Construction Risk in PPP Projects

Special Comment
August 2006
Contact
Toronto
Phone
1.416.214.1635
Andrew Kriegler Catherine Deluz Allan Mclean

London
Andrew Blease Neal Shah

Sydney
44.20.7772.5454
Nicola O'Brien David Howell Tim See

New York
61.2.9270.8100
Bart Oosterveld Daniel Gates Nicolas Weill
1.212.553.1653
Request for Comment on:
Construction Risk in Privately-Financed Public Infrastructure (PFI / PPP / P3) Projects

Summary
Moody's is seeking comment on a plan to augment its current approach to evaluating construction risk in privately financed, public infrastructure projects ("PFI/PPP") by introducing a quantitative model designed to improve the consistency and transparency of its analysis. The approach will also standardize the presentation of the qualitative factors considered in assigning ratings on these project financings. This Special Comment outlines Moody's proposed methodology and invites interested parties to comment. Please send comments to cpc@moodys.com with the heading Construction Risk by September 30, 2006. A subsequent Special Comment, to be issued shortly, will request comment on Moody's revised approach to rating the operating period risks of PFI / PPP projects. The quantitative model, which is an application of the approach underlying Moody's CDOROM software for synthetic collateralized debt obligations ("CDOs") will, for four standard project classes, explicitly incorporate Moody's assumptions about the likelihood of construction over-runs, the value of parent guarantees from construction contractors as well as the benefit of various types of external financial supports including performance bonds and letters of credit. The proposed approach models PFI / PPP projects on an expected loss basis. As a result, the analysis of project liquidity - ensuring that projects have sufficient cash reserves so that default probability is broadly consistent with the expected loss-based rating - is conducted in parallel with the quantitative modeling. The model results are an input to a rating framework that incorporates aspects of construction programs that are less suited to modeling but that are nonetheless critical to the analysis of PFI / PPP infrastructure financing. The methodology is not an exhaustive treatment of all factors reflected in Moody's ratings, but it should enable the reader to understand the key considerations and financial factors used by Moody's in the final rating determination. If implemented, the proposed methodology would be applied to the construction phase of qualifying new PFI / PPP projects globally. It would also be applied to those existing projects that have a significant portion of their construction tasks ahead of them. Some existing PFI / PPP projects may see their debt ratings changed as a result of implementation but a majority are expected to remain unaffected.
Consistent with existing practice, the proposed methodology separately considers risks in the construction and operations phases of PFI / PPP projects. As a result, should projects be structured with a significantly higher risk profile in construction than in operations, Moody's ratings on those projects, post-construction, will adjust to reflect a successful transition to project operations. Moody's will comment on its opinion of the relative risk of the construction and operations phases of new PFI / PPP projects as ratings are assigned. A beta test version of the construction simulation model will shortly be available to market participants at no cost on execution of a usage agreement. For further information on how to obtain the model, please contact Grant Headrick in Moodys Toronto office at grant.headrick@moodys.com.
Moodys Special Comment
Table of Contents
What is PFI / PPP? .......................................................................................................................... 4 About this Rating Methodology ....................................................................................................... 4
Methodology Overview ...........................................................................................................................5
Understanding Construction Risk in a PFI / PPP Concession ............................................................ 6

The PFI / PPP Structure...........................................................................................................................6 The Design-Build Agreement ..................................................................................................................7 Lenders' Exposure to Construction Risk ..................................................................................................7 Expressing Construction Risk as Credit Risk..........................................................................................10 What is CDOROM?............................................................................................................................10
Modeling Construction and the Risk Mitigation Package ............................................................... 11

Modeling Raw Construction Risk...........................................................................................................11 Construction Loss Severity Distributions ...........................................................................................11 Modeling Contractor Support ................................................................................................................13 The Value of Contractor Support .......................................................................................................13 Effects of a Contractor Default ..........................................................................................................14 Modeling Financial and Performance Supports......................................................................................15 Modeling the Value of Equity.................................................................................................................16 Correlations within the Construction Package .......................................................................................16
Qualitative Considerations ............................................................................................................ 17

Combining Qualitative Adjustments with the Model Results...................................................................17 When Qualitative Factors Dominate ......................................................................................................17 Assessing the Qualitative Factors..........................................................................................................19 The Construction Contractor ..............................................................................................................19 Complexity of the Project / Project Management Task........................................................................19 Contractual Framework .....................................................................................................................21 Related Party Effects .........................................................................................................................23
Liquidity ....................................................................................................................................... 25 Ratings after the Transition to Operations ..................................................................................... 26 Appendix A: Project Complexity Definitions ................................................................................... 27 Appendix B: Typical Performance Supports ................................................................................... 28 Appendix C: Technical Information about the Model ...................................................................... 29
What is PFI / PPP?

Used extensively to finance public infrastructure in the U.K., where they were launched in the mid-1990s, PFI / PPP structures are designed to shift certain financing, construction, and operating risks of public infrastructure projects to the private sector. Private sector consortia are engaged through a public bidding process, to design, build, and operate various types of public infrastructure projects under long-term concession agreements from a sponsoring government or one of its agencies. In addition to the U.K., where the structures are part of the government's Private Financing Initiative (PFI), the technique is being increasingly used in Australia and Canada, where the transactions are referred to as Public-Private Partnerships (PPP's or P3's) and they have also been introduced across continental Europe. PFI / PPP financing is not common in the United States, where most public infrastructure remains government owned, but there is increasing interest in examining a range of private financing options for public infrastructure, as evidenced by the proposed Port of Miami Tunnel PPP. Most commonly known for the construction and operation of public hospitals, the PFI / PPP financing approach has also been used to facilitate the construction and operation of government buildings, such as courthouses or other administrative centres, as well as to carry out a range of other public policy-mandated services such as the construction and operation of schools or mass transit terminals. PFI projects are usually distinguished from traditional government procurement arrangements by the fact that they feature fixed-price, fixed-term construction contracts and incorporate a requirement to operate the completed facilities pursuant to pre-agreed performance standards over a long term (25+ years) concession agreement. Performance failures, during either the construction or operational periods, can lead to abatement of the payment stream and, ultimately, to termination of the concession agreement. The risk of abatement or termination of the concession is usually offset by the experience of the contractor or operator and various levels of performance or surety support. In addition, contractual provisions usually exist that permit the project company to replace sub-standard contractors well before project termination itself becomes a possibility. Typically, the capital structure of the project company also incorporates a tranche of loss-absorbing equity to provide additional credit support.
About this Rating Methodology

This report explains how Moody's proposes to evaluate the credit quality of PFI / PPP projects during their construction periods. A subsequent methodology to be released shortly will speak to Moody's approach to rating the operating period risks of PFI / PPP projects. This methodology is intended to apply only to those projects that feature availability-based revenue streams. That is, projects that will earn their revenues if they ensure availability of the facility while meeting performance standards common in their industries. Public infrastructure financings with significant patronage risk (also known as volume or market risk) such as traditional toll roads and airports, share many characteristics with PFI / PPP transactions but are not included within the scope of this methodology. This methodology is also not intended to apply to infrastructure projects with highly complex operating requirements or significant post-construction, pre-operation commissioning risks, such as waste treatment plants or defense projects. However, the construction modeling aspects of this methodology may be extended to other classes of project finance which feature relatively homogenous construction requirements. Moody's may in the future also consider applying elements of this methodology to project financings with construction risk that are rated on Moody's U.S. Municipal Bond Rating Scale1. A mapping of Moody's U.S. Municipal Bond Rating Scale to Moody's Corporate Rating Scale is currently the subject of a separate request for comment2. Should the proposal outlined in that request for comment be implemented, it would allow the modeling results to be expressed on either the corporate or municipal rating scales.
1. The current approach to assessing construction risk for financings rated on the U.S. Municipal Bond Rating Scale is outlined in "Construction Risk: Mitigation Strategies for U.S. Public Finance", Moody's Special Comment, December 2004 2. "Request for Comment: Mapping of Moody's U.S. Municipal Bond Rating Scale to Moody's Corporate Rating Scale and Assignment of Corporate Equivalent Ratings to Municipal Obligations" Moody's Special Comment, June 2006
METHODOLOGY OVERVIEW
In this report we will describe each of the components of a construction project in terms of its default and recovery characteristics and its correlation with the other project elements. The expected loss - and therefore the credit risk - of the project is based on a Monte Carlo simulation of each of the project's components. To obtain a final rating, the result from a large number of simulations is overlaid with a series of qualitative factors that reflect each project's unique characteristics. This report is divided into three main sections as follows:
1. Construction Risk in a PFI / PPP Framework

This section provides background on the typical structure of a PFI / PPP project that includes a construction obligation. It describes the generic contractual structure governing the parties and how the design and construction responsibilities are assigned by the government / agency sponsor to the project company and ultimately passed on to the general and sub-contractors. It then discusses how lenders are exposed to construction risk via cost or schedule over-runs and the way in which those risks are limited through the operation of the 'cost to complete' tests that are typical to PFI / PPP concessions. We then summarize the various financial alternatives available to mitigate the risks during construction. As construction risk in a PFI / PPP project is usually considered as a physical risk with credit risk consequences rather than as a credit risk per se, we explain how we translate construction risk into the expected loss framework of credit risk used by Moody's ratings.
2. Modeling Construction Risk

This section describes how each aspect of a PFI / PPP construction project is modeled. It begins by explaining how the modeling of raw construction risk on an individual project is simplified by classifying each project into one of four basic types: Standard Buildings, Standard Civil Infrastructure, Complex Buildings and Complex Civil Infrastructure. It also presents Moody's assumed default and loss characteristics for each project type as well as the differences between the loss profiles of construction projects and corporate obligations. We also explain Moody's views on the ability and willingness of a construction contractor to support the performance of a project as well as the negative effects on a project's economics should a contractor default and have to be replaced. The modeling parameters for construction, the contractor and various financial and performance supports are then presented along with a brief discussion of how components of the project are correlated.
3. The Application of Qualitative Rating Factors

