Citation
Investment Risk Based on Traffic Forecasting Accuracy

Material Information

Title:
Investment Risk Based on Traffic Forecasting Accuracy Case Studies of U.S. Highway Public-Private Partnerships
Creator:
Fisher, Kyle S
Place of Publication:
[Gainesville, Fla.]
Florida
Publisher:
University of Florida
Publication Date:
Language:
english
Physical Description:
1 online resource (281 p.)

Thesis/Dissertation Information

Degree:
Doctorate ( Ph.D.)
Degree Grantor:
University of Florida
Degree Disciplines:
Civil Engineering
Civil and Coastal Engineering
Committee Chair:
ELLIS,RALPH D,JR
Committee Co-Chair:
GLAGOLA,CHARLES ROBERT
Committee Members:
YIN,YAFENG
HOUSTON,JOEL F
Graduation Date:
12/13/2013

Subjects

Subjects / Keywords:
Analytical forecasting ( jstor )
Cash flow ( jstor )
Debt service ( jstor )
Loan defaults ( jstor )
Modeling ( jstor )
Optimism ( jstor )
Purchasing power parity ( jstor )
Risk premiums ( jstor )
Tolls ( jstor )
Traffic forecasting ( jstor )
Civil and Coastal Engineering -- Dissertations, Academic -- UF
finance -- infrastructure -- partnerships -- public -- public-private
City of Miami ( local )
Genre:
bibliography ( marcgt )
theses ( marcgt )
government publication (state, provincial, terriorial, dependent) ( marcgt )
born-digital ( sobekcm )
Electronic Thesis or Dissertation
Civil Engineering thesis, Ph.D.

Notes

Abstract:
The scarcity of public funding, coupled with crumbling infrastructure, and demands for infrastructure improvements has led to increased usage of public-private partnerships (PPP) as an alternative delivery method for surface transportation infrastructure projects in the United States. PPPs however have a somewhat mixed track record; an unusually high number of PPPs have defaulted on debt due to lower than expected traffic and revenue.  These observations led this study to investigate the potential losses to debt and equity due to the inaccuracies in traffic and revenue forecasting.  Building on the traffic forecast research of Dr. Bent Flyvbjerg and Dr. Robert Bain, a top down approach was taken to investigate the problem.  Five U.S.highway PPP projects were selected for investigation based on project type,location and availability of financial data. Financial data from Private Activity Bond transcripts was used to construct cash flow models that emulated the concessionaires’ cash flow models and accounted for funds during the project term.  Next a Traffic Distribution Instrument was created to vary the lender’s base case traffic forecasts to a user-specified confidence interval on a normal probability distribution function.  These new traffic levels were entered into the Simulated Cash Flow Model and their impacts were observed on equity IRR and debt default.  The effects of optimism bias in traffic forecasting were investigated by creating two traffic forecasting accuracy distributions, one with no optimism bias (u=1.00, sd=0.34)and one with optimism bias (u=0.77, sd=0.26). It was found that the risk of default on debt in PPP projects is three times greater than would be anticipated by a projects BBB- credit rating.  The equity IRR and equity risk premium for the base case scenarios is consistent with other investments but returns are at serious risk when accounting for the volatility of traffic forecasting.  It was also shown that concessionaires do not accurately price the cost of traffic revenue risk.  From these results it was concluded that there is a general underestimation of risk in direct toll PPP projects by both debt investors and equity participants. It was not possible to conclude that optimism bias is present or the cause of the unusually high default rate in PPP projects; however, it was possible to demonstrate that if present, optimism bias would be detrimental to returns on investments in the selected cases. ( en )
General Note:
In the series University of Florida Digital Collections.
General Note:
Includes vita.
Bibliography:
Includes bibliographical references.
Source of Description:
Description based on online resource; title from PDF title page.
Source of Description:
This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Thesis:
Thesis (Ph.D.)--University of Florida, 2013.
Local:
Adviser: ELLIS,RALPH D,JR.
Local:
Co-adviser: GLAGOLA,CHARLES ROBERT.
Electronic Access:
RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2014-12-31
Statement of Responsibility:
by Kyle S Fisher.

