reconstruction and Development

ECA Export Credit Agency

ECF Equity Cash Flows EI Economic Indicator

EIB European Investment Bank

Stochastic Approach to Risk Assessment of Project Finance Structures under Public Private Partnerships Leandro F Alves

Trang 9

Eletrobras Centrais Elétricas Brasileiras S.A., (the national electricity company) EMA Exponentially Moving Average

EMU Economic and Monetary Union

EU European Union

EWMA Exponentially Weighted Moving Average FCF Free cash flow

FDI Foreign Direct Investment

Fiduciary Fund Brazilian Federal Government Fiduciary Fund

Forecast Methodology A forecast methodology based on Barnhill and Maxwell’s (2000) paper

GARCH Generalized Auto-Regressive Conditional Heteroskedasticity GDP Gross Domestic Product

GTIP Golden Horseshoe Transit Investment Partnerships GWU The George Washington University

HK$ Hong Kong Dollar

HPP Hydroelectric Power Project HS Historical Simulation

IBRD International Bank of Reconstruction and Development (World Bank) IDB Inter-American Development Bank

IFC International Finance Corporation IGCC Integrated Gasification Combined Cycle IGPM Indice Geral de Precos do Mercado

l£ Irish Pound

IPC Indice de Precos ao Consumidor

IPEA Instituto de Pesquisa Econémica Aplicada IPP Independent Power Producer

IRR Internal Rate of Return IRS Internal Revenue Services

ISS Imposto Sobre Servicos, a Value added Tax (VAT) ITF Inside the Fence Projects

JBIC Japan Bank for International Cooperation JERI Japan Economic Research Institute

km Kilometer

Ke Cost of equity

KW Kilowatt

£ British Pound

LIBOR London Inter-Bank Offer Rate LNG Liquefied Natural Gas

LTO Philippines’ Land Transportation Office

M$ Malaysian Ringgit

MGFF State of Minas Gerais’ fiduciary fund

MME Ministerio de Minas e Energia, (Ministry of Mines and Energy) MPT Modern Portfolio Theory

MW Megawatt

MWh Megawatt hour

NHS Act National Health Services Act (or Private Finance Act) NIC National Infrastructure Corporation

NK Norwegian Kroner

NPC National Power Corporation, a Philippines federal utility company

Leandro F Alves

Trang 10

NPV NSW OP OPI OPT PaR PC Petrobras PFI Picko PIS PLC PPA PPP PSC QMV Quotas Reais SMA SPC SPV PRI PSA RAROC REI RGR RPI R$ S&P Scut TENs TIP TJLP TOR UK US US Exim US$ UTFP VAT VEM WACC World Bank v ZZL

Net present value New South Wales Outright Privatization Omnitel Pronto Italia Option Pricing Theory Payment-at-Risk, Percentage change

Petroleo Brasileiro S.A., (the Brazilian state owned petroleum company) Private Finance Initiative

Private Infrastructure Investment Center of Korea

Programa de Integracdo Social (Brazilian Federal tax for social programs)

Percentage log change Power Purchase Agreements Public Private Partnership Public Sector Comparator Quasi-Market Valuation

Ownership interest in Fiduciary Fund Brazilian Reais or Real (Brazilian currency) Simple Moving Average

Special Purpose Company, also known as SPV Special Purpose Vehicle, also known as SPC Political Risk Insurance

Portfolio Simulation Approach Risk adjusted return on capital Regional Economic Indicator

Receita Global da ReversGo (Brazilian Federal value added tax) Retail price index

Brazilian Real Standard & Poors’

Sem Cobranga ao Utilizador, Government of Portugal paid shadow tolls Trans-European Network

Transit Investment Partnerships

Taxa de Juros a Longo Prazo (long-term interest rates) Transfer of Operating Rights

United Kingdom

United States of America

Export Import Bank of the United States United States Dollar

Unita Tecnica per la Finanza di Progetto (a task force created by the

Italian Government for PPP projects)

Value added Tax (in Portuguese it is ISS) Value for Money

Weighted Average Cost of Capital

International Bank of Reconstruction and Development (IBRD) Japanese Yen

Zuiderzeelijn Rail Line Project in the Netherlands

Trang 11

Table of Contents

I INTRODUCTION AND LITERATURE REVIEW .:::ssccssssscccsccssscsssssscsssccsssensssees 12

1.1 0112:0141) — 12

1.2 PROJECT FINANCE RESEARCH (THE NEED FOR EXPANSION) các nen se 13

1.3 RESEARCH QUESTIONS .- GQ nung và 17

TW 0 1v001) 000/3 0 18 1.4.1 Project Finance LH€F@IUF€ ÍĐ€VI@W LH TH kh kh như Hy 18 14.11 Financial Assessment of Project Finance SUructur es .ccccccccceccccsseseessseecsssness 20 1.4.1.2 Monte Carlo Simulation for Project Finance SITHCÍUF€S cecccesS- 22 142 Public Private Partnerships (PPP) Literature Đ€VI€W à cào 26 143 Value at Risk (VaR) Methodology Literature J€Vi@YW àà St Seo 27

1.5 THEORY FOR THE RESEARCH .:ccseseceeecuceescececccsceceeusesseeaueneusssesenessecseseseseseeeereseenenes 31

1.6 APPLICATION OF THE METHODOLOGY Q0 HH HH gu ve 31

1.7 MAIN OBJECTIVES OF THE RESEARCH << SH TH HH nh 32

H PROJECT EFINANCE STRUCTURE OVERV LƯY su con HH 11990 s56 33 2.1 PROJECT FINNANCE STAKEHOLDERS ¬- 34 211 Á SDOHSOFS/EQUHIHV (ƯWHƠF.Ả Gv Hy HH TT HH HH nàn 34 2.1.1.1 Appeal oƒ ProjeCL FÌHqHCIHW ong kg ki kh, 35 2.1.1.2 MH 6 5 nh .ằeằ 35 21.3 Strategic EXPANSION icccccccccccscceesssecceneecsneeeseeeeesceneeteaneesesescentessuanessueeessunnenaes 36