This section explains the qualitative factors that are overlaid on the model result to generate a final rating. Each of the factors, The Construction Contractor, Project Complexity, The Contractual Framework and Related Party Effects speak to elements of PFI / PPP projects that are either not suitable for modeling or which are modeled via the use of simplifying assumptions. The model result implicitly assumes that each project is "average" on each of the qualitative factors. To the degree that an individual project is materially stronger or weaker than the model-assumed average, the factor rating is notched up or down from the model result. Factor ratings are then combined to obtain a final rating. In this section we define the factors considered and provide examples of the ways in which projects may demonstrate that they are average, above average or below average for each. Finally, the methodology discusses how the analysis of liquidity applies to PFI / PPP construction ratings as well as the way in which a project that successfully transitions from construction to operations may expect to see its rating change over the project life.
Understanding Construction Risk in a PFI / PPP Concession

THE PFI / PPP STRUCTURE
The typical PFI / PPP transaction is governed by a master concession agreement ("Concession Agreement" or "CA") between the sponsoring government or agency (the "Off-taker") and a single purpose project company (the "Project Company" or "Concessionaire"). The Concession Agreement sets out the obligations of the Project Company to design, build and operate the project over the life of the concession. The CA defines the construction deliverables and timetable3. The construction obligation is usually defined on a fixed-price, fixed term basis although, depending on the level of design requirements finalized when the concession is awarded, some construction elements may not be completely specified. In addition to a target date for project completion and acceptance, the CA also specifies a "Sunset Date"4 - the date by which the construction and acceptance must be complete or the concession will be terminated. Should construction stretch into the period between the target and sunset dates - also known as the "construction tail" - the Concessionaire continues to be responsible for completing the project within the original budget. In addition, the Concessionaire may also be responsible for paying penalties or liquidated damages to the Off-taker. As operating funds from the Off-taker generally will not begin flowing until construction is complete and the project is accepted, the Concessionaire's capital5 is also the only source of funds for debt service for the whole of the construction period including the construction tail. The original project financing is either accreting during construction or it includes enough funding to provide for the payment of interest until the target construction date and at least part of the way to the sunset date. Should the project not be completed by the sunset date, the Off-taker can - and in some cases is obligated to - terminate the CA. Generally, the Off-taker is required to make a termination payment to the concessionaire based on either the NPV6 of the future income and expenses of the project - including the expenses of finishing the project - or the original construction contract amount less the remaining cost to complete the project to its agreed upon specifications. The cost to complete is usually defined as including not only labour and materials but also the Off-taker's general costs of remediation of any improperly completed work, contract re-tendering, etc. The project company's lenders are thus exposed to a loss if there is a termination based on a failure to complete the project on time, on budget and on specification. Losses result from the project's exposure to the cost to complete and the costs of carrying the project from the target date to the sunset date - and ultimately to the date that the Offtaker pays the termination payment 7. The Project Company attempts to ensure that lenders will be repaid by passing on the construction risks in the CA to a construction contractor and/or by requiring the contractor to provide additional security in support of its obligations.
3.
4. 5. 6. 7.
The concession agreement and its schedules also define all of the operational parameters of the project including the responsibilities of the concessionaire, performance standards, hand-back requirements, etc. Discussion of Moody's proposed approach to rating PFI / PPP projects during the operating period will be the subject of a forthcoming request for comment. Also known as the "long-stop date". As well as any external credit supports that have been put in place for such a purpose. Either directly, via a formula payout, or indirectly through a market tender mechanism. Other termination scenarios in PFI / PPP transactions are possible. As an example, the government usually reserves the right to terminate the project for convenience. These scenarios are not covered in this report as they usually provide for creditors - or even creditors and equity investors - to be paid out.
THE DESIGN-BUILD AGREEMENT

The Project Company enters into a fixed price, fixed term design-build agreement (the "DBA") with one or more construction contractors8. The DBA provisions generally mirror the obligations of the Project Company to the Contractor but typically feature even more stringent performance thresholds.
Typical PFI/PPP Transaction Structure

Shareholders Equity Lenders Debt
Concession Off-taker Agreement
Project Company
Design - Build Agreement
Construction General Contractor
Operating
Sub Contracts
Construction
Sub Contracts
For example, if the sunset date for construction completion and project acceptance in the CA is 36 months, the sunset date in the DBA may be 3 - 6 months shorter. Similarly, if under the terms of the CA,the Project Company is responsible for the payment of liquidated damages to the Off-taker in a delay scenario, the DBA typically requires that the Contractor be responsible for those payments in addition to any senior debt-service payments that are not otherwise provided for. The difference in performance thresholds is designed to give the Project Company the time to react to construction performance failures and, ultimately, to replace the Contractor and finish the Project with a substitute before the Off-taker can terminate the overall Concession Agreement.
LENDERS' EXPOSURE TO CONSTRUCTION RISK

PFI / PPP creditors are rarely willing to accept the raw construction risk that the project doesn't get built on time and on budget. So, the Project Company will - in addition to structuring the DBA on generally more restrictive terms than the obligations in the CA - either require the Contractor provide a construction risk mitigation package or arrange one itself and pass some of the costs on to the Contractor. In some jurisdictions, regulations also specify minimum levels of external support that must be provided in order to be eligible to bid on government contracts. The design of a construction risk mitigation package always begins with consideration of the stand-alone risk of construction in a given project. It then incorporates any of several possible financial and performance supports. The supports are designed to shield the Project Company and ultimately lenders from cost or schedule over-runs on the project and are also intended to raise the credit quality of the project debt during construction.
8.
The company contracted to do the construction is typically a subsidiary of one or more construction companies. The subsidiary is often an operating entity in its own right (dedicated, for example to construction in certain region). Multiple contractors may also join together in a (usually) unincorporated joint-venture.
Typical PFI/PPP Design-Build Structure
Bank
Construction Company Ultimate Parent
Letter of Credit Full/Partial Guarantee of Performance
Project Company
DBA
Construction Company
Performance/ Adjudication Bond
Insurance Co./Bank
1. Raw Construction Risk
Expressed in credit risk terms, raw construction risk is the expected loss on the stand-alone project. It can be analogized as the potential exposure of the sponsoring government or agency to cost or schedule over-runs for an average project of its type should the project have been let on a cost-plus basis to a contractor of average experience and ability9. As a result, this expected loss does not take into account the willingness or the ability of contractors or other parties to absorb the consequences of cost or schedule over-runs. The level of raw construction risk in a given project is first and foremost a function of the project's complexity. As is discussed below, Moody's has grouped the majority of projects seen in PFI / PPP transactions into four project types for ease of analysis. The four are Standard Buildings, Standard Civil Infrastructure, Complex Buildings and Complex Civil Infrastructure. Recognizing that it is difficult to capture any particular project's complexity given only four categories, Moody's also considers a range of qualitative factors that can influence the amount of raw construction risk. Examples of these factors include the experience of the contractor with the particular type of project being constructed, the degree of reliance on new and/or unproven technologies and the degree to which local economic conditions around the project site will affect the cost and availability of labour and materials over the construction period.
2. Contractor Support
The DBA governing construction in PFI/PPP transactions is typically structured on a fixed-price, fixed term basis. The Contractor is obliged to complete the project, absorb the costs of over-runs and often pay debt service otherwise expected to be covered from operating income as well as any liquidated damages or penalties assessed against the Project Company. While in many cases the obligation is open-ended, the Contractor can often cap its liability at some percentage of the contract amount - particularly the liability to pay liquidated damages to the sponsoring government / agency or Project Company.
9.
The "cost plus" analogy is clearly an imperfect one as cost-plus contracts are also generally at risk of not featuring the same degree of oversight, cost control, etc., by contractors as when the contracting firms are responsible for overages.
Moody's evaluation of the value of contractor support to a transaction incorporates an opinion of both the ability of the contractor to pay as required and its willingness to do so - rather than walk away and face reputation or legal challenges. Importantly, Moody's also considers the impact of the scenario where the Contractor, despite its best intentions, is unable to complete its responsibilities and needs to be replaced. In those circumstances, the project may suffer not only from the time lost while a new contractor is located but also from the likelihood that substitute contractors may charge a premium price to assume a project mid-contract.
3. Performance / Insurance-based Supports

Performance supports are designed to guard against the default by a Contractor or one of its sub-contractors on their construction obligations. Common types of performance supports include performance bonds, adjudication bonds, sub-contractor insurance and completion insurance10. These performance supports are distinguished from financial supports (e.g. bank guarantees) - referred to below in section 4. Usually written by highly rated financial institutions, most performance supports generally have the character of insurance policies written by multi-line insurance companies. In other words, in order to receive compensation from the supporting institution following a default on performance, the beneficiary is required to submit a claim for adjudication, prove that the default is covered under the policy and substantiate the amount claimed for. Performance supports are rarely settled on a timely basis and often settle for less than the face amount of the claim. As a result, although performance supports can improve the credit quality of a transaction, Moody's considers them to be of only limited liquidity value.
4. Liquid / Financial Supports

The most common examples of financial supports are bank letters of credit, bank guarantees and demand deposits at regulated financial institutions. To be considered a financial rather than a performance support, the instrument must be able to be drawn upon on a timely basis and there must not be any conditions precedent to the bank honouring the drawdown either in the instrument itself or in any of the other project documents. In addition, while drawing on a support usually gives the support provider a claim on the project for the amount advanced, categorization as a financial support requires that the support provider's claim be properly subordinated to rated project debt. As a result of the unconditional and irrevocable nature of the support, Moody's focuses on the ability of the support provider to make good on its obligation - as represented by its rating - as the key attribute of a financial support.
5. Equity
The capital structure of the Project Company usually features a tranche of equity ranging from 10% - 20% of the total capitalization which is provided by the project sponsors. Equity provides funding for the project budget and, in the event of a termination, may provide credit support to the project debt by absorbing losses. The elements of construction risk packages can differ significantly across project types and by jurisdiction. Generally, transactions that are wrapped by financial guaranty insurers before funding feature only contractor support and relatively low levels of performance and financial support while those placed directly into the debt markets usually include a high level of financial support (along with contractor and performance support)11. As a termination payment generally provides for the Project Company to receive the contract amount less the cost to complete, the loss severity on a project - after considering the benefits of the construction risk mitigation package and loss absorption by equity - equals the loss on the senior debt.
10. The most common types of performance supports are defined in Appendix B. 11. For example, most PFI / PPP transactions in the UK feature only a modest level (10% - 20% of the construction contract amount) of performance support in the form of adjudication bonds while many Australian transactions have featured full (100% of the debt amount) financial support via bank letters of credit.
EXPRESSING CONSTRUCTION RISK AS CREDIT RISK