Record Information

Source Institution:
UFRGP
Rights Management:
Applicable rights reserved.
Embargo Date:
12/31/2014
Resource Identifier:
907780682 ( OCLC )
Classification:
LD1780 2013 ( lcc )

Downloads

This item has the following downloads:


Full Text

PAGE 1

1 INVESTMENT RISK BASED ON TRAFFIC FORECASTING ACCURACY: CASE STUDIES OF U.S. HIGHWAY PUBLIC PRIVATE PARTNERSHIPS By KYLE FISHER A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLME NT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2013

PAGE 2

2 2013 K yle F isher

PAGE 3

3 To my family

PAGE 4

4 ACKNOWLEDGMENTS To my advisor, Dr. Ellis for all of his support throughout the Ph.D. process and his assistance in obtaining the Universit y of Florida Alumni Fellowship. To the University of Florida Alumni for the funding to purs ue my Ph.D.; t he importance of th is funding cannot be overstated. To my fellow graduate students and o fficemates, for t heir support and understanding. To my parents, thank you for your eternal support on my nev er ending journey as a student. To my beautiful wife Danielle, to who m I owe everything; she is my rock, partner and equal. Thank you for your hel p collecting data, assisting in analysis and countless hours of editing.

PAGE 5

5 TABLE OF CONTENTS page ACKNOWLEDGMENTS ................................ ................................ ................................ ............... 4 LIST OF TABLES ................................ ................................ ................................ ........................... 8 LIST OF FIGURES ................................ ................................ ................................ ....................... 13 LIST OF ABBREVIATIONS ................................ ................................ ................................ ........ 15 ABSTRACT ................................ ................................ ................................ ................................ ... 18 CHAPTER 1 INTRODUCTION ................................ ................................ ................................ .................. 20 Background on Public Private Partnerships ................................ ................................ ........... 20 PPP Project Delivery Types ................................ ................................ ............................ 21 Private Finance ................................ ................................ ................................ ................ 23 Public Private Partnerships in the U.S. ................................ ................................ ................... 25 Increasing Demand ................................ ................................ ................................ ................. 27 Deteriorating Facilities ................................ ................................ ................................ ........... 28 Declining Resources ................................ ................................ ................................ ............... 30 Advantages and Disadvantages of PPPs ................................ ................................ ................. 30 Potential Advantages of PPPs ................................ ................................ ......................... 31 Potential Disadvantages of PPPs ................................ ................................ ..................... 31 PPP Default ................................ ................................ ................................ ............................. 3 2 Purpose of the Study ................................ ................................ ................................ ............... 34 Research Questions ................................ ................................ ................................ ................. 35 2 LITERATURE REVIEW ................................ ................................ ................................ ....... 37 Funding Legislation for Public Private Partnerships ................................ .............................. 37 Federal Highway Trust Fund ................................ ................................ ........................... 38 Transportation Infrastructure Finance and Innovation Act (TIFIA) ............................... 39 SAFETEA LU ................................ ................................ ................................ ................. 40 Moving Ahead for America in the 21 s t Century MAP 21 ................................ ............... 41 Project Finance Structure for PPP projects ................................ ................................ ............. 42 Debt ................................ ................................ ................................ ................................ 43 Seni or d ebt ................................ ................................ ................................ ................ 44 Subordinate d ebt ................................ ................................ ................................ ....... 45 Equity ................................ ................................ ................................ .............................. 46 Credit En hancement ................................ ................................ ................................ ........ 46 Risk in PPP ................................ ................................ ................................ ............................. 47 Traffic Revenue Risk ................................ ................................ ................................ ....... 49 Traffic and Revenue Forecasting ................................ ................................ ............................ 50 The Flyvbjerg Distribution ................................ ................................ .............................. 52 Bain Distribution ................................ ................................ ................................ ............. 54 Credit Ratings on PPP Project Bonds ................................ ................................ ..................... 56