2.1.1.4 The Sale of Goods 6s nẽeốnốee 36 2.1.2 HOSI ŒOVEYHIHGHHÍ Q S ST ng nh HT TT vn Tự 3ĩ PI PT H ) á8 aaa- Ố.ỐỐ.ỐốỐ.Ố 37 PI /ŠZN/‹ si : 2a<a 37 2.1.2.3 Attracting New Capital na ốốe 38 2.1.2.4 Technology TYaHSƒGT/TTYÌHÌHđ Q SH HT RA T4 ng ngàn 38 2.1.2.5 Competitive AAVANLAC 0.0 ccccccccccesccsceesscssceeeneceneccaeeeseeeseesseecsetsatenaeeeaeensesaneseeeeas 39 213 @/27/1174727/,86027/174/;s PS PP- 39 2.14 [1//278 80,1.2051/)/)721.22EN0aẢ 39 215 0/72 40 2.1.6 (0) 11.) ađaadđđđiiiiiiỒỠ - 4] 2.17 LOA S .QQ TTnnHn ng KT gi HH cà TT ĐT ng 5 ng kc 4] 2.18 Adbisors qHd COHSHÏÍQPES Q Q SG TH TH ng HH KH ng ng nen 42

2.2 FINANCIAL STRUCTURE AND MODALTTIES OF PRIVATIZATIƠN c.c.< << <ss<- 43

2.2.1 Build Conmtract (BC) C10101 TS ĐK KTS C115 1102 104 8 k vn HE 44 2.2.2 Build, Transfer, Operate (BTO) cccccccccccccceccccecesesesseceseneeesesaesesesssessssssnssecesassanaes 44 2.2.3 Build, Lease, Qperadfe (BL() cá cá KH KT ng HH ng iu 44 2.2.4 Lease, Operate, TransƒeF (LOT) à.- St TH TT kHỦ 45 2.2.5 101,00 N W (0 ) 20/17/00 00006 ga 45 2.2.6 Bulld, Operale and Transƒer (BIT) c kkckxnknHTHnH KH HH ke 45 2.2.7 Build, Own, Operate and Transfer (BOOT) .cccccccccceccetetseeeeetiesseennnsseeeesnseees 45 2.2.8 Build, Own and Operafe (B) ST HT ng nghe rệt 46 2.2.9 Build, Own, Operate, and Sell (B(Š) - -G SH HH rệt 46 22.10 Rehabilitate-Operate-Transƒfer (NOT) Ặ ST HH ào 47

Trang 12

2.2.11 Lease-Rehabhilitate-Transƒfer (LÌT) ĂẶ SG S TS HH HH Hước 47 2.2.12 Transfer of OQperating Rigls (TOR) cS S2 S SA nSnH HH kiệt 47 2.213 Trside-the-Fence ProJeCfS (TÏÍF) TH ng HH ng kg kh cu 47 2.2.14 Outright PrivatizaHon (OP) SH c HT ng TH TH kg tr 46

2.3 RISK ALLOCATION IN PROJECT FINANCE ccccccccccsesssececescseececceeueveveseeseueeeauaeseesceseeenens 48

2.3.1 Contractual AFVQAGCMENES oo N0 Ha 48 2.3.1.1 Off-take Algreemeni (Take-or-Pay, Take-and-PAqy) -sc cà kiệ, 49 2.3.2 J/ nh Ầ ẦẦỐỐS 20 2.3.3 TẪ Q QQQ QC KE ĐC KP k4 99 30 2.4 A SAMPLE OF PROJECT FINANCE STRUCTURES BY REGION .cccccscsssesssecseeeeceeseceeeees 51 2.4.1] ya 6) 1 RE TT nGŨỘŨ 5 32 2.4.2 NV 78.1, 27692NEENNNớA tgớ 52 2.43 N/247/8.1/24// 0 NYYNNGHaa á Iáä4ỈI HH 33 2.4.4 “¡0 33 2.4.5 “r3;;// 00788 33 2.46 2.0001 Ề7®e —- J4 2.4.7 l/12/2/I-2721211 0N NNMẳẳũẮŸÃẮỶẢỶẲẢ 34 2.4.8 [24/120 EESEnh 33 Ill PUBLIC PRIVATE PARTNERSHIP (PPP) OVERVIEW V SỐ

3.1 EFFICIENCY GAINS BY PPP SCHEMES .cccccccscssesencsccvececcessecceceeeceaeaessuececaneseeaaasereres 57 3.2 A SAMPLE OF PUBLIC PRIVATE PARTNERSHIPS BY REGIƠN - - co c2 58 327 EHẨOJĐƠ HH HH TT KT Tà KT Tà k1 1k T00 8k krp ĩ0 3.2.2 Nyz7;/.0.:,, 27.00707876 ộ.ộ.ộ.aaAAẦ 62 3.2.3 D(2041/5-1//127142 Ea ‹{đaa 62 3.2.4 ẢÁ SỔ Q LG TT ng TT ng cá HH S9 Ti TK tk 63 3.2.5 AUSUF LAG o.oo Q QQQQ TQ TS HH ng ng ng ng TK go no C0 1 k K KEEve 64 3.2.6 171,084 64

3.3 PRACTICAL ISSUES FACED BY GOVERNMENT”S PPP PROGRAMS - 65 IV Š INERASTRUCTURE PROJECTS IN BNAZÌL - 5555555 <<<s<s<<<ssxse=se 66

4.1 BRAZIL’S PPP INITIATIVE .cc ccccececccsscocsccecccuccecececvsseveeeeaveveceueusecersauecessunesseeenesesenseees 68 4.2 PUBLIC SECTOR ISK - LG G0 gu Ki cá Km cv vs 72 4.3 OVERVIEW OF TRANSPORTATION SECTOR .cccccccccssseecscecccseeesevensessueesseanceserceescsensaees 74 4.3.1 V)N E0GỎ 74 V CONCEPTUAL FRAMEWORK IN PUBLIC PRIVATE PARTNERSHIPS 75 5.1 I9 (0 29):9/.009)06)9019 11 76 5.11 VI () 0.0159.1.).) 7 TINN ẽ raỤỤỤỪDỮD ỮƯG 76 ` “NHI na nenene 77 (101 n eH 77 "h9 ri nan 77 2n N.I.,./,/2: n Ơ 78 TT .,/ L 8n .4AAĂ all ááảằ 78 3.1.LĨ COHC€SSIOH TGFIHIHGEOH TH HH Tu nh ph TT kế 79 Z NA: nnn TA ẫăĂäĂ 80 5.1.2 AlloCAtiON Of RiSK cccccccccccccesssssccccsseeccsssecsseeeceseesseecseeeeseseessaaesssaasasensseseessseess 80 VI ROADMAP FOR EMPIRICAL ANALYSIS OF RISK ASSESSMENT .ccc00 32

Trang 13

6.1 BACKGROUND FOR THE MODELING EXERCISE cccccssecsececnccvecesccvscssonsacecaveenecessanseees 82

6.2 — EXAMPLE MODEL OF TWO PROJECTS Ăn HH nh ni 83 6.2.1 Example \áodel — Toll ĐOdidÏ Ï cà tt HH TH ro 83