Moody's long term ratings are primarily opinions about expected loss: they reflect both the probability of default on an instrument and the likely level of financial loss suffered in the event of default. To assess expected loss on a construction project, raw construction risk - which is associated with project complexity and the likelihood of budget over-runs - must be translated into credit risk. This enables Moody's to express the lender's net exposure to the construction package in terms of a Moody's rating equivalent. To accomplish this, we use the following general analogies: Moody's definition of a default on a corporate bond holds that a default occurs if a scheduled interest or principal payment is missed, the borrower seeks judicial or regulatory relief from the obligation to pay creditors or if the holders of the instruments in question agree to a distressed exchange of their holdings12. Our definition of default in a stand-alone construction project is analogous: the project goes over budget or over schedule. Similarly, the construction analogue for the loss severity of a defaulted financial instrument - the percentage of par not recovered by investors - is the amount of the budget overage plus the monetary consequence of a schedule delay, in each case measured as a percentage of the Project Amount13. As it is impossible to determine a single loss severity in advance, the generic term "loss severity" refers to the probability distribution of various levels of loss rather than a single point assumption.
Table 1: Mapping Construction Risk to Financial Risk

Financial Instrument Probability of an instrument going into default (PD) Construction Project Probability of the project going over time / over budget
Loss suffered by investors as a percentage of par (LS) Amount of budget overage / cost of schedule overage1 Expected Loss = PD*LS
1. Implemented as a probability distribution of different losses / budget and cost over-runs. See Modeling Raw Construction Risk.
Once raw construction risk is translated into its credit risk equivalent, the construction risk package can be expressed as a portfolio of financial assets supporting a financial liability. The asset portfolio is made up of the construction obligation, the contactor's support and any performance or financial supports, while the senior liability to be rated is generally the amount of senior debt to be issued. While it might seem unusual to characterize a construction obligation as a financial asset, one can consider it as a bond that, rather than paying par at maturity, "pays" the project the right to earn the operating period cash-flows. If the construction obligation "defaults", the loss to the project is the amount that the termination payment from the Off-taker will not cover - the cost-to-complete - plus the cost of carry until the termination payment is received. Recognizing that the construction risk package on most PFI / PPP transactions can be viewed as an analogue of a portfolio of financial assets supporting a financial liability - a CDO - Moody's has adapted its widely used CDOROM software, originally designed to assess the credit quality of synthetic CDOs to allow it to be used to assess the credit quality of a construction package.
CDOROM
CDOROM is a Monte Carlo-based simulation model used by Moody's analysts to rate synthetic CDOs. It calculates the expected loss (i.e. the credit risk) for a portfolio of financial assets based on each asset's default probability and recovery rate distribution as well as how each asset's performance is correlated with the others in the portfolio.
How does the model work?

A Monte Carlo simulation calculates the credit outcome for the assets in the portfolio for a large number of possibilities. It is not a cash-flow model of the portfolio. Rather, each possible outcome (a path) sees a certain combination of asset defaults and recoveries. The expected loss on the asset portfolio is the weighted average of the path outcomes. The default and recovery rate distributions chosen by the user, along with the pair-wise correlations between assets, define how likely certain paths are to be taken. Together with a user-specified liability structure, which defines the priority of payment between the debt tranches that financed the project, the simulation outcomes provide an expected loss and therefore a rating for each tranche of debt.
12. The analogous events for structured securities - which may contemplate that interest or principal due in a given period may be deferred to later periods - are more complex. The focus for such securities is on uncured payment defaults or downgrades to Ca or C in anticipation of expected future payment losses, collectively known as Material Impairments. See "Default and Recovery Rates of Corporate Bond Issuers, 1920-2005", Moody's Special Comment, January 2006, and "Default and Loss Rates of Structured Finance Securities: 1993-2005", Moody's Special Comment, April 2006 for details. 13. The Project Amount is defined by reference to the contractual structure of the concession and, particularly, the termination regime that would apply in the event of a construction default causing concession termination.
10
Modeling Construction and the Risk Mitigation Package

To simulate the performance of a PFI / PPP construction transaction, we must determine the default probabilities and loss distributions for the project and translate the value of contractor and performance supports into their financial support equivalents. Table 2 sets out the major components of typical PFI / PPP construction risk packages, their character and the requirements necessary to model them.
Table 2: Modeling Requirements for PFI / PPP Construction Risk Packages

Component Raw construction risk Character Physical project complexity Requirements for Modeling Determinination of default probabilities and probability distributions for cost over-runs by project. Determinination of ability and willingness of contractor to stand behind project. Impact on the project of a contractor default. Performance support Multi-line insurance contract Determinination of relative value and timeliness v. an equivalent face value amount of a financial support. Bank rating. Timing and certainty of cash receipts. Bank rating.
Contractor support
Commercial contract
Financial Support Cash holdbacks, other
Bank obligation Cash deposits or obligations supported by LC's
MODELING RAW CONSTRUCTION RISK

Rank ordering all PFI / PPP projects in terms of construction complexity and then determining individual default probabilities and loss distributions is not practical - nor are minor distinctions in complexity between similar types of projects likely to be material in the context of the credit rating result. As a result, Moody's has simplified the analysis by grouping similar types of projects together for the purposes of modeling. The four types are: Standard Buildings; Standard Civil Infrastructure; Complex Buildings; and Complex Civil Infrastructure. The types are based on a classification system defined in a study of the UK public procurement system conducted for the UK Treasury in 200214. In addition to grouping projects by type, Moody's has also simplified the analysis by assuming the default probability for the raw construction project is 1. In other words, Moody's assumes that no project will come in under budget. It could be argued that ignoring the possibility of projects coming in under budget is a harsh assumption since many do. However because the project doesn't generally get to "keep" any of the savings realized in such circumstances, such an outcome is equivalent to a zero loss scenario from the perspective of the project's creditors15.
CONSTRUCTION LOSS SEVERITY DISTRIBUTIONS

Loss severity distributions on construction projects are much more tightly bounded than the distribution of losses given default for (non-financial) corporations. This is primarily because the ability of corporations to make decisions to change strategy, to enter or exit businesses or to modify their capital structures is far broader than in the project finance sector generally or PFI / PPP specifically.
14. See Appendix A, Project Complexity Definitions. For Mott MacDonald's definitions, see Review of Large Public Procurement in the UK, Mott MacDonald (2002), available at www.hm-treasury.gov.uk/greenbook 15. Such savings generally are for the benefit of the contractor.
11
Moody's review of ultimate recoveries from a sample of approximately 400 US bankruptcies and distressed exchanges of non-financial corporations suggests that historical variation in firm-wide average recovery rates is consistent with a beta distribution for enterprise value bounded between 0% and 120% of liabilities that has a mean of 52% and a standard deviation of 28%16. The chart below presents the same results from the creditors' perspective: the distribution of losses on enterprises' liabilities given default17 .
Loss Given Default Non-Financial Speculative Grade Corporations

2.00%
1.50% Probabilty
1.00%
0.50%
0.00% -20% 0% 20% 40% 60% Loss (%) 80% 100% 120%
In contrast to the relative freedom of corporations in unregulated industries to operate, the structure of construction contracts generally and the structure of design-build agreements in PFI / PPP transactions in particular, tend to limit the amount of cost and schedule over-runs to far below the levels experienced by unsecured creditors of defaulted corporations. This is because: The construction obligation is well defined in advance, has been explicitly priced by the Contractor and generally cannot be altered. The cost of changes to the project requirements after the design has been agreed ("variations") are generally the responsibility of the Off-taker; In most transactions, the Project Company engages a technical advisor that reviews the scope of the construction, the budget sufficiency and the feasibility of the schedule. The Off-taker generally has also done a capability review of the budget, schedule, etc., before awarding the contract to a particular concessionaire; During construction, equity investors in and/or lenders to the project company engage an independent engineer18 to monitor the progress of project and authorize payments to the Contractor. The independent engineer's responsibility is to review completed work, to satisfy itself that the works meets specifications and confirm that the project is on budget and on schedule. The independent engineer is generally not permitted to release funds to the Contractor unless the amounts remaining in the construction account are sufficient to complete the project after considering the work completed to date; and The overwhelming majority of projects subject to PFI / PPP transactions involve design and construction techniques that are well understood and have been repeatedly applied elsewhere. Moody's has determined stressed case loss severity distributions for each project class based on: (i) our existing PFI / PPP ratings, (ii) an internal survey of Moody's project finance analysts globally, and (iii) discussions with major contractors in various jurisdictions.
16. Setting the endpoint of the enterprise value / liability ratio at 120% allows for the (generally low probability) outcome that firm value in default will exceed firm liabilities in order to capture those circumstances in which debt recoveries are so strong that the firm's preferred and common stockholders emerge from bankruptcy with some recoveries. 17. Where Recovery Rate = 1- Loss Given Default. Presented in this form, the mean of the distribution is 48%. Note that since creditors can generally not recover more than they are owed, the LGD distribution is truncated at 0. 18. Also known as an "Independent Certifier".
12
Presented in comparison with the loss-given-default distribution for non-financial corporations cited above, the results for PFI / PPP construction projects are consistent with beta distributions with the following characteristics:
a
Table 3: Project LGD Means and Standard Deviations by Type1

Project Type Non-Financial Corporate Standard Building Complex Building Standard Civil Complex Civil
1. Rounded to whole numbers.
Mean Loss 48% 8% 15% 11% 18%
Loss Std Deviation 28% 8% 12% 12% 14%
Loss Given Default Non-Financial Corporations & Construction Projects