PAGE 6

6 ................................ ................................ ...... 58 Municipal Bond Default ................................ ................................ ................................ .. 60 Summary of Literature Review ................................ ................................ .............................. 61 3 METHODOLOGY ................................ ................................ ................................ ................. 63 Research Questions and H ypotheses ................................ ................................ ...................... 63 Methodological Steps to Answer Each Question/Hypotheses ................................ ................ 64 Case Selection ................................ ................................ ................................ ......................... 65 Types of Cases ................................ ................................ ................................ ................. 65 Number of Cases ................................ ................................ ................................ ............. 67 Selecting Conditions ................................ ................................ ................................ ........ 68 Data Collection ................................ ................................ ................................ ....................... 68 Instrumentation ................................ ................................ ................................ ....................... 69 Traffic Distributions ................................ ................................ ................................ ........ 69 The Cash Flow Model ................................ ................................ ................................ ..... 71 Step 1: Convert Nominal to Real Dollars ................................ ................................ 75 Step 2: Determine Length of Concession t o be Modeled ................................ ......... 76 Step 3: Select Traffic & Revenue Forecasts to be Modeled ................................ ..... 77 Step 4: Perform Regression Analyses on Revenues ................................ ................. 79 Step 5: Perform Regression Analyses on Operations & Maintenance Costs ........... 81 Step 6: Perform Regression Analyses on Major Maintenance Costs ....................... 83 Step 7: Estimate Revenue ................................ ................................ ......................... 85 Step 8: Estimate O&M Expense ................................ ................................ ............... 86 Step 9: Estimate Major Maintenance Expense ................................ ......................... 86 Step 10: Calculate Net Cash Flow Available for Debt Service. ............................... 87 Step 11: Det ermine Debt Service on Private Activity Bonds (PABs) ..................... 87 Step 12: Determine Debt Service on TIFIA Loans ................................ .................. 88 Step 13: An Explanatio n of Reserve Accounts ................................ ........................ 89 Step 14: Calculate Cash Flow Available after Debt Service. ................................ ... 89 Step 15: Calculate Revenue Sharing ................................ ................................ ........ 89 Step 16: Calculate Payments to Equity ................................ ................................ .... 90 Step 17: Calculate Equity IRR ................................ ................................ ................. 90 Data Analysis ................................ ................................ ................................ .......................... 90 Limitations and Delimitations ................................ ................................ ................................ 92 Limitations ................................ ................................ ................................ ....................... 92 Delimitations ................................ ................................ ................................ ................... 92 Interpretation and Refinement of Model ................................ ................................ ................ 92 Summary of Methodology ................................ ................................ ................................ ...... 94 4 RESULTS ................................ ................................ ................................ ............................... 95 Results of the Cash Flow Model Analysis of Selected Case Data ................................ ......... 95 Findings to Answer Ea ch Research Question/Hypotheses ................................ ..................... 96 Research Question 1 ................................ ................................ ................................ ........ 96 Research Question 2 ................................ ................................ ................................ ........ 98 Research Question 3 ................................ ................................ ................................ ........ 99 Research Question 4 ................................ ................................ ................................ ...... 100 Research Question 5 ................................ ................................ ................................ ...... 101

PAGE 7

7 Research Question 6 ................................ ................................ ................................ ...... 103 Research Question 7 ................................ ................................ ................................ ...... 104 5 CONCLUSION ................................ ................................ ................................ ..................... 106 Summary of Conclusions ................................ ................................ ................................ ...... 108 Default Risk ................................ ................................ ................................ ................... 108 Traffic Revenue Risk Effects on Equity ................................ ................................ ........ 109 Ability of Concessionaires to Identify Risks ................................ ................................ 109 Effect of Optimism Bias ................................ ................................ ................................ 109 Di scussion ................................ ................................ ................................ ............................. 110 Debt in PPP Projects ................................ ................................ ................................ ...... 110 Equity in PPP Projects ................................ ................................ ................................ ... 111 Optimism Bias in PPP Projects ................................ ................................ ..................... 112 Recommendations for Practice ................................ ................................ ............................. 113 Recommendations for Further Study ................................ ................................ .................... 113 APPENDIX A TIFIA PROJECT PROFILES ................................ ................................ ............................... 114 B SIMULATED CASH FLOW MODEL CONSTRUCTION ................................ ................ 139 C SIMULATED CASH FLOW MODEL (BASE CASE) RESULTS ................................ ..... 208 D DEFAULT RATE CALCULATIONS ................................ ................................ ................. 272 E EFFECTS OF OPTIMISM BIAS ON SELECTED CASES ................................ ................ 274 LIST OF REFERENCES ................................ ................................ ................................ ............. 277 BIOGRAPHICAL SKETCH ................................ ................................ ................................ ....... 282