6.2.2 Example Model — Toll ROG 2 sac HH HH HH nh 83

6.3 THEVAR APPROACH IN THE ANALYSIS OF RiSK PROCESS ẶĂ Sex 84 6.4 MODELING THE ECONOMIC SCENARIO SH HH HC 86 6.4.1 hy/ /22/27//.20000n0nẺ8n858 &7 6.4.2 Estimating Historical VolaHlities and CorrelHOHS Series 88 6.5 90961.090.010 88 TA N cài a0 090G 89 6.6.1 Parameters for PAR ANQlysis .ccccccccccccsccccescseesceenseeesseetenseceenieeeseeeneveeeeesteeeeneteas 90 VIL DATA UTILIZED IN THE EXAMPLE MODELS u ccssscsssssssseescsessssesessssneeness 91 7.1 ALL SECTORS IN THE BRAZILIAN ECONOMY .esceseeeeeeteeeccssseeseesesasentneueeseeeeeeseeeaes 92 7.11 Financing Assumptions for All Sectors the Brazilian ECOHOI 92

7.2 TRANSPORTATION SECTOR (TOLL ROADS) ¬ ne ees eeee eine eeeeaeeeagnenaeeeen tans 96

7.2.1 CÀ/27/1/.89/31 2/8 017)//70779,/ 00000806 na ằẶằaa 9ĩ 7.2.2 6/110//1.2/0.111,.//J2727.0000n0n8n08 3 97 MIIR I32701231.5)(0)5Ẽ0.097.9054)17 98

8.1 INTRODUCTION TO TOLL ROAD REGRESSION ANALYSIS cccccccccceccevesseeeeeseeeseeseeseeess 98 8.2 DATASET OF BRAZILIAN EXISTING TOLL ĐOADS c ch re rkee 100 8.3 REGRESSION EQUATION FOR BRAZILIAN TOLL ROADS cà <cccrceeres 100

8.3.1 Specifications of Newly Concessioned Toll Road Traffic Projections 100 8.3.1.1 Regression 0ƒ Newly Concessioned Toil Đoddl§ net ieeeied 100 8.3.1.2 Regression Equation for Newly Concessioned Toll Roadls 102 &.3.2 Specifications of Existing Concessioned Toll Road Traffic Projections 103 8.3.2.1 Regression of Existing Concessioned Toll RĐoddlS cĂ sài 103 8.3.2.2 Regression Equation for Existing ToÏlÏ RoddiS Ăn si eiixey 104

§.4 SPECIFICATIONS FOR EXAMPLE MODEL (1.E TOLL ROAD TRAFFIC MODEL PROJECTIONS)

106

IX Š RESULTS OEF PAR ANALYSĨS -<c cĩ HH n9 0000088844068666 109 9.1 OVERVIEW OF A PORTFOLIO APPROACH VIA PAR METHODOLOGY 109 9.11 Government Fiduciary Fund for PPP Contingemt QbligaHows 110 9.2 99)::120.00090.1.000) 7 .— 110 9.3 COMPARISON BETWEEN BBW AND PAR - - Linh TH HH kh 112 9.3.1 [17422/117/7290227//58625./7.000N0n808 112 9.4 FINANCIAL STATEMENT OUTPUTS .c:cccccceceesseeeessuecenesesaaeseeanecencsesseenneasenaeeeseaneesenes 117 9.41 Financial Statements for Toll ĐOđdl Ì kh KH ng kh 117 9.4.2 Financial Statements for Toll RĐOdđdl 2 à HH KT Hy 122

9.5 09c ẽ ch 125

95.1 Forecasting PaR ƒor Toll ĐoddÏ Ï ác xxx HH HH 125 9.5.2 Forecasting PAR for Toll ĐOddl 2 sa kh tk ng KH kg hu 127

96 ANALYSIS OF PAR ON A PORTFOLIO OF PPP PROJECTS -.- cà sxxcss 128

9.7 FINDINGS OF THE RESEARCH Ăn HH nọ Ho HH 130

971 Measurement of Individual Project ĐSK- chinh 130

9.7.2 Measurement of Portfolio Risk im Project Fimance SIYHCHIF€S' 131

Trang 14

X LILMTTA TION AND EUTURE RESEARCTH 0 n1 rseer 131

II LIMITATIONS woo ˆ ẻẽe.- 131

10.2 DIFFICULTY N FINANCIAL PROJECTIONS FOR TOLL ROADS - 131

10.2.1 Car demand forecast — the standard pFOC€đÏMF6 SĂằẶcseseisecee 132

10.22 — Car demand ƒforecast— the simplifled DFOC€dÏHF€ Ă ằằS Site, 133 102.3 — Traffic forecast ƒOr tỌÏ FOddlS à TS SH HH HH HH nh nh te 133 XI NATURE OF RISK AND GENERAL CONCEPTS, AND THE PRACTICAL APPLICABILITY OF METHODOLOGY TO OTHER LINES OF WORK 134

11.1 NATURE OF RISK AND GENERAL CONCEPTS ST St tk key 134

11.2 PRACTICAL APPLICATIONS Ăn HH KT Ho HT in TH 135 11.3 THE EDUCATION SECTOR AS AN EXAMPLE OF A PRACTICAL APPLICATION 136

11.4 MARKETING AND FUTURE OPPORTUNITIES - Ăn kg khiết 137

39:33 0 1 139

ANNEX I - EXTENDED LITERATURE REVIEW uc co ch mg 0686060158 139 ANNEX H - A SAMPLE OF PROJECT FINANCE STRUCTURES BY REGION 139 ANNEX ITI - A SAMPLE OF PUBLIC PRIVATE PARTNERSHIPS BY REGION 139 ANNEX TV — ONE YEAR SIMULA TION so H00 0 600508066566 139

Leandro F, Alves

Trang 15

I Introduction and Literature Review

1.1 Overview

There are two distinct types of finance modality, (i) project finance and (11) corporate finance, or a combination of the two (limited recourse financing) Project finance utilizes the underlying project revenues as the source of funds from which loans are to be repaid; In contrast, corporate finance seeks financial guarantees from the sponsor (i.e developers of the project) for debt repayment

Every project financing is inherently different; nevertheless, there are some common traits in the majority project financing These commonalities include projects developed through a separate legal entity, and the obligations of the sponsors are limited

and typically off-balance sheet (in the case of limited sponsor recourse), if at all (no

sponsor obligation is known as non-recourse project finance) Additionally, project finances structures also allow the sponsors seek to leverage the project (in order to increase the returns on equity) by maximizing the debt (the level of debt is based on the cash flow potential of the project) Project assets and revenues are given as collateral to the lenders Lastly, contractual arrangements related to the project (1.e construction contracts, fuel supply, or purchaser of the project’s outputs) are interrelated to the viability of a bankable project