16.00% 14.00% 12.00% Probability 10.00% 8.00% 6.00% 4.00% 2.00% 0.00% 0.00% 10.00% 20.00% 30.00% 40.00% 50.00% 60.00% 70.00% 80.00% 90.00% 100.00% Loss (%) Standard Building Complex Building Standard Civil Complex Civil
Standard Corporate
MODELING CONTRACTOR SUPPORT

The relationship between a contractor and a PFI / PPP project is more complex than that seen in a typical financial services transaction. To begin with, the Contractor is very closely linked to the project - indeed the Contractor is often an equity participant in the project itself. Yet, PFI / PPP transactions are explicitly designed to enable a contractor may be replaced if it fails to meet the performance specifications set out in the DBA. As well, by assuming the obligation to build the project on a fixed price, fixed term basis, the Contractor provides potential credit support to the transaction. For example, the contractor is obligated to absorb excess labour or material costs within contractual limits. However, this support differs significantly from a typical external credit support like a third party guarantee or an insurance policy because the failure of a Contractor does not just simply remove the support itself, it hurts the primary obligation by increasing the cost of completing the construction contract.
THE VALUE OF CONTRACTOR SUPPORT

In most PFI / PPP transactions, the legal entity responsible for construction is an operating subsidiary of the Contractor's ultimate parent company. The financial strength behind the Contractor's commitment can come from the operating subsidiary or from a full or partial guarantee of the subsidiary's obligation by its ultimate parent. However, obligations under a construction contract are not debt obligations. As a result, while traditional measures of credit quality - such as a rating - capture both an organization's ability and willingness to pay its debts, that same level of willingness may not be applicable to a non-financial obligation. In other words, when faced with a sig-
13
nificant cost over-run, a contractor may be more likely choose to walk away from the contract - and face the subsequent legal and reputation consequences - than to walk away from a comparable bond or bank debt. It is also important to consider that in the context of a complex design-build agreement, legitimate disagreements may arise as to the cause of cost or schedule over-runs and therefore whether the responsibility to pay for them falls to the Contractor, the Project Company or the Off-taker. Moody's captures differences between debt obligations and construction obligations by applying a haircut to the value of the support provided by the Contractor and using that reduced amount as the input into the modeling. Haircuts are based on a review of existing rated PFI / PPP transactions and an internal survey of Moody's project finance analysts.
Table 4: Mapping Construction Contractor Support to Model Inputs

Notional amount Amount of the construction contract obligation guaranteed by a substantial contractor operating company or the contractor's ultimate parent 66% of the first 50% of the guaranteed construction obligation PLUS 33% of the remaining guaranteed construction obligation
Value after haircut
Moody's credit rating is used as the model input for the contractor's credit quality. Moody's will typically generate an internal credit estimate for unrated contractors to serve as the input to the construction model. When a joint venture (J-V) rather than a single company is acting as the Contractor, Moody's will follow the structure of the J-V agreement when modeling contractor support and will take into account whether the obligations are several or joint and several and whether they are capped at a particular level.
EFFECTS OF A CONTRACTOR DEFAULT

Replacing a defaulting Contractor takes time. Thus, in the context of a fixed price, fixed term DBA, any delay moving a project forward also has a monetary cost. In quantitative terms, the impact of a Contractor default can be described as moving the loss distribution for the project to the right. Adding to the problem is the potential difficulty of finding a qualified replacement contractor that is willing to take on the responsibilities of the defaulted Contractor. Not only would the replacement contractor need to familiarize itself with the project, it would have to either step into existing sub-trade contracts or go to the expense of replacing them. In a worst case, some already completed work might even need to be re-done. For example, if the contractor on a $100 construction project defaults, the project - which had, as all projects do, an intrinsic risk of coming in over-budget - still retains its original risk of coming in over-budget, but it must also "pay" both for the additional cost of finding a replacement contractor and for whatever premium over the original contract price is demanded by the replacement. For a standard building, this cost - measured in both out of pocket costs and in terms of the cost associated with the delay - is assumed to total an additional $12.
Loss Given Default Std Building before / after 12% Contractor Default Effect
14.00% 12.00% 10.00% Probability 8.00% 6.00% 4.00% 2.00% 0.00% 0.00%
12%
10.00%
20.00% Standard Building
30.00% 40.00% 50.00% Loss (%) Std Bldg w Contractor Default /
60.00%
14
Further, the more complex the project, the higher the likely costs associated with replacing a defaulted Contractor. Moody's will thus apply higher costs to project loss severity for complex projects than for simple projects. As is the case with the loss distributions themselves, Moody's determined the contractor default effects through a review of its existing project finance ratings and by surveying our project finance analysts.
Table 5: Project Mean Losses w/ Contractor Defaults

Project Type Standard Building Complex Building Standard Civil Complex Civil Mean Loss 8% 15% 11% 18% Effect of Contractor Default +12% +20% +12% +20% Mean Loss w/ Contractor Default 20% 35% 23% 38%
In practice, the replacement cost is likely to vary depending on when in the project the Contractor defaults. For example, if a Contractor default occurs early in construction, a replacement contractor would have to assume a large construction obligation based on another contractor's contract and design - which suggests that the replacement premium may be higher than the level above. In contrast, late stage defaults may have lower replacement costs relative to the original total budget, as the absolute construction amounts that remain to be spent are relatively low. Given the questionable increase in precision from modeling a time-variable default effect, however, Moody's employs a fixed Contractor default effect as a reasonable simplifying assumption.
MODELING FINANCIAL AND PERFORMANCE SUPPORTS

The value of financial and performance supports to a construction project are a function of: the credit quality of the issuing institution; the structure of the support; and the conditions precedent, if any, to their use. To be considered a financial rather than a performance support, the instrument must generally be issued by a regulated, deposit-taking institution, be available to be drawn on a timely basis and there must not be any conditions precedent to the drawdown, either in the instrument itself or in any of the other contracts governing the project. As a result of their unconditional and irrevocable nature and - importantly - the consequences to the issuer of not honouring the instrument on a timely basis, financial supports are modeled at face value using their Moody's rating as the indicator of credit quality. Bank letters of credit, bank guarantees and demand deposits are typically the only obligations treated as financial supports19. All other supports are treated as performance supports. This includes instruments drafted identically to a bank letter of credit but issued by a multi-line insurance company. This treatment reflects the nature of insurance contracts as compared to bank letters of credit. Specifically, disputing payment under an insurance contract has limited capital markets consequences for an insurer while failing to pay a letter of credit on demand would generally be regarded as an event of default for a bank. For similar reasons, a bank-issued adjudication bond issued in support of a PFI / PPP transaction would also be treated as a performance support20 . Based on internal estimates, the credit value of a performance support is modeled using its Moody's rating and 50% of its face value. Moody's considers performance supports as providing only limited liquidity value to a PFI / PPP transaction21.
Table 6: Modeling Inputs - Financial and Performance Supports

Support Type Financial Support Performance Support Credit Quality Bank Deposit Rating Bank Deposit Rating or Insurance Financial Strength Rating Value Face Value 50% of Face Value
19. Financial Guaranty insurance policies from monoline insurers would generally also qualify as financial supports. 20. In rare circumstances, Moody's has evaluated financial obligations from multi-line insurance companies in support of certain structured finance transactions as being of equivalent timeliness to those provided by monoline financial guarantors. See "Analyzing the Role of Multi-Line Insurance Companies as Primary Obligor in Structured Transactions", Moody's Special Comment, January 2004. 21. See LIQUIDITY.
15
Generally, Moody's will explicitly model only top-level performance supports - i.e. instruments that support the project as a whole and that name the project company and/or lenders as beneficiaries. In many cases, a contractor will also require that its sub-contractors provide their own performance supports - for the benefit of the contractor - or the contractor will take out its own insurance against sub-contractor default. Secondary levels of performance support such as these can be valuable to the project and are considered as part of the qualitative assessment of each transaction.
MODELING THE VALUE OF EQUITY

Typically, PFI / PPP transactions are highly levered with no more than 10% - 20% of the Project Company's capitalization made up of equity provided by the project sponsors. Equity is either contributed up front at financial close or, more commonly, is contributed near the end of construction in the form of funding to complete the project. If the equity isn't contributed in cash, it is typically supported by an unconditional, irrevocable letter of credit from a highly rated financial institution. The value of equity is indirectly captured in the model. The raw construction "asset" reflects the size of the gross construction obligation while the "liability" being rated is set at the lesser of the construction obligation or the senior debt amount. As a result, the model reflects the extent to which the equity commitment is available for loss absorption on the construction project rather than for the payment of other transaction costs. Project sponsors often seek to provide a portion of their project contribution in the form of deeply subordinated debt rather than as common equity of the Project Company. The ability to structure an equity-like contribution as subordinated debt will vary by jurisdiction. Generally, Moody's will consider some level of deeply subordinated debt equity-like as long is it is permanently subordinated and postponed during construction22 and it confers no rights of action on the subordinated lender - including acceleration or cross default.
CORRELATIONS WITHIN THE CONSTRUCTION PACKAGE

Modeling construction risk and a construction risk mitigation package as the equivalent of a portfolio of financial assets requires not only that each of the "assets" be properly characterized but also that the relationship between the assets is captured. In modeling portfolios of financial instruments in CDOROM, Moody's assumptions about correlations are based primarily on two sets of data: Moody's corporate rating transitions study and Moody's KMV's calculated asset correlations. These assumptions capture regional correlations in conjunction with a designation of each industry as local, semi-local, or global. The financial elements of the construction mitigation package are assigned correlations as if they were financial obligations of the organizations that issued them. For example, contractor support, performance bonds and letters of credit are treated for correlation purposes as if they were bonds issued by the contractor, the multi-line insurance company and the LC bank, respectively. Similarly, the construction project is characterized - for the purposes of determining its correlation with the other assets in the portfolio - as if it were the debt obligation of a speculative grade company in the construction industry. In other words, the correlation between the project and its contractor is the same as would normally be assumed when modeling the debt obligations of two different contractors operating in the same market. This gives the project an 18% correlation with investment grade contractors or a 36% correlation with a speculative grade contractor23.
Table 7: Modeling Inputs - Contractor / Project Correlations