PAGE 8

8 LIST OF TABLES Table page 1 1 Common PPP designations ................................ ................................ ................................ 21 1 2 U.S. transportation DBFOM concessions 1993 2013 ................................ ...................... 26 1 3 ASCE grading scale ................................ ................................ ................................ ........... 28 2 1 PABs issued for PPP projects in the U.S. since May 2013 ................................ ................ 45 2 2 Critical s uccess factors ................................ ................................ ................................ ....... 47 2 3 Credit ratings defined ................................ ................................ ................................ ......... 57 2 4 owned toll facilities in the United States ................................ ................................ ........... 58 3 1 Selected cases ................................ ................................ ................................ ..................... 67 3 2 Ratio of actual to forecast traffic ................................ ................................ ....................... 71 3 3 NTE convert nominal to real dollars ................................ ................................ .................. 75 3 4 NTE traffic forecast with annualization ................................ ................................ ............. 78 3 5 NTE traffic & revenue forecast ................................ ................................ .......................... 80 3 6 ................................ ................................ .......... 82 3 7 NTE accu mulated major maintenance ................................ ................................ ............... 84 4 1 Equity IRR given probability of exceedance without optimism bias ................................ 95 4 2 Equity IRR given prob ability of exceedance with optimism bias ................................ ..... 95 4 3 Probability of default without optimism bias ................................ ................................ ..... 96 4 4 Probability of default wit h optimism bias ................................ ................................ .......... 96 4 5 Selected cases bond ratings ................................ ................................ ................................ 97 4 6 Probability of default PABs without optimism bias ................................ .......................... 97 4 7 Probability of default on TIFIA loans without optimism bias ................................ ........... 98 4 8 Probability of default on PABs with optimism bias ................................ .......................... 99

PAGE 9

9 4 9 Probability of default on TIFIA loans with optimism bias ................................ ................ 99 4 10 Equity risk premium for selected cases ................................ ................................ ........... 100 4 11 Equity risk premiums for selected cases assuming optimism bias ................................ .. 102 4 12 Effects of optimism bias on base case ................................ ................................ ............. 102 4 13 Availability payment PPP projects equity risk premium, r isk free rates, and risk premium ................................ ................................ ...................... 103 4 14 Equity risk premium and TIFIA sc heduled change on selected cases ............................. 10 5 B 1 I 495 summary ................................ ................................ ................................ ................. 139 B 2 I 495 convert nominal to real dollars ................................ ................................ ............... 139 B 3 I 495 traffic forecast ................................ ................................ ................................ ........ 140 B 4 I 495 traffic & revenue forecast ................................ ................................ ....................... 141 B 5 I 495 T&R regression analysis summary output ................................ ............................. 143 B 6 I 495 O&M expenses ................................ ................................ ................................ ....... 144 B 7 I 495 O&M regression analysis summary output ................................ ............................ 146 B 8 I 495 accumulated major maintenance ................................ ................................ ............ 147 B 9 I 495 accumulated major maintenance & accumulated t raffic regression analysis summary output ................................ ................................ ..... 149 B 10 I 495 revenue sharing ................................ ................................ ................................ ...... 150 B 11 LBJ summary ................................ ................................ ................................ ................... 151 B 12 LBJ convert nominal to real dollars ................................ ................................ ................. 151 B 13 LBJ traffic forecast ................................ ................................ ................................ .......... 152 B 14 LBJ traffic & revenue forec ast ................................ ................................ ......................... 152 B 15 LBJ T&R regression analysis summary output ................................ ............................... 154 B 16 LBJ O&M expenses ................................ ................................ ................................ ......... 155 B 17 LBJ O&M regression analysis summary output ................................ .............................. 157 B 18 LBJ accumulated major maintenance ................................ ................................ ............... 158