Past literature in project finance, included research related to the risks associated with such structures Most papers in the field are based on the traditional risk analysis under uncertainty that focuses on scenario analysis, or Best case / Base case / Worst case

Trang 16

(BBW) analysis Such approach has a numerous deficiencies including limitations on the number of possible outcomes associated with the project, in which the assessment of the project’s fundamental value is vulnerable and may not actually reflect the true risk associated with the project The traditional approach includes the Net Present Value (NPV)! exercise, but NPV alone is not sufficient to accurately predict the risk of the project, since NPV does not particularly defined the risk of not accurately forecasting

future cash flow stream

1.2 Project Finance Research (The Need for Expansion)

There have been papers have surpassed the traditional risk analysis approach (see above), but are still limited to an analysis of one project, which does not take into consideration a group of project finance structures Research papers have focused on one project with simulation analysis, or a number of projects that have an underlying commodity, but there has been limited research in analyzing a group of projects that do not have an underlying commodity

In practice, more companies are engaging in numerous project finance structures, governments are concessioning a variety of project finance structures whereby the government assumes some of the risks, and suppliers and other related companies (both in the debt and equity side of the balance sheet) are encountering multiple projects The summation of their individual project risks does not necessary equal their aggregate risk

! The NPV exercise is based on the view that the same money received in the future is not worth as much today (this is because money received today can be applied and obtain a higher value in the future) The NPV exercise establishes a current value for the future cash flows, by applying a discount rate to such future cash flows

The George Washington University, June, 2006

Trang 17

of the portfolio of projects This is largely due to the correlation of national and regional economic variables that affect the various projects differently.” Lastly, there is limited research in analyzing a group of heterogeneous projects (of a project finance structure) that do not have an underlying tradable commodity (i.e oil, energy, or natural gas)

The dissertation thesis (Thesis) takes a step forward in answering the question: “How to accurately quantify the risks of a portfolio of project finance structures.”, by utilizing a methodology, traditionally applied in the banking sector The combination of a stochastic approach with Portfolio Simulation Analysis (PSA) furthers the understanding of the underlying risks associated with project finance structure, and brings a new approach to examining risk of multiple projects without an underlying tradable commodity

On a practical level, the Thesis provides a framework for valuing contingent government risk in Public Private Partnerships (PPP) for project finance structures PPP

projects differ from traditional public concession project finance structures due to the

modifications of the regulatory and legal framework, which enables the respective

governments that establish a PPP framework to take substantial risk, on a project-by-

project basis, with the private sector Another difference between a project finance of publicly concessioned infrastructure projects and PPP is that under the PPP programs, the government risk and rewards are similar to equity participants, whereby the government

Projects maybe positively or negatively correlated, depending on the correlation of the national economic indicators or regional economic indicators that impact the projects In the case of toll roads,

if both toll roads are owned by the same company, clearly the risk of both toll roads under performing

is less due to its inherent dynamics of the economic indicators

Trang 18

takes a larger portion of the risk in particular occurrences of each project and may take an up-side of the project if the project performs better than expected

In the case of toll roads, an example of a traditional toll road concession largely involves the government concessioning out the road in order for the private sector winner of the concession to be able to toll the road (existing or greenfield’) In some cases the concessionaire is required to pay an annual fee to the government (known as a cannon

payment), if this modality is selected the highest bidder (in terms of cannon payment to

the government) wins the concession, and the toll prices are pre-set prior to the bidding of the concession Adjustments to such tolls are based on local economic indices (i.e in the case of Brazil, IGPM)* Another modality, the concessionaire wins the bid if they offer the lowest toll for that project (such modality allows the user to benefit from the concession process) In either case, the government does not assume project risk, in terms of government cash outlays for the toll road

In case of PPP concession project the government may assume some risk, (also using toll roads as an example), such as providing a minimum revenue guarantee to the winner of the concession (as such the government is obligated to pay a limited amount, if the traffic of the concessioned project is insufficient, as stipulated in the concession agreement) PPP modality allow, previously unviable projects, to be concessioned to the

A greenfield toll road is a toll road that does not currently exist, and the majority of the road will have

to be built, including tunnels and bridges Rights of way from existing property owners, where the toll

road will pass, will need to be obtained

Further details on Brazilian indices, including IGPM are described in detail in Chapter IV, titled: “Infrastructure Projects in Brazil”

Stochastic Approach to Risk Assessment of Project Finance Structures under Public Private Partnerships Leandro F Alves The George Washington University, June, 2006

Trang 19

private sector, whereby the government assumes some risk sharing of the project Another possible PPP style transportation project is by shadow tolling, whereby the government pays a fixed amount per car that passes through the tollbooth As such, the toll price can be significantly reduced, possibly generating additional traffic This approach transfers a portion of the cost from the end user (drivers paying the tolls) to the government It is also common practice for the government to receive an up side if the

project performs above expectations (as defined in the concession contract) This is a

result of the government is assuming some of the downside risk (which is typically an equity characteristic, although the government is not involved in the day to day management of the project, nor does it participate on any decision making process)

The study provides a methodology for governments to size their contingent risk

obligations under their PPP arrangements, and utilizes the Government of Brazil’s (GOB) PPP program in transportation as an example Since the early 1990s, there has been

evidence that, due to fiscal restrictions, governments are utilizing PPPs to fulfill their country’s basic infrastructure requirements that were traditionally provided by the public sector PPP project’s viability lies in the government’s ability to bear particular risks In order for investors and international financiers to be comfortable that the government fulfills its obligations under the PPP, a financial vehicle is envisioned Such financial vehicle is to be collateralized and directed to make payments on behalf of the government’s obligations under particular scenarios, as defined in the respective concession agreements The study provides a methodology, which could be used to size such financial vehicle Utilizing the Monte Carlo simulation on economic indicators and

Trang 20

regional economic indicators, and their correlation, allows one to determine the level of contingent obligations the government faces over a pre-determined time-period of a group of designated projects under the PPP program Such analysis further expands the field of project finance in the area of risk assessment

A Payment-at-Risk (PaR) methodology together with a PSA is demonstrated, and can be used to determine the size of a fund the government needs to establish to cover its contingent obligations under their respective PPP programs for a determined time period The term PaR has been coined in this Thesis, by the author of the Thesis, and utilized instead of the well know Value-at-Risk (VaR) in the banking sector The shift from VaR to PaR is primarily because the VaR is utilized for valuing investments, while the PaR is particular for assessing the payment risk of the government due to the government’s contingent obligations in their respective PPP program (as such, this Thesis refers to PaR and VaR interchangeably)