Project Construction Project 100% S. Grade Contractor 36% I. Grade Contractor 18%
22. In operations, distributions on sub-debt are permitted, subject to the Project meeting the same financial thresholds that would have permitted distributions on the equity that the sub-debt replaces. 23. More detailed information about the asset correlation structure in CDOROM is provided in its User Guide, available on www.moodys.com.
16
Qualitive Considerations
The simulation model provides a consistent measure of credit risk within and between projects as well as a transparent assessment of the value of various types of external credit supports. However, quantitative models are limited by the range and complexity of the inputs they are designed to accept. Moody's ratings on debt instruments issued to fund PFI / PPP construction necessarily consider a broader range of factors than can be completely modeled. As a result, the model results are the input into a framework that considers four qualitative rating factors: the Construction Contractor; Project Complexity; Contract Framework; and Related Party Effects The application of these rating factors enables us consider whether the simplifying assumptions that are inherent in the use of the model are appropriate for a given transaction or there is a need to adjust the model results upward or downward. These qualitative factors may not be applicable if creditors are fully insulated from a project's construction risk - as is common, for example, in Australia through the use of bank letters of credit covering 100% of the rated debt during the construction period.
COMBINING QUALITATIVE ADJUSTMENTS WITH THE MODEL RESULTS

The PFI / PPP construction approach assumes, for the purposes of modeling, that the project meets a certain expected norm for each of the four factors above. To the degree that a particular PFI / PPP construction project is materially above or below the expected level it will score a higher or lower rating than the model result would otherwise indicate. The qualitative factors are then combined to generate a final rating for the project 24. The usual outcome for each of the attributes that make up a factor evaluation is for the project to receive one of three grades: Above Average, Average and Below Average. A fourth possible outcome, Not Suitable for Modeling is discussed further below. The outcomes are then added together to form the final rating. Generally, the combination results are symmetric (an Above Average result on one characteristic will offset a Below Average on another), reflecting the offsetting influences that the different factors can have on a transaction. However, as projects can score above or below the norm in more than one attribute per factor and as the marginal credit effect of additional positive or negative attributes diminishes, it is unlikely that qualitative factors will move the final rating more than two notches up or down from the model result.
WHEN QUALITATIVE FACTORS DOMINATE

When consideration of the qualitative factors applicable to a particular transaction lead Moody's to conclude that the model is a not suitable basis for rating the project, Moody's will apply its traditional project finance methodologies to rate the transaction debt, but will use the model result to inform the rating committee discussion. Some examples of cases in which where a model-based approach would likely not be appropriate are projects in which: 1. the construction task is not comparable to or is more complex than the four standard project types; 2. the technology underlying the project's operations has not been previously used in a similar application and/or in similar environments; or 3. the project is not situated in an economy with a track record of private sector financing for public infrastructure.
ASSESSING THE QUALITATIVE FACTORS

This section explains some of the most common qualitative characteristics that may influence a project's rating and how we factor these into our final rating. Given the complexity of PFI / PPP structures and the variations in the way in which they are implemented from jurisdiction to jurisdiction however, the listing is not exhaustive. To the extent that a project's final rating reflects qualitative adjustments of the type listed below, Moody's will describe the expected characteristic for a given factor, the way in which the project deviated from the expected level and the way in which all of the qualitative factors collectively affected the final rating on the project.
24. This approach differs slightly from that generally used in Moody's rating methodologies for non-financial industries. Typical corporate methodologies map the scores of issuers on various factors - including quantitative measures - to ratings and then weight the factor results to obtain an overall rating.
17
THE CONSTRUCTION CONTRACTOR

The simulation model captures Moody's view of the contractor's credit quality and the likelihood that it will fulfill its obligation to build the project at a fixed cost within a fixed time. In addition to credit quality, Moody's also considers the circumstances in which a particular contractor is more or less likely than the average contractor to be called upon to fulfill its credit support obligation. To make this assessment, we consider a contractor's track record of completing similar projects. We also take into account whether the contractor is tied to the success of the project - beyond its fee - by virtue of an equity investment in the project itself, the degree to which the contractor has shielded itself from sub-contractor performance failures and contractor's size relative both to project and to its other commitments. Thus, if a contractor has a particularly strong track record with a project type or if it has insulated itself - and therefore the project - from the risk of subcontractor underperformance, the credit quality of the project may be higher than otherwise indicated by the model Relevant Experience / Track Record As part of the contractor assessment, Moody's determines whether the contractor is experienced with the type of project at hand. Our standard assumes a track record of largely on-time, on-budget projects completed over several years. A contractor with an exceptionally strong record - in complex projects - can improve the rating. Projects relying on contractors with limited or no experience with the project type are likely to see a factor rating lower than the model result. In evaluating the contractor's track record, Moody's will consider its project history and examine the criteria used by the design-build consortium in selecting the contractor.
Table 9: Contractor Track Record

Standard Projects Above Average N/A Complex Projects DBA consortium is composed of recognized industry leaders in the design / construction of projects in this class and jurisdiction Multi-year track record of successful completion of similar projects Limited experience with similar projects in jurisdictions with similar legal and regulatory frameworks. Project is very large relative to the size of the Contractor. Project complexity / scope beyond the standard types
Average
Multi-year track record of successful completion of similar projects No track record with similar projects.
Below Average
Project is very large relative to the size of the Contractor. Not Suitable for Modeling N/A
Equity Investment in the Project It has become common in some jurisdictions for contractors to make an equity investment in the project. Contractors with an economic stake beyond their fee are generally be regarded as being more committed to the project's success and are less likely than the average contractor to walk away from a severely under-performing project, bolstering the rating.
18
Table 10: Contractor's Equity Investment

All Projects Above Average Contractor's equity investment exceeds its expected margin on the project and is locked in beyond completion and acceptance. Contractor either has no equity investment or the investment is less than its expected margin on the project. N/A N/A
Average
Below Average Not Suitable for Modeling
COMPLEXITY OF THE PROJECT / PROJECT MANAGEMENT TASK

Even within a specific project type there will be differences in risk profile from project to project. As well, the degree of conservatism of the construction schedule, the level of design risk and local factors - such as the predictability of pricing for labour and materials over the life of the contract - can also differentiate one project from another. Project Simplicity & Feasibility of Schedule Moody's assesses a project's simplicity and the feasibility of its construction schedule by comparing it to the average for its type. We do this by reviewing reports from the technical advisor and/or independent engineer as well as the off-taker's feasibility studies (where available) and the performance of similar projects rated previously by Moody's
Table 11: Project Simplicity

Building Project Above Average Construction is to take place on greenfield or brownfield sites with no material external elements likely to delay construction. Civil Engineering Project Construction takes place within an existing known space (e.g. for standard civil engineering - road extension / widening along existing alignment or for complex civil engineering construction construction of a new bridge adjacent to an existing one using the same basic design. All technologies and techniques are very well established within the field with very high degree of confidence in outcome Construction is to take place on a greenfield and / or un-congested site with no particular environmental issues concerns or structures to cause concern. Technologies and process used in construction are considered normal with usual application risks outstanding New technologies or processes stretching the limits of what has been done previously, or construction is to take place in a challenging environment (e.g. construction of a structure in a highly congested urban environment, requiring extensive and difficult traffic management) The project is employing untried and untested material or processes, which provides material uncertainty that construction can be completed within a reasonable timeframe
All technologies and techniques are very well established within the field with very high degree of confidence in outcome Average Where construction is to take place on existing sites, there is unlikely to be any material impact on construction works. Technologies and process used in construction are considered normal with usual application risks outstanding
Below Average
Construction takes place on live operating sites, involving a complex process of decanting services from one building to another and demolition works that may affect continuing operations; or construction processes are considered particularly unusual and challenging Rarely, when the construction environment is considered extremely difficult - e.g., a requirement to construct in a very contaminated environment
Not Suitable for Modeling
19
Table 12: Reasonability of the Construction Schedule

Building Project Above Average Project involves the construction of multiple buildings but where interdependencies are minimal allowing rescheduling of tasks without affecting the critical path. Time contingencies provide good protection against uncertainties. Design work is well advanced Time contingencies provide good protection against uncertainties with total time floats equivalent to 5% - 10% of the total construction time allowed. Design work is well advanced by financial close, with outline designs complete Time contingencies are tight,and material slippage in the critical path would likely endanger final completion dates. Design work is not very well advanced. Either time contingencies are wholly inadequate, with material delays possible if construction takes take longer than expected or there is very limited design work by financial close. Civil Engineering Project Project is split into phases or can be modularised for partial completion, enabling contractor to deliver specific phases of work earlier and counteract delays elsewhere. Time contingencies provide good protection against uncertainties. Design work is well advanced Time contingencies provide good protection against uncertainties with total time floats equivalent to 5% - 10% of the total construction time allowed. Design work is well advanced by financial close, with outline designs complete Time contingencies are tight and material slippage in the critical path would likely endanger final completion dates. Design work is not very well advanced. Either time contingencies are wholly inadequate, with material delays possible if construction takes take longer than expected or there is very limited design work by financial close.
Average
Below Average
Not suitable for Modeling
Construction Budget A contractor operating with a very aggressively priced construction budget, with small contingencies and narrow margins is more likely to experience cost over-runs than is one with conservative cost estimates and generous contingencies. Moody's will look to the technical advisor / independent engineer as well as to its own experience in rating project financings to judge the reasonability of the budget and building schedule. Given that PFI / PPP projects are awarded pursuant to a competitive bid process and contractors are under pressure to submit the lowest possible construction bid, Moody's does not expect to see Above Average outcomes on this factor very often.
Table 13: Construction Budget