PAGE 10

10 B 19 LBJ major maintenance regression analysis summary output ................................ ......... 160 B 20 LBJ revenue sharing ................................ ................................ ................................ ......... 161 B 21 NTE summary ................................ ................................ ................................ .................. 164 B 22 NTE convert nominal to real dollars ................................ ................................ ................ 164 B 23 NTE traffic forecast ................................ ................................ ................................ .......... 165 B 24 NTE traffic & revenue forecast ................................ ................................ ........................ 166 B 25 NTE T&R regression analysis summary output ................................ ............................... 168 B 26 NTE operations & maint enance expenses ................................ ................................ ........ 170 B 27 NTE O&M regression analysis summary output ................................ ............................. 171 B 28 NTE accumulated major maintenance ................................ ................................ ............. 172 B 29 NTE major maintenance regression analysis summary output ................................ ........ 174 B 30 NTE revenue sharing schedule ................................ ................................ ......................... 176 B 31 ERC summary ................................ ................................ ................................ .................. 180 B 3 2 ERC convert nominal to real dollars ................................ ................................ ................ 180 B 33 ERC traffic for ecast ................................ ................................ ................................ .......... 181 B 34 ERC traffic & revenue forecast ................................ ................................ ........................ 182 B 35 ERC T&R regression analysis summary output ................................ ............................... 183 B 36 ERC O&M expenses ................................ ................................ ................................ ........ 185 B 37 ERC O&M regression analysis summary output ................................ ............................. 186 B 38 ERC accumulated major maintenance ................................ ................................ ............. 188 B 39 ERC accumulated major maintenance & accumulated traffic regression analysis summary output ................................ ................................ ..... 189 B 40 ERC revenue sharing ................................ ................................ ................................ ........ 191 B 41 I 95 summary ................................ ................................ ................................ ................... 195 B 42 I 95 conve rt nominal to real dollars ................................ ................................ ................. 195 B 43 I 95 traffic forecast ................................ ................................ ................................ .......... 197

PAGE 11

11 B 44 I 95 traffic & revenue forecast ................................ ................................ ......................... 199 B 45 I 95 T& R regression analysis summary output ................................ .............................. 200 B 47 I 95 O&M regression analysis summary output ................................ .............................. 204 B 48 I 95 accumulated major maintenance ................................ ................................ ............. 205 B 49 I 95 major maintenance regression analysis summary output ................................ ......... 206 C 1 I 495 traffic & revenue cash flow section ................................ ................................ ........ 208 C 2 I 495 PABs cash flow section ................................ ................................ .......................... 211 C 3 I 495 TIFIA debt sched ules section ................................ ................................ ................. 214 C 4 I 495 equity cash flow section ................................ ................................ ......................... 218 C 5 NTE traffic & revenue cash flow section ................................ ................................ ........ 222 C 6 NTE PABs cash flow section ................................ ................................ ........................... 224 C 7 NTE TIFIA mandatory debt cash flow section .. 226 C 8 NTE TIFIA scheduled debt cash flow secti on ................................ ................................ 228 C 9 NTE equity cash flow section ................................ ................................ .......................... 230 C 11 LBJ PABs cash flow section ................................ ................................ ............................ 234 C 12 LBJ TIFIA mandatory debt cash flow section ................................ ................................ 235 C 13 LBJ reserve accounts cash flow section ................................ ................................ ........... 237 C 14 LBJ equity cash flow section ................................ ................................ ........................... 240 C 15 ERC traffic & revenue cash flow section ................................ ................................ ........ 242 C 16 ERC PABs cash flow section ................................ ................................ ........................... 245 C 17 ERC TIFIA mandatory debt cash flow section ................................ ................................ 247 C 18 ERC reserve accounts cash flow section ................................ ................................ ......... 250 C 20 ERC equity cash flow section ................................ ................................ .......................... 254 C 22 I 95 PABs cash flow section ................................ ................................ ............................ 260 C 23 I 95 T IFIA mandatory debt cash flow section ................................ ................................ 263

PAGE 12

12 C 24 I 95 reserve accounts cash flow section ................................ ................................ ........... 266 D 1 Probability of default on PABs ................................ ................................ ........................ 273