1.3 Research Questions

The study provides a methodology for assessing risk for a portfolio of project finance structures, and utilizes example models in the transportation sector, toll road sub- sector of Brazil to demonstrate its usefulness The reason for selecting the Brazilian toll roads as the example models is due to numerous factors, principally being the availability of historical traffic data, and the availability of tested financial models of project finance structures in Brazil Other factors for selecting the toll road sub-sector include the availability of national and regional economic data inputs for such project finance models, and the GOB’s interest in such methodology The GOB’s approvals of PPP laws Stochastic Approach to Risk Assessment of Project Finance Structures under Public Private Partnerships Leandro F Alves

Trang 21

in 2005, enables PPP projects to materialize, and provide immediate relevance to this

Thesis

The main research question the dissertation addresses to the overall contribution in the project finance literature with regards to accurately measuring the risks in a portfolio of project finance structure for a pre-determined time period Sub-questions address the ability of the methodology to accurately measure the resources necessary to mitigate numerous project finance structures simultaneously

Research Question: How to quantify the risks of a portfolio of project finance

structures

Sub-Research Question 1: Will the VaR methodology with a PSA provide a measurement for the Government of Brazil’s contingent obligations under their PPP initiative

Sub-Research Question 2: How to estimate government’s contingent liabilities?

1.4 Literature Review

1.4.1 Project Finance Literature Review

The project finance literature, given its diverse nature, incorporates a number of disciplines including: legal, engineering, mathematics, finance, environment, economics, insurance, and public policy The relevant literature for the research focuses exclusively on the financial modeling applications dealing with project risk, although all the above fields play a significant role in project finance and the financial models ultimately are outputs to the many disciplines listed above

Trang 22

Although the label “project finance structures” has been around since the 1950s, academic research and advancements in the field are relatively recent, from the late 1980s The literature review presented in this Thesis focuses exclusively on the aspects relevant to this research as they pertain to project finance, such as: (i) financial modeling,

and (ii) probability of an occurrence utilizing Monte Carlo simulation Only the literature

review that directly contributes, as a basis for the Thesis, is described in the subsequent

paragraphs, an extended literature review can be found in Annex I

Under the financial modeling approach in project finance, there have been a number of recent advances in practice and encompassing theories from the general finance literature The relevant research to date in financial modeling of project finance is diverse and touches upon a number of areas including Risk Adjusted Return on Capital (Raroc), and Weighted Average Cost of Capital (WACC) In addition, other research has also analyzed the discounting Free Cash Flows (FCF), discounting Equity Cash Flows (ECF), and the Quasi-Market Valuation (QMV) Lastly, studies that are more recent have

explored the net book value methodology, comparable sales methodology, and the

income capitalization approach

With regards to the probability of an occurrence, project finances research to date has focused on: (i) general risk analysis including utilizing the BBW approach, and (ii) Monte Carlo exercise for commodity based project finance The Thesis takes an evolutionary step in providing a methodology to calculate a portfolio of projects, which its fundamental revenues do not require to have a transferable underlying commodity

The next paragraphs illustrate the most recent literature that directly contributes to the

Trang 23

Thesis The extended literature review in Annex I, highlights the authors contribution to the field of project finance, and their limitations, which provides the reader with the

various approaches researchers have taken in the field of project finance in analyzing risk

1.4.1.1 Financial Assessment of Project Finance Structures

The papers mentioned in this paragraph illustrate the recent trend of project finance literature, which are further detailed in Annex I One of the earlier papers where a financial tool was applied is the Lee and Eom (1989) paper, where the authors presented a goal-programming model for project financing strategies to support negotiations between lenders and borrowers Han and Shi (1997) also displayed a quantitative approach in assessing project finance structures by measuring the return on equity and Risk Adjusted Return On Capital (Raroc) Miller and Zhang (2003) utilized a benchmark for local market returns and Chinese stock exchanges, the authors estimated individual company risk parameters and industry sector average beta coefficients using the Weighted Average Cost of Capital (WACC) In a different approach, Urquhart and Siegel (1999) employ gaming theory in the financial model of United States (US) power assets The Dell et al (2004) paper, the authors analyze Fitch Rating Agency’s approach to project finance In Percopo and Haller (1999), the authors took a different approach than the

previous papers mentioned, and emphasize the ability of insurance to bridge many gaps in

project finance structures

The above-mentioned papers offer different perspectives in assessing project

finance structures, although none attempted to quantify a portfolio of project that do not have an underlying commodity as the basis The Esty (1999) paper more closely is

Trang 24

applicable to the Thesis, in which it demonstrates the deficiencies in using the traditional approach (constant discount rate for cash flow for all periods) in project valuation, and provides a glimpse of the direction of future research The author illustrates the two most common approaches: (1) discounting Free Cash Flow (FCF) using the WACC and subtracting debt values; and (ii) value of equity directly, by discounting Equity Cash Flows (ECF) using the cost of equity (Kg) Both of these methods utilize a constant

discount rate, which the author demonstrates to be incorrect, since the cost of equity and

WACC change as leverage changes Project finance, by its very nature, has differentiating capital structure each year, as debt is obtained and later amortized The author describes the Quasi-Market Valuation (QMV) as an alternative approach to

valuing project finance structures, where multiple discount rates and QMV, both improve

the valuation technique for large infrastructure projects The author also discusses Monte Carlo simulation to analyze the uncertainty of the cash flow stream and real options methodology for manager’s flexibility with regards to strategy The benefit of the paper

is its clear approach highlighting the deficiencies in the constant discounted cash flow and

the possible improvements in valuation technique Its shortcoming is that it relies on a traded underlying commodity (i.e crude oil) for the real options approach utilizing Monte Carlo simulations The approach presented in the Thesis advances the authors’ research by applying the Monte Carlo exercise combined with the PSA to any project finance

structure, without requiring to have an underlying value based on a traded commodity

Leandro F Alves

Trang 25

1.4.1.2 Monte Carlo Simulation for Project Finance Structures

The more applicable literature to the Thesis is the literature related to the Monte Carlo approach to project finance structures Savvides (1994) introduced the Monte Carlo simulation technique of investment projects to analyze and assess risk His paper

provided a detailed account of risk analysis and the use of Monte Carlo simulation to

enhance such analysis on a project-by-project basis The Thesis utilizes many of the concepts set forth in Savvides paper, but takes a further step in applying the Monte Carlo simulation for a portfolio of projects