All Projects Above Average Budgeted unit costs are materially higher than expected based on current market forecasts for labour, materials, energy etc. and the contractor's ability to hedge purchases, OR Cost contingencies are generous, providing material support for possible cost overruns. Budgeted unit costs are consistent with current market forecasts for labour, materials, etc., AND Cost contingencies are appropriate for the project. Budgeted unit costs or cost contingencies are optimistic. No independent reports / insufficient reporting, no market data or independent consultants' reports to provide a view on costs
Average
Below Average Not Suitable for Modeling
Vulnerability to Local Economic Conditions As the construction task in most PFI / PPP projects is a multi-year exercise, the contractor is potentially exposed to changes in the local markets for labour, materials and energy. This issue is likely to be a problem only rarely, as local economic conditions are generally well considered when a project company prepares its bid for a concession. However, if cited as a separate characteristic, it will generally put downward pressure on the model rating. Certain markets may be particularly vulnerable because of high projected construction activity and shortages of labour and materials. This condition is associated most often with venues that are preparing for a particular event, such as the infrastructure build associated with hosting the Olympic Games or in a sector experiencing an economic peak such as the minerals boom in Western Australia. This factor typically results in a generally high level of construction activity in the local economy and an inability to offset or transfer the risks of rising input costs, in both cases putting downward pressure on the rating. Moodys Special Comment
20
Table 14: Vulnerability to Local Economic Conditions

All Projects Above Average Average N/A Economic conditions foreseen and considered as part of the concession bid. Contractor has engaged in forward purchases of materials and energy for the project to the degree possible and transferred labour costs through sub-contracting or collective agreements that extend over the life of the construction project. Economic pressures arose after bid was submitted or contractor did not take actions to limit exposures to labour, materials and energy costs over the life of the construction project. Local / regional economy is suffering extreme distress.
Below Average
CONTRACTUAL FRAMEWORK
PFI / PPP transactions are designed to shift certain of the financing, construction and operating risks of public infrastructure projects on to the private sector. To the degree that a particular concession agreement transfers significantly more or less risk than is typical, the model rating will be adjusted accordingly. In a PFI / PPP project's operating period, an analysis of the risk sharing embedded in the concession's contractual framework will focus largely on the balance between the complexity of the operators' responsibilities and the severity of the performance management, abatement and termination regimes. It may have a positive or negative impact on the risk characteristics of the project as a whole. In contrast, the construction period's single performance objective is to deliver the project on time, on budget and fit for its designed purpose. Therefore the construction framework analysis is more concerned with the degree to which the concession agreement assigns risks to the project that are beyond the project company's ability or intention to manage or control. In addition to the characteristics cited below, the assessment of a project's contractual framework also considers several pass/fail issues. A 'fail' would typically result in the project being characterized as Not Suitable For Modeling and might make it impossible for Moody's to assign a rating at all. Two common examples of pass / fail issues would be the presence or absence of an independent engineer / certifier to monitor the construction budget and control payments and the presence or absence of a comprehensive insurance package from a credit worthy insurer which has been vetted by a credible, independent insurance consultant. Site Related Issues Sites for PFI / PPP projects are usually selected by the off-taker or government long before the project company becomes involved. However, as part of the bid process, the project company is generally given access to any geotechnical or environmental assessments conducted by the off-taker so they may price their bid appropriately. The exposure of a project company to known or unknown site-related issues varies widely.
Table 15: Site Related Issues

All Projects Above Average Government / off-taker is responsible for all pre-existing site conditions including all environmental, geotechnical and archaeological issues and holds the project company harmless for their effects. Government / off-taker is responsible for all pre-existing site conditions not disclosed to the project company prior to financial close. Project company is responsible for the costs of dealing with any disclosed environmental or geotechnical issues. Project company assumes full responsibility for the site at financial close but has had access to site assessments that are comparable in detail and scope to those reasonably expected to be completed in connection with an arms-length sale of the site itself. Project company assumes full responsibility for the site at financial close without access to a full site assessment (unlikely).
Average
Below Average
21
Site Access, Acquisition and Planning PFI / PPP projects may be exposed to increased costs and delays as a result of a failure by the Off-taker to acquire the land and / or provide access to the site; a failure to obtain the relevant planning and other permits required to build the project in accordance with the agreed design, or issues surrounding utility connections and diversions.
Table 16: Site Access, Acquisition and Planning

Building Project Above Average Risks associated with obtaining all land, construction permits and procuring all utilities services (time and cost implications) remains with the off-taker. Civil Engineering Project Risks associated with obtaining all land and construction permits (time and cost implications) remains with the off-taker. No required utilities diversions that would likely have a negative effect on construction. Land has been / will be acquired by off-taker and handed over to project company at financial close. Failure to do so will be a full compensation (time and costs) event for project company. The acquisition of permits is not considered to be a problem. Off-taker commits to support the application for such permits where appropriate. Utilities diversions, while a risk of the project company, are not likely to hold up construction or result in significant cost overruns. Most land has been / will be acquired by off-taker and handed over to project company. Modest residual land acquisition risks are with the project company, but are expected to be manageable. There may be material utilities diversion risks or acquisition of permits risks (possibly arising form local authorities). There are material land acquisition risks residing with the project company
Average
Land has been / will be acquired by off-taker and handed over to project company at financial close. Failure to do so will be a full compensation (time and costs) event for project company. All permits are received or are well in hand with none likely to hold up construction. Off-taker commits to support the application for such permits where appropriate. No utilities diversions that could have a significant impact on project completion.
Below Average
Most land has been / will be acquired by off-taker and handed over to project company. The risk of delay is shared between the off-taker and the project company. Procurement of permits or utility connections has some potential to cause project delays or higher costs
Not suitable for Modeling Land acquisition risk (cost and / or timing) lies with the project company.
A failure by the off-taker to acquire the site and / or provide access to the project company is generally outside the control of the project company - making it difficult to mitigate the risk of delays or increased costs. Obtaining planning and other permits - whilst part of the contractor's day to day business and taken into account in pricing and scheduling - is also outside the control of the project company / contractor. In contrast, the off-taker may have the ability to exercise its statutory and regulatory powers to facilitate planning and other permits. Similarly, the off-taker will often be in a better position to ensure that the site is connected to all appropriate utilities and undertake appropriate diversions. Extraordinary Event Provisions (Force Majeure and Change in Law) Similar to the evaluation of risks regarding the project site, the evaluation of the contractual provisions regarding extraordinary events focuses on the degree to which the project company is exposed to risks it either cannot control or decides not to mitigate.
22
Table 17: Extraordinary Event Provisions

Force Majeure - All Projects Above Average Government / off-taker is explicitly responsible for making payments sufficient to service debt during all force majeure events. Government / off-taker and project company responsible for making limited payments during force majeure but: (i) project company liquidity is sufficient to cover the entire period from declaration of force majeure through concession termination and receipt of termination payment without relying on any ongoing payments, or (ii) Force Majeure risk is passed down to a subcontractor that has backed its obligations with adequate security. Government / off-taker and project company responsible for making limited payments during force majeure and project company does not have sufficient liquidity to cover the entire period from declaration of force majeure through concession termination and receipt of termination payment without relying on any ongoing payments. An extended force majeure likely to cause a cashflow default on the project. Change in Law - All Projects Government / off-taker accept all change in law risk.
Average
Government / off-taker share risk of a nondiscriminatory change in law. Project has either reserved up to the limit of its exposure, the amount of potential exposure not sufficient to cause financial distress to the project or the project has passed change in law risk down to a sub-contractor that has backed its obligations with adequate security.
Below Average
Project company responsible for the effects of a non-discriminatory change in law; no risks are passed down or reserves maintained and project agreement does not provide for any resetting of fees in the event of a significant externality such as a change in law.
Contemplated changes in law likely to cause a default on the project.
RELATED PARTY EFFECTS