PAGE 13

13 LIST OF FIGURES Figure page 1 1 PPP continuum ................................ ................................ ................................ ................... 22 1 2 Research structu re ................................ ................................ ................................ .............. 35 2 1 PPP Legislation timeline ................................ ................................ ................................ .... 38 2 2 Cash flow of the highway account of the highway trust fund ................................ ........... 39 2 3 General U.S. PPP project financing structure ................................ ................................ .... 43 2 4 Four steps used in traffic modeling ................................ ................................ .................... 51 2 5 Flyvbjerg inaccuracy for road projects ................................ ................................ .............. 53 2 6 Flyvbjerg inaccuracy for rail projects ................................ ................................ ................ 54 2 7 Bain normal traffic distr ibution ................................ ................................ ......................... 55 2 8 Bain traffic distribution adjusting for optimism bias ................................ ......................... 56 2 9 year cumulative proba bility of default rates ................. 61 3 1 Bain probability distribution function forecasting accuracy ................................ .............. 70 3 2 Simulated cash flow model flow of funds ................................ ................................ ......... 74 3 3 NTE revenue vs. traffic graph ................................ ................................ ............................ 81 3 4 NTE O&M vs traffic graph ................................ ................................ ................................ 82 3 5 NTE major maintenance vs time graph ................................ ................................ ............. 84 3 6 NTE accumulated cost vs accumulated traffic graph ................................ ........................ 85 4 1 Equity risk premium for selected cases ................................ ................................ ........... 101 4 2 Effects of optimism bias on I 495 ................................ ................................ .................... 103 4 3 Risk premium comparison of direct toll and AP projects ................................ ................ 104 4 4 Equity risk premium and scheduled change on selected cases ................................ ........ 105 B 1 I 495 revenue vs. traffic ................................ ................................ ................................ ... 14 2 B 2 I 495 O&M vs. traffic ................................ ................................ ................................ ...... 145

PAGE 14

14 B 3 I 495 cost vs. accumulated traffic ................................ ................................ .................... 148 B 4 LBJ revenue vs. traffic ................................ ................................ ................................ ..... 153 B 5 LBJ O&M vs. traffic ................................ ................................ ................................ ........ 156 B 6 LBJ cost vs. accumulated traffic ................................ ................................ ...................... 159 B 7 NTE revenue vs. traffic ................................ ................................ ................................ .... 168 B 8 NTE O&M vs. traffic ................................ ................................ ................................ ....... 171 B 9 NTE cost vs. accumulated traffi c ................................ ................................ ..................... 173 B 10 ERC revenue vs. traffic ................................ ................................ ................................ .... 183 B 11 ERC O&M vs. traffic ................................ ................................ ................................ ....... 186 B 12 ERC cost vs. accumulated traffic ................................ ................................ ..................... 189 B 13 I 95 revenue vs. traffic ................................ ................................ ................................ ..... 200 B 14 I 95 O&M vs. traffic ................................ ................................ ................................ ........ 203 B 15 I 95 cost vs. accumulated traffic ................................ ................................ ...................... 206 D 1 Year Cumulative Probability of Default Rates ................................ ................................ ............................. 272 E 1 Effects of optimism bias on I 495 equity IRR (%) ................................ .......................... 274 E 2 Effects of optimism bia s on LBJ equity IRR (%) ................................ ............................ 274 E 3 Effects of optimism bias on NTE equity IRR (%) ................................ ........................... 275 E 4 Effects of optimism bias on ERC equity IR R (%) ................................ ........................... 275 E 5 Effects of optimism bias on I 95 equity IRR (%) ................................ ............................ 276

PAGE 15

15 LIST OF ABBREVIATIONS 1 BOT Build Transfer Operate is a variation on the design build operate maintain (DBOM) project delivery structure in which the private contractor transfers ownership to the public sponsor after construction is completed, and then is authorized to operate the facility for a period of time. This model also includes some privat e financing of the design, construction, operation and maintenance of a facility. BTO is similar to design build finance operate (DBFO) and build (own) operate transfer (BOT or BOOT). DB Design Build contracting method is structured so that a single entit y is responsible for both the design and construction of a project and both procurements are combined into one fixed fee contract. Potential benefits can include time savings, cost savings, risk sharing and quality improvement. A variation is design build with a warranty (also known as design build warrant), in which a contractor guarantees to meet material workmanship and/or performance measures for a specified period of time after project delivery. Design build is sometimes considered the PPP approach wit h the least private involvement; others have excluded it from Thirty eight states and Puerto Rico have design build enabling legislation, whereas fewer have authorized other PPP mode ls for highway projects DBB Design Bid Build is the traditional procurement approach for transportation projects in the United States, in which the design and construction of a facility are sequential steps in the project development process and each act ivity is bid separately. This is not a PPP. DBFOM Design Build Finance Operate (maintain/manage models are variations on the design build operate maintain PPP project delivery structure that also include some private financing of the design, construction, operation and/or maintenance of a facility. Under a DBFO or DBFOM, the public sponsor retains ownership of the facility and uses revenues generated from operation of the facility (such as tolls) to repay the private and other financing used to construct i t. These approaches may include an up front payment to the public sector agency or a revenue sharing agreement (see ansfer of financial risk to the private contractor. These models are similar to build operate transfer (BOT) and build transfer operate (BTO). 1 List of Abbreviations definitions are provided by the National Conference of State Legislatures report Public Private Partnerships for Transportation: A Toolkit for Legislators An update to the report was released on Jan. 25, 2013.