A later paper, the authors’ Songer, Diekmann, and Pecsok (1997) demonstrate Monte Carlo risk assessment methodology for revenue dependent infrastructure project Traditional toll road privatization projects utilized financing statements that do not

include adequate risk assessment for uncertainty The authors state that traditional debt

coverage calculations (used in project financing), which are point values do not include the likelihood of such values are to occur nor does it identify the most critical variables that affect project risk Their paper highlights the importance of Monte Carlo simulation to both identify the most critical values and to assess the likelihood that the debt service would not be paid Their paper has the deficiency that it focuses on one toll road project and not on a number of projects to analyze the projects’ correlation with one another,

given by regional economic factors that affect the underlying projects Thus, the Thesis

advances in the field, which captures the portfolio of projects and the correlation of the factors that affect each project

Trang 26

Rode, Fischbeck, and Dean (2001) also utilized Monte Carlo method, as the previous authors, and demonstrated, using a nuclear power plant as an example, that such methodology encapsulates project valuation in a higher degree than other appraisal methods such as: (i) net book value, (ii) comparable sales, (iil) reproduction or replacement cost, and (iv) income capitalization The authors defined the methods

including the net book value, which is the original cost of the asset plus additional capital,

costs minus capital departure and depreciation The comparable sales method is used when the asset is liquid and there is a market for such asset (based on similar assets) The authors have also defined the reproduction or replacement valuation method, which is based on replacing the original asset with the same craftsmanship and materials as the original Lastly, the authors examine the income capitalization approach, where future cash flow is discounted providing a value for the asset

All four traditional appraisal methods described above do not fully consider the inherent risks involved in the underlying project The first three are based on an exact moment in time, normally the moment of possible sale of the asset, while the forth method, income capitalization, view the future net cash flow of the project, but also disregards the occurrences that may affect negatively/positively the net cash flow of the

project The authors develop a simulation model that incorporates uncertainty in three levels: (i) level one is that an event occurs, (ii) level two is when the event occurs, and (iii) level three is the event’s cost, in particular, how the event affects the valuation of the

Stochastic Approach to Risk Assessment of Project Finance Structures under Public Private Partnerships Leandro F, Alves The George Washington University, June, 2006

Trang 27

project The authors also depict correlation between variables” that ultimately affect the project valuation

The authors group the risk into three categories: (1) market or economic risk, (11) technical risk, and (iii) political risk The main deficiencies of the work is that it only focuses on one project, and that the project has an underlying commodity (i.e energy) that can be compared to other generation projects in the energy grid, facilitating the assumptions and the valuation of such project The Thesis advances such methodology by utilizing PSA and not requiring a tradable underlying output of the project finance

structure for evaluation

Of all of the literature reviewed, the most recent, Rode, Lewis, and Dean, (2003), and Rode, Lewis, and Fischbeck (2003), build a framework for utilizing simulation analysis as a base to derive financing risk and capital structure for project finance structures where there is an underlying commodity (i.e energy) The authors examine the problems associated with BBW and the usefulness of simulation analysis Using such simulation analysis, the authors are able to explore issues in capital structure and decision

making under two scenarios: (1) when equity shareholders are in control, and (ii) when

lenders assume control Their use of simulation analysis enables the authors to assess risks and potentials not captured by BBW scenarios and include: (1) approximating default likelihood, (ii) level of debt service reserve account, (111) assessment of capital structure, and (iv) how to best restructure in case of financial distress

° The correlation is inter-variable, where one occurrence affects another variable, (i.e increase in

Trang 28

The authors contend that BBW scenario analysis recognizes the uncertainty of the environment but does not specify (or specifies based on subjective data) the probabilities to such scenarios The authors contend that the predominant use of BBW analysis in project finance is due to: (i) the original notion that projects where sufficiently straight forward that a sophisticated analysis was unnecessary, and (ii) the cost factor involved for

utilizing more elaborate methods Furthermore, the BBW scenarios both the up side and

downside assume all assumptions simultaneously went well or all assumptions simultaneously went poorly respectively In the base-case, the assumption is that inputs are uncorrelated The characterization of the uncertainty of cash flows by only three scenarios is the principal drawback to BBW analysis, and leads to misinterpretation of the risk associated with the project Whereby, the true value of the cash flow can have an infinite number of distributions shapes and thus possible outcomes This variety of

distribution scenarios and outcomes is the basis for VaR models, which are common

practice in the commodities markets and derivative instruments

The authors contend that such scenario can be applied to interest rates and commodity prices, which capture the probability distribution Their examples demonstrate that simulation approach show the possible outcomes for particular stakeholders and provide an instrument for analyzing managerial reaction to uncertainties Their study is based on project that has an underling transferable commodity (1.e energy), and is limited to evaluating one project

unexpected repairs also increases the level of unavailability of the project)

Trang 29

This Thesis takes an evolutionary approach and builds on the many academic papers presented in this section The utilization of a PSA in valuing projects with non- homogenous outputs further develops the area of project finance risk analysis

1.4.2 Public Private Partnerships (PPP) Literature Review

The PPP literature overlaps with the project finance literature, whereby the

difference between a PPP structure and a traditional project finance structure is that the PPP structure envisions more risk sharing in the part of the government Therefore, the above project finance literature review provides the greatest insight to the advancement of the inherent risks involved in a portfolio of PPP projects Nevertheless, the specific government obligations set forth in the PPP enable this overlap to have particularities of its own There is little analysis on how to measure PPP risk in project finance structures and more importantly established methodologies that can cater to PPP style project finance structures From the late 1990s, there have been a number of publications reflecting on the PPP process in developed economies The Thesis contributes to the PPP literature providing a forward-looking methodology in evaluating government’s contingent obligations on a portfolio of projects under a PPP framework

The most important research regarding PPP, in which this Thesis research follows is Frank and Merna (2003) study of bundling a range of projects under the PPP program The authors describe the United Kingdom (UK) and Germany experience in the PPP for

the purposes of financial analysis Such bundling can be considered a portfolio of

projects, whereby some non-viable projects are, in essence, subsidized by the more lucrative projects The authors conclude that in order for non-viable projects to be