Moody's generally expects that all parties involved in a PFI / PPP will be sufficiently experienced and capable of carrying out their responsibilities, be motivated to act in support of the Project's success and that they will be faithful to their contractual responsibilities as written. We do not assume that any of the parties will go beyond their contractual obligations and support a PFI / PPP transaction when they are not required to do so, nor that they will actively seek to avoid their responsibilities. To the extent that actions by a related party call these assumptions into question, a qualitative adjustment may be warranted. However, it would be rare to see the effect - positive or negative - at the outset of a project. As a result, it would be unusual for Moody's to find anything other than an Average result for a new transaction in the absence of recent actions by the same parties in the same jurisdiction. Government / Off-taker PFI / PPP transactions represent an alternative infrastructure procurement mechanism for governments. However, because PFI / PPP projects deliver public infrastructure, factors beyond just pure economics may affect government decisions about their dealings with PFI / PPP concessions. In particular, governments may in some circumstances be inclined to support a PFI / PPP project beyond contractual requirements - or more mildly, not enforce their contractual rights to the letter. This may happen when a project is important to the jurisdiction or if it is feared that the failure of the project may impair the jurisdiction's future access to the PFI / PPP market for politically or socially important infrastructure. On the other hand, some aspects of PFI / PPP transactions can be deeply unpopular with some groups able to bring pressure to bear on governments. This may be, simply, a groups' opposition to contracting out of public services to the private sector in principle or, more often, because some aspects of a transaction have a negative effect on a certain constituency in some way. Governments themselves have also been known to change their policy perspectives or have objectives overturned by voters. Existing concession agreements can be enforced to the letter, challenged in the courts or, at the theoretical extreme, cancelled by the exercise of a government's sovereign power. In evaluating the effect of a particular government / off-taker on a project, Moody's will consider its demonstrated willingness to support previous PFI / PPP projects beyond the contractual requirements or, conversely, the degree to which they are challenging existing concessions in the courts.
23
Other Related Party Effects Other parties to a PFI / PPP may also go beyond their contractual obligations and support a PFI / PPP project if it is in their long term economic interest to do so. For example, the lead equity investor in a Project may think it economic to make a small additional investment to complete construction - and by extension, earn the rights to the project's operating cash flows - rather than lose the entire investment should the concession be terminated for non-completion. Because, however it is difficult to predict such support in advance, Moody's is unlikely to consider the credit benefit of non-contractual support from other related parties in advance of evidence that it is occurring.
24
Liquidity
Liquidity analysis in the context of a typical industrial company considers the likelihood that it can survive for a period of time without access to the capital markets to refinance maturing obligations. In contrast, liquidity analysis for a PFI / PPP project in construction considers whether it can continue paying debt service either until construction is complete and the project begins to receive its operating payments or the concession is terminated and the termination payment has been received. If the qualitatively-adjusted rating of the project during construction already reflects its inherent credit profile (project complexity, the quality of the contractor, value of financial supports, etc.) then the requirement for liquidity is simply that there be enough of it so that the possibility of a liquidity-induced default does not add to the expected loss. Consider the impact on creditors should a project default on its debt before the construction sunset date, solely because of a lack of liquidity. Generally, the loss to debt holders in this situation would be greater than if construction was able to continue until the sunset date - if for no other reason than the cost-to-complete deduction from a termination payment usually falls as the project advances. Therefore, if a lack of liquidity can cause a default prior to the sunset date (or the date at which the construction risk mitigation package is entirely exhausted) the rating on the project should be lower than if that is not possible. Theoretically, the required liquidity would be the amount needed to ensure that the probability of a liquidityinduced default was consistent with the default probability component of the rating of the project as a whole. While it would be mechanically possible to construct probability distributions that considered the likelihood of construction schedule over-runs of various durations - and therefore measure the need for various amounts of liquidity - such an exercise is not practical in light of the limited additional precision it would bring to the model result25. As a result, Moody's has looked to its traditional practices for rating project financings in establishing liquidity thresholds for various rating categories in PFI / PPP projects.
Table 19: Minimum liquidity levels for Rating Categories in PFI / PPP Projects
Target Rating Range Aa Standard Projects Committed liquidity sufficient to pay all debt service from the scheduled construction date through to the date by which the termination payment would be received or the debt redeemed from letter of credit or other proceeds. Sufficient liquidity to cover a 30% construction schedule over-run, subject to a three-six month minimum. Sufficient liquidity to cover a 15% construction schedule over-run, subject to a three month minimum. Complex Projects Committed liquidity sufficient to pay all debt service from the scheduled construction date through to the date by which the termination payment would be received or the debt redeemed from letter of credit or other proceeds. Sufficient liquidity to cover a 30% construction schedule over-run, subject to a three-six month minimum. Sufficient liquidity to cover a 15% construction schedule over-run subject to a three month minimum.
Baa
The most credit is assigned to liquidity support structured as a cash-funded debt service reserve account in a regulated deposit-taking institution with a Moody's rating at least as high as the target rating for the project's senior debt. Liquidity support via letters of credit or other financial supports - as defined above - are also valuable. As previously discussed however, Moody's considers performance supports to have only limited liquidity value. To the extent that any portion of the risk mitigation package's financial supports are either notionally or potentially dedicated to liquidity, those amounts will be subtracted from the amounts available for credit support during the modeling of the transaction. It should be noted that the role of liquidity in construction is different from its role in operations. Once a project has been completed and begun operations, the role of liquidity is to provide a buffer against the effects of payment abatements for non-performance by providing enough time for the operational difficulties that caused the abatement to be corrected. Operations period liquidity will be addressed in the operational period methodology to be published in the coming weeks.
25. Not only is data on schedule over-runs likely even harder to obtain than on overall cost over-runs, such data would not be predicatively useful. Construction contractors generally have the ability to convert time to money and are likely to do so if it makes economic sense - rendering purely schedule-based data unreliable. For example, it may be much less expensive for a contractor to spend on overtime and accelerate a project than to suffer the expense of liquidated damages for the same period.
25
Ratings after the Transition to Operations

In many PFI / PPP financings, construction period risks are sufficiently mitigated to cause the risk profile to be broadly similar to that expected during the project's operational life. As a result, Moody's ratings on PFI / PPP project financings have not generally experienced significant upgrades post-construction, regardless of jurisdiction or analytical area. However, through the use of different amounts and types of external support, it is relatively easy to structure the credit risk of the construction period for a PFI / PPP transaction to the desired credit outcome. Depending on credit market demand and the cost of different credit enhancement options, projects may intentionally be structured to very different risk levels during construction than in operations. Moody's will comment at the time a rating is assigned on whether it believes there to be a significant difference between the risk profile in construction and operations and whether the rating is depressed during construction as a result. Should the construction period risk profile be lower than that is expected to prevail in operations, the higher risk operational period will depress the rating throughout the project's life. Ratings changes as a result of a transition to project operations will generally not be considered until Moody's is satisfied that the project is successfully operating at or near expected levels and that it has done so long enough for any design or construction-related operational issues or problems to have been identified.
26
Appendix A: Project Complexity Definitions

STANDARD BUILDINGS
Buildings not requiring unique or otherwise unusual design considerations. These include most office or residential buildings, out-patient or ambulatory care medical facilities, medium and lower security prisons as well as green-field airport terminal buildings.
COMPLEX BUILDINGS
Buildings requiring unique or unusual design considerations as a result of any of space constraints, complicated site characteristics or requirements to maintain services throughout the construction period. Generally, brown-field redevelopment or refurbishment projects will be categorized as complex. Other examples include general or specialty hospitals or high security prisons.
STANDARD CIVIL INFRASTRUCTURE

Projects involving the construction of facilities, in addition to buildings, not requiring unique or unusual considerations e.g. most new roads in areas without unusual geotechnical issues.
COMPLEX CIVIL INFRASTRUCTURE

Projects involving the construction of facilities, in addition to buildings, that require special design considerations as a result of space constraints, complex geotechnical issues or unusual output specifications. Also included are projects that incorporate innovative design, architectural or construction features or those that incorporate technically complex requirements such as bridges or roadways in difficult terrain, any tunneling other than minimum amounts of 'cut and cover' tunnels and any projects requiring that existing services be maintained throughout the construction period.
27
Appendix B: Typical Performance Supports

AUSTRALIA
Performance Bond / Insurance Bond
Provided by highly rated insurers licensed to do business in Australia. Insurer undertakes to pay the beneficiary on presentation of the bank guarantee at its office. The obligation to pay is unconditional and on demand. The insurer agrees to pay the beneficiary without reference to the insurer's customer, regardless of any notice from the insurer's customer not to pay, and without reference to whether there is any subsisting breach or default under the underlying contract in respect of which the guarantee is issued. These instruments are drafted identically to an Australian bank guarantee except that the provider is a multi-line insurer.
Adjudication Bond
Not usual in the Australian market. Provided by a highly-rated bank licensed to do business in Australia and drafted in the form of a letter of credit. The obligation to pay is conditional on a statutory declaration from the beneficiary to the effect that (i) other security has been exhausted (ii) the contractor is either insolvent OR the adjudicator has made a determination under the Construction Contract in favour of the beneficiary.
NORTH AMERICA
Performance Bond
Provided by highly-rated multi-line insurers. Insures performance by the contractor of the construction contract obligation. On a default by the Contractor, the insurer is required to (i) support the existing contractor in the completion of its obligation; (ii) perform the obligation itself - usually by engaging a substitute contractor; or (iii) pay out a cash amount in lieu of performance. A performance bond is generally only callable if the default is clearly and demonstrably the fault of the insured. Unless explicitly specified, performance bonds typically do not provide for the payment of liquidated damages or other amounts for delays.
Sub-Contractor Insurance
Typically purchased by the contractor to insure the performance of sub-contractors it has engaged, it is an alternative to requiring that sub-contractors provide their own bonding. The project does not benefit directly from the purchase of such insurance. However, by making it more likely that the contractor will be able to absorb the default of its subcontractors, sub-contractor insurance contributes to the likelihood of the project being completed on time.
UNITED KINGDOM
Adjudication Bond
Pays against a claim from the project company against the contractor, where such claim has been approved by an adjudicator in accordance with the adjudication procedure set out in the construction contract. As a consequence, all claims contested by the contractor need to go through an adjudication procedure. The adjudication procedure will have clear timelines for the adjudicator to reach a final decision and issue a certificate of its decision. The adjudication bond will respond to the claim supported by the certificate of decision irrespective of whether or not the contractor subsequently seeks a court remedy to overturn the adjudicator's decision. If the contractor becomes insolvent, any outstanding or post-insolvency claim on the contractor (including claims for liability under construction contract termination) are referred to a referee under the terms of the adjudication bond. The referee is required to determine the contractor's liability and issue a decision within a defined timeline. The adjudication bond then pays against the referee's decision irrespective of whether the contractor or the surety provider subsequently seeks a court remedy to overturn the referee's decision. The referee procedure is designed to cover the concern that an insolvency of the contractor may effectively frustrate the operation of the construction contract adjudication procedure.
28
Appendix C: Technical Information About The Model

The expected loss on the project over-run has been calculated using a Monte Carlo-based simulation model. This has been implemented using the functionality of CDOROMTM, the model used by Moody's analysts to rate synthetic CDO transactions. As an introduction, assume that the over-runs on the project follow a beta distribution L ~ B ( , ) with probability density function
f (l ) =
1 l ( l ) 1 B( , )
with mean =
and standard deviation = + ( + + 1)( + )2
The first loss on the project is absorbed by the building contractor up to a pre-agreed level K. Hence when no defaults are considered, the exposure to loss on the project by the investors is given by Max(0,(L-K)). In other words, if the project over-run L is less than the amount K absorbed by the building contractor, then the loss to the investors is zero. If L exceeds K, then the investors must absorb the excess loss (L-K). The expected loss on the project over-run is then E[Max(0,(L-K))], which has a closed form analytical solution when no defaults are considered. In this case, it can be shown that
E[Max (0, (L K ))] = Max (0, (L K ))dL = (l K ) f (l )dl