Trang 30

financed by the private sector such bundling approach is necessary This research takes the idea a step forward by introducing the bundling of project finance structure They verify the mechanism by comparing deterministic to stochastic economic outputs from a number of projects The authors combine the economic parameters and illustrate a BBW analysis The main contribution of the research is the idea of bundling various projects

together to make them more commercially attractive They examine four points in the

project’s financials, (i) cash, (ii) lock-up time®, (ili) payback period, and (iv) NPV, and simulate the three BBW scenarios The authors conclude that by combining projects in bundles they are able to alter the three parameters mentioned above and classify project

traditionally non-bankable to a bankable status within a portfolio

A shortcoming of the study is that the authors only used the four parameters

mentioned above, and the three traditional risk classifications related to BBW This

research intends to follow the same general ideas set forth in the paper and take a further step in analyzing via a VaR approach, whereby probability distributions is combined with the possibility of a government to pay-out contingent obligations under a PPP program Lastly, the Thesis also applies the government’s contingent obligations to a portfolio of

projects not necessarily concessioned to one sector as described by the authors, but to

multiple concessions in different sectors and exploring the covariance between sectors

1.4.3 Value at Risk (VaR) Methodology Literature Review

Trang 31

Although there has been extensive research regarding VaR in the banking sector,

such approach has not been widely applied to project finance structure, with the exception of project finance structures that have an underlying commodity as the Cabedo and

Moya’s (2003) paper

The reason for the difficulty in applying VaR methodology to project finance structures is their inherent complication and difficulty in standardization Each project’s finance model is distinct, depending on the underlying concession contract (if publicly concessioned project) or the demographics As such, for each project finance structure, one needs to identify the risks associated with such structure, assess probabilities to that risk and perform VaR models for each project at the time of the assessment Such intricacies have reduced the application of VaR methodology to the project finance arena Given the surge in project finance and in particular, the possibility of multiple projects being owned by one entity, the VaR approach is a common tool, which serves to efficiently inform management of the risks associated with each of the various projects in their portfolio

In the past, project finance structures were in limited numbers; as such, had little financial impact on the sponsors/shareholders From a research point of view, the difficulty lies in the heterogeneous nature of project finance structures, in which there has

been little application for such an effort until now, where the PPP trend is now taking

Trang 32

As mentioned above, the only papers utilizing VaR methodology in project

finance incorporates an underlying commodity In Cabedo and Moya’s (2003) paper, the

VaR approach was used for quantification based on oil prices, including the likelihood of

a particular price Such study is beneficial to design risk management strategies for that industry The authors calculated VaR on three methods: (i) Historical Simulation (HS); (ii) Historical Simulation with Autoregressive Moving Average (ARMA) forecasts; and ( the variance-covariance Method based on Autoregressive Conditional Heteroskedasticity (ARCH) model forecasts The HS, an empirical distribution, is derived for the change in price, and the VaR is calculated based on a statistic likelihood percentile The HS approach takes for granted that price change characteristics recur over time The ARMA utilizes the same base for the empirical distribution, and a Monte Carlo exercise is utilized to find the maximum loss associated with the statistic likelihood percentile

The authors conclude that the ARMA model does not obtain VaR directly from past returns, and does not require knowledge of the statistical distribution of past returns, nevertheless, the ARMA model was able to provide a better VaR estimation than HS Such paper is beneficial to the Thesis given its direct application of VaR for an oil project The Thesis explores VaR methodology for two projects, which such projects do

not require an underlying tradable commodity

The predominant literature in VaR application has been in the banking sector The Thesis follows such methodology, although it applies to project finance structures under the PPP framework The similarities in methodology with the Thesis can be found

Trang 33

in Barnhill and Maxwell’s (2002) paper, where they acknowledge the significance of an integrated risk assessment methodology through a forecast methodology (Forecast

Methodology) The authors develop a methodology for assessing VaR of a portfolio of

securities with correlated items such as: (i) interest rate; (ii) interest rate spread; (1i1) exchange rate; and (iv) credit risk This approach concurrently simulates the future financial environment (in which financial instruments are valued) and the credit rating of specific firms For each simulation, a different financial environment is created (based on correlated items mentioned above) as well as firm specific values In the case of the Thesis, similar approach are used where with each simulation, the project finance forecast methodology incorporates the up-dated financial environment, as well as the particulars in the project After running a number of simulations, a distribution of portfolio values of the project is generated Barnhill and Maxwell (2002) conclude that the standard deviation of the portfolios are less than the total of the market and credit risk combined, illustrating that market and credit risk are not additive, and provides value to utilizing

PSA in correlated variables

The Barnhill, Papapanagiotou, and Schumacher’s (2000) research, utilized the same methodology as in the Barnhill and Maxwell (2002) paper, and extend the Forecast Methodology for South African banks The authors apply the Forecast Methodology to various theoretical banks operating in the South Africa Their results suggest that combining credit risk and geographical concentration in the loan portfolios leads to bank portfolios with higher risk The authors further demonstrate that credit quality of the bank’s loan portfolios worsens by the level of geographic intensity In the same spirit, the

Trang 34

Thesis evaluates geographical concentration’s relation to risk, as such, it evaluates two transportation projects that affected by regional economics Another paper (i.e Barnhill and Kopitz (2004)), utilized a similar Forecast Methodology to extract further findings In their paper, the Forecast Methodology provides a forward-looking methodology for assessing bank risk levels A similar forecast methodology is used to evaluate PaR in project finance structures

l5 Theory for the Research

The Thesis research is based mainly on Option Pricing Theory (OPT) and Modern Portfolio Theory (MPT) MPT is based on Markowitz’s (1959) paper, where the author developed MPT, and is the predominant theory in assessing and measuring risk, (as defined by Markowitz, portfolio risk is the standard deviation of portfolio returns) Markowitz paper led the way to a consensus that risk can be reduced by diversification In OPT, Black and Scholes (1973) led the way to develop this theory, whereby a contingent claims approach is utilized to value debt and equity via a options

1.6 Application of the Methodology

The combination of VaR methodology, which has been developed tn numerous applications as mentioned in the preceding section, particularly in the Banking sector, enables the assessment of multiple projects and the correlation of regional economic indicators that affect such projects Such methodology advances the research in project finance by measuring PaR of government contingent obligations

In order to use a PSA in the PPP project scenario and to model such variables via a stochastic processes that encapsulates the regional economic variables’ dynamics, the

Trang 35

model requires the following components: (i) a set of economic variables (regional or national), which its variation affects the projects being studied; and (11) estimate

volatilities and correlations for such variables

Appropriately establishing the state variables that model the regional and national economic and financial environment, which affects PPP projects, allows for assessing the diverse factors of risk faced by government in the PPP infrastructure projects The government’s default under the PPP is based on a systemic tisk’ given the characteristics of the government contingent obligations under the PPP program In a subsequent section in the paper, a list of the state variables that are envisioned to be selected, including its source Such list of variables has its origin on previous studies mentioned in the previous section

1.7 Main Objectives of the Research

This dissertation accomplishes three objectives:

1 Assess, first on a project-by-project basis and later on a portfolio basis, the PaR via an integrated risk methodology

2 Assess PPP systemic risk

3 Assess the reliability of an integrated risk methodology in modeling PaR probabilities

7 Systemic risk can be measured as the probability that a number projects of the PPP require government

support (as specified in the concession contracts) at the same time and create, on a sequential basis, further defaults on other PPP projects

Trang 36

II Project Finance Structure Overview

Project finance is challenging to define, Milbank (2004) states the following interesting quote: “Most definitions [of project finance structures| focus on the financing of a specific facility in which lenders look principally to the revenues generated by the operation of the project for the source of funds from which loans are to be repaid The primary security for the loan consists of the assets of the project including, most importantly, the cash flow the project generates and the contracts that assure the stability

of both its costs and its revenues.”

Milbank continues the analysis indicating, “Project financings are used most commonly in the development of large infrastructure projects (e.g., power generation, toll

roads, telecommunications) and the exploitation of natural resources Although each

financing is specifically designed to meet the requirements of a particular project and the objectives of its sponsors, the following characteristics are common to most project financings: (i) the project 1s developed through a separate financial and legal entity; (11) the debt of the project company is often completely separate (at least for balance sheet purposes) from the sponsors’ direct obligations; (i1i) the sponsors seek to maximize the debt to equity leverage of the project and the amount of debt is linked directly to the cash flow potential, and to a lesser extent the liquidation value, of the project; (iv) the sponsors’ guarantees (if any) to lenders generally do not cover all the risks involved in the project; (v) project assets (including contracts with third parties) and revenues are often pledged as security for the lenders; and (vi) firm contractual commitments of a range of third parties (such as construction contractors, fuel and other suppliers, purchasers of the

Trang 37

project’s output and government authorities) represent significant components of the credit support for the project.”

Project finance structure gamed popularity as the benefits in Private Finance Initiatives (PFI) became more common and private sector more knowledgeable in the

exploring such opportunities PFI is the one form of public private partnership where

governments concession a construction (i.e construction of a government building or bridge) or a product (i.e the design and construction of fighter jets by Lockheed Martin or missiles) or a service (i.e the operations of a federal prison) The most notable early project finance structures include the Suez Canal and more recently the Euro tunnel The subsequent paragraphs highlight project finance arrangement that had a significant impact in the project finance field

2.1 Project Finance Stakeholders®

Project finance has a number of stakeholders with particular incentives the following section describes each potential player in the project finance arena.”

2.1.1 A Sponsors/Equity Owners

The principal objective of private sector sponsors and equity investors is to make profits Their tolerance for assuming risk may differ, and many equity investors may have other, perhaps more compelling, interests in the project, including supply or offtake contracts Most sponsors assess their investment in a project based on an IRR analysis of

the proposed investment, generally calculated on an after tax basis Equity investors

Section 2.1.1 through Section 2.1.8 was obtained as a direct quote from Milbank, 2004

Trang 38

generally invest through a project vehicle, which is most often a single purpose company, partnership, joint venture or trust, organized by its sponsors/equity owners with no assets other than the project The shareholding or other equity interests may be held by one person or, as is often the case in large international projects, a consortium Such consortia may include local participants, foreign operators, equipment suppliers, and other investors seeking sufficient returns to justify the risks being taken

2.1.1.1 Appeal of Project Financing

Infrastructure projects typically have significant funding requirements and entail risks, often in excess of what the sponsors may be willing or able to assume The project finance structure can be appealing to sponsors because it: (i) provides financing that is legally non-recourse to the sponsors; (ii) achieves off balance sheet accounting treatment of project debt by not showing any borrowing for the project among the sponsor’s own borrowings in its consolidated accounts; (iii) allows highly leveraged structures, which often means a reduction in the cost of capital by substituting lower-cost, tax deductible interest for higher cost, taxable returns on equity (some projects have been financed on a 100 percent debt basis, although a level of 70 percent to 90 percent is more typical); and (iv) allocates project risks among participants, thereby reducing each participant's risk of loss

2.1.1.2 Maximizing Return on Equity

Virtually all sponsors seek opportunities to obtain attractive rates of return on their investment To attract private foreign investment, host countries may need to afford

° Section A through H was a quote obtained from Milbank, 2004

Leandro F Alves The George Washington University, June, 2006

Trang 39

investors greater returns than are available internationally in other markets There is, of course, a natural limit to the returns available to investors in that the real cost of a project (manifested, for example, in monthly tolls that can be charged to the end user in case of a toll road) may be a large part of a local commuter’s basic cost of living One can easily imagine the political sensitivity in a developing country to increases in commuter tolls to satisfy the demands of foreign investors

2.1.1.3 Strategic Expansion

Sponsors, particularly utilities, may seek to expand into new markets at a time of

limited growth in demand in their domestic markets The global infrastructure market is entering a period of restructuring, with national utilities and international private developers expanding into new regions

2.1.1.4 The Sale of Goods or Services

Certain project participants, such as equipment manufacturers and fuel suppliers, have as one of their principal objectives securing contracts for the sale of equipment or supplies or for the operation of the facility Although these parties may be prepared to invest in private projects, a portion of their profits lies in securing the related supply contracts They may be willing, therefore, to take risks in relation to their equity investments (which form only part of their overall return on the project) that others would be reluctant to accept

2.1.2 Host Government

Many countries need a credible alternative to the traditional method of central government financing and operation of major infrastructure assets The reluctance of host

Trang 40

governments to increase domestic borrowing, together with the emerging political will to involve the private sector (including foreign investors), underlie the visible efforts to find ways to involve private initiative and private capital in the promotion of public interest

objectives (i.e the development of infrastructure) Once it decides to involve the private

sector in infrastructure development, the government has some of the following objectives:

2.1.2.1 Minimizing Costs

Private participation in infrastructure development can lower overall costs Effectively structured and transparent bidding procedures inevitably heighten competition among private-sector sponsors and suppliers

2.1.2.2 Risk Transfer

The objective of the government, generally stated, is to transfer the cost and risk of infrastructure development from the public sector to the private sector This objective involves: (i) Demanding a safe, efficiently run project The government (a) demands that the project be completed (to the government's specifications) as quickly as possible and (b) seeks adequate safeguards and assurances that the project operates properly and in line with the public’s interests; (i) Retaining control over the project Should the original sponsors fail to provide the required level of service or should the project run into insurmountable difficulties, the government has the ability to (a) regain control of the project or even, as a last resort, bring it back to public ownership or (b) offer the ownership or operation of the project to other private-sector entities These options may involve issues relating to expropriation, in which case the question of adequate

Định dạng
Số trang	250
Dung lượng	10,94 MB