K 0
=
where L ~ B( + 1, ) .
P L K K P(L K ) +
Once the assumption that the contractor may default is introduced, the variable K becomes a function of X, K(X), where X is the indicator of the contractor default,
K K (X ) = 0
X =0 X =1
Further, if it is assumed that the contractors obligations are covered by a third party which could also default we have Expected Loss = E[Max(0,(L-K(X,Y)))] where Y is the indicator of the third part default. In this case
K1 K K (X , Y ) = 1 K 2 0
for pre-determined K1 and K2.
K = 0, Y = 0 K = 0, Y = 1 K = 1, Y = 0 K = 1, Y = 1
29
This can be extended to the case where the contractor has multiple layers of support, where each counterparty can default, and dependant on which counterparties have not defaulted, the attachment point varies. Moody's compares all projects against a 4 year idealized loss rate to generate ratings rather than the actual term of the construction project. Given this idealized 4 year timescale together with Moody's rating of each entity, the probability of default over the lifetime of the project for each entity can be determined from Moody's table of 'Idealized' Cumulative Expected Loss Rates. From these default probabilities, P (X=1) = x and P (Y=1) = yi where i represents support level number, we can find the default thresholds x and yi such that (x) = x and (yi) = yi where represents the cumulative distribution function of the standard normal distribution. Moody's uses a standard normal or Gaussian dependency structure for pair-wise asset correlations based on a set of assumptions defined mainly on the sector and geography of the entities26. The correlated random variables are drawn using a standard multi-factor model (see CDOROMTM v2.3 User Guide27 for a technical description) -
X j = G * Z G + I * Z I + I ,R * Z I ,R + 1 G I I ,R * j
where
j is the idiosyncratic factor specific to asset j.
Currently Moody's corporate CDS asset correlation assumptions are G =3% I takes the following values
Sector's Geographical Impact Global ("G") Semi-Local ("SL") Local ("L") Asset Correlation 6% 3% 0%
Sector's Geographical Impact Global ("G") Semi-Local ("SL") Local ("L")
High ("H") 11% 14% 17%
Sector's Correlation Intensity Medium ("M") Low("L") 6% 1% 9% 4% 12% 7%
In addition, the correlation between the project and the contractor has been defined as 18% for an investment grade contractor and 36% for a speculative grade contractor. The default thresholds x and yi can then be compared to the standardized asset values xi. If the standardized asset value is below the default threshold, the asset is considered to have defaulted. The beta distribution recovery rate mean and standard deviation for the project over-run are obtained from the mean loss and loss standard deviation for the type of project being modelled. The recoveries on the contractor default have been modelled using a fixed negative digital recovery rate to allow for the extra cost incurred by the project on contractor default. This has the effect of shifting the loss distribution for the project to the right. The support level recoveries have been modelled using a beta distribution with assumed recovery rates mean and standard deviation for an equivalent SU Bond.
26. For corporate correlations, see Moody's rating methodology, "Moody's Revisits its Assumptions Regarding Corporate Default (and Asset) Correlations for CDOs", 30 November 2004 27. CDOROMTM v2.3 User Guide, Moody's Investors Service, 12 May 2006
30
Due to the use of multiple beta distributions within this model, a closed form analytical solution is no longer available, and so Monte Carlo simulations have been run to predict the expected loss on the project. In each Monte Carlo trial, defaults and recovery values upon default are simulated for the project over-run, the contractor and each level of support. Losses on the project are then computed and averaged over a large number of simulations to give the expected loss on the project.
31
To order reprints of this report (100 copies minimum), please call 1.212.553.1658. Report Number: 98409
Author Andrew J. Kriegler Production Associate Tara Cheparev
Copyright 2006, Moodys Investors Service, Inc. and/or its licensors and affiliates including Moodys Assurance Company, Inc. (together, "MOODYS"). All rights reserved. ALL INFORMATION CONTAINED HEREIN IS PROTECTED BY COPYRIGHT LAW AND NONE OF SUCH INFORMATION MAY BE COPIED OR OTHERWISE REPRODUCED, REPACKAGED, FURTHER TRANSMITTED, TRANSFERRED, DISSEMINATED, REDISTRIBUTED OR RESOLD, OR STORED FOR SUBSEQUENT USE FOR ANY SUCH PURPOSE, IN WHOLE OR IN PART, IN ANY FORM OR MANNER OR BY ANY MEANS WHATSOEVER, BY ANY PERSON WITHOUT MOODYS PRIOR WRITTEN CONSENT. All information contained herein is obtained by MOODYS from sources believed by it to be accurate and reliable. Because of the possibility of human or mechanical error as well as other factors, however, such information is provided as is without warranty of any kind and MOODYS, in particular, makes no representation or warranty, express or implied, as to the accuracy, timeliness, completeness, merchantability or fitness for any particular purpose of any such information. Under no circumstances shall MOODYS have any liability to any person or entity for (a) any loss or damage in whole or in part caused by, resulting from, or relating to, any error (negligent or otherwise) or other circumstance or contingency within or outside the control of MOODYS or any of its directors, officers, employees or agents in connection with the procurement, collection, compilation, analysis, interpretation, communication, publication or delivery of any such information, or (b) any direct, indirect, special, consequential, compensatory or incidental damages whatsoever (including without limitation, lost profits), even if MOODYS is advised in advance of the possibility of such damages, resulting from the use of or inability to use, any such information. The credit ratings and financial reporting analysis observations, if any, constituting part of the information contained herein are, and must be construed solely as, statements of opinion and not statements of fact or recommendations to purchase, sell or hold any securities. NO WARRANTY, EXPRESS OR IMPLIED, AS TO THE ACCURACY, TIMELINESS, COMPLETENESS, MERCHANTABILITY OR FITNESS FOR ANY PARTICULAR PURPOSE OF ANY SUCH RATING OR OTHER OPINION OR INFORMATION IS GIVEN OR MADE BY MOODYS IN ANY FORM OR MANNER WHATSOEVER. Each rating or other opinion must be weighed solely as one factor in any investment decision made by or on behalf of any user of the information contained herein, and each such user must accordingly make its own study and evaluation of each security and of each issuer and guarantor of, and each provider of credit support for, each security that it may consider purchasing, holding or selling. MOODYS hereby discloses that most issuers of debt securities (including corporate and municipal bonds, debentures, notes and commercial paper) and preferred stock rated by MOODYS have, prior to assignment of any rating, agreed to pay to MOODYS for appraisal and rating services rendered by it fees ranging from $1,500 to $2,400,000. Moodys Corporation (MCO) and its wholly-owned credit rating agency subsidiary, Moodys Investors Service (MIS), also maintain policies and procedures to address the independence of MISs ratings and rating processes. Information regarding certain affiliations that may exist between directors of MCO and rated entities, and between entities who hold ratings from MIS and have also publicly reported to the SEC an ownership interest in MCO of more than 5%, is posted annually on Moodys website at www.moodys.com under the heading Shareholder Relations Corporate Governance Director and Shareholder Affiliation Policy. This credit rating opinion has been prepared without taking into account any of your objectives, financial situation or needs. You should, before acting on the opinion, consider the appropriateness of the opinion having regard to your own objectives, financial situation and needs.
32

Construction Risk in PPP Projects

Enviado por

Dados do documento

Descrição original:

Direitos autorais

Formatos disponíveis

Compartilhar este documento

Compartilhar ou incorporar documento

Opções de compartilhamento

Você considera este documento útil?

Este conteúdo é inapropriado?

Direitos autorais:

Formatos disponíveis

Construction Risk in PPP Projects

Enviado por

Direitos autorais:

Formatos disponíveis

Special Comment

Andrew Kriegler Catherine Deluz Allan Mclean

Andrew Blease Neal Shah

Nicola O'Brien David Howell Tim See

Bart Oosterveld Daniel Gates Nicolas Weill

Request for Comment on:

Construction Risk in Privately-Financed Public Infrastructure (PFI / PPP / P3) Projects

Moodys Special Comment

Understanding Construction Risk in a PFI / PPP Concession ............................................................ 6

Modeling Construction and the Risk Mitigation Package ............................................................... 11

Qualitative Considerations ............................................................................................................ 17

Moodys Special Comment

What is PFI / PPP?

About this Rating Methodology

Moodys Special Comment

1. Construction Risk in a PFI / PPP Framework

2. Modeling Construction Risk

3. The Application of Qualitative Rating Factors

Moodys Special Comment

Understanding Construction Risk in a PFI / PPP Concession

Moodys Special Comment

THE DESIGN-BUILD AGREEMENT

Typical PFI/PPP Transaction Structure

Concession Off-taker Agreement

Design - Build Agreement

Construction General Contractor

LENDERS' EXPOSURE TO CONSTRUCTION RISK

Moodys Special Comment

Typical PFI/PPP Design-Build Structure

Construction Company Ultimate Parent

Letter of Credit Full/Partial Guarantee of Performance

Performance/ Adjudication Bond

Moodys Special Comment

3. Performance / Insurance-based Supports

4. Liquid / Financial Supports

Moodys Special Comment

EXPRESSING CONSTRUCTION RISK AS CREDIT RISK

Table 1: Mapping Construction Risk to Financial Risk

How does the model work?

Moodys Special Comment

Modeling Construction and the Risk Mitigation Package

Table 2: Modeling Requirements for PFI / PPP Construction Risk Packages

Financial Support Cash holdbacks, other

Bank obligation Cash deposits or obligations supported by LC's

MODELING RAW CONSTRUCTION RISK

CONSTRUCTION LOSS SEVERITY DISTRIBUTIONS

Moodys Special Comment

Loss Given Default Non-Financial Speculative Grade Corporations

Moodys Special Comment

Table 3: Project LGD Means and Standard Deviations by Type1

Mean Loss 48% 8% 15% 11% 18%

Loss Std Deviation 28% 8% 12% 12% 14%

Loss Given Default Non-Financial Corporations & Construction Projects

MODELING CONTRACTOR SUPPORT

THE VALUE OF CONTRACTOR SUPPORT

Moodys Special Comment

Table 4: Mapping Construction Contractor Support to Model Inputs

Value after haircut

EFFECTS OF A CONTRACTOR DEFAULT

20.00% Standard Building

30.00% 40.00% 50.00% Loss (%) Std Bldg w Contractor Default /

Moodys Special Comment

Table 5: Project Mean Losses w/ Contractor Defaults