PAGE 16

16 DBOM Design Build Operate Maintain project delivery contracts are structured so that the private contractor is responsible for the design and construction of a facility, as well as its operations and maintenance for a specified period of time after construction. One potential benefit is the incentive to a private contractor to deliver a higher quality product in or der to avoid higher maintenance and improvement costs during the operations phase. DOT U.S. Department of Transportation FHWA The Federal Highway Administration is an agency within the U.S. Department of Transportation that supports State and local gove rnments in (Federal Aid Highway Program) and various federally and tribal owned lands (Federal Lands Highway Program). Through financial and technical assistance to State and local go vernments, the Federal Highway highways continue to be among the safest and most technologically sound in the world.( https://www.fhwa.dot.gov/about/) FDOT Florida Department of Transport ation is a decentralized agency charged with the establishment, maintenance, and regulation of public transportation in the state of Florida. (http://www.flsenate.gov/Laws/Statutes/2012/Title26/#Title26) O&M Under an d Operations and Maintenance contract, a selected contractor is responsible for operating and maintaining a facility for a specified time. PAB Private Activity Bond is an innovative financing tool that can be used for transportation PPPs, public activity bonds are a form of tax exempt bond financing that can be issued by or on behalf of state or local governments for privately developed and operated projects. This gives private entities access to tax exempt interest rates. All Transportation Infrastructure Finance and Innovation Act of 1998 (TIFIA) projects are eligible for PABs. Under current law, the total amount of such bonds is limited to $15 billion. As of January 2010, PAB allocations totaled $6.3 billion for seven projects. Private Activity Bonds PP P public and private sector partners, which allow more private sector participation than is traditional. The agreements usually involve a government agency contracting with a private c ompany to renovate, construct, operate, maintain, and/or manage a facility or system. While the public sector usually retains ownership in the facility or system, the private party will be given additional decision rights in determining how the project or may also finance some or all of a project.

PAGE 17

17 TIFIA Transportation Infrastructure Finance and Innovation Act of 1998 An innovative financing tool that can be used for transportation PPPs, the Transp ortation Infrastructure Finance and Innovation Act of 1998 provides federal credit assistance in the form of direct loans, loan guarantees or standby lines of credit to public or private sponsors of major surface by attracting substantial private and other non federal co investment in Various criteria must be met to qualify for TIFIA assistance, and only 33 percent of eligible project costs can be supported. Congress authorized $122 million per year for TIFIA for FY 2005 through FY 2009, which can support on average more than $2 billion of annual credit assistance. From its inception to July 2010, the program provide d $7.9 billion in assistance for projects worth $29.4 billion total

PAGE 18

18 Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy I NVESTMENT RISK BASED ON TRAFFIC FORECASTING ACCURACY: CASE STUDIES OF U.S. HIGHWAY PUBLIC PRIVATE PARTNERSHIPS By Kyle Fisher December 2013 Chair: Ralph Ellis Major: Civil Engineering The scarcity of public funding, coupled with crumbling infrastructur e, and demands for infrastructure improvements has l ed to increased usage of public private partnerships (PPP) as an alternative delivery method for surface transportation infrastructure projects in the United States PPPs however have a somewhat mixed tr ack record; an unusually high number of PPPs have defaulted on debt due to lower than expected traffic and revenue. These observations le d t o t his study to investigate the potential losses to debt and equity due to the inaccuracies in traffic and revenue forecasting. Building on the traffic forecast research of Dr. Bent Flyvb j erg and Dr. Robert Bain, a top down approach was taken to investigate the problem. F ive U.S. highway PPP project s were selected for investigation based on project type, location and availability of financial data. Financial data from Private Activity Bond transcripts was used to construct Simulated Cash Flow M odels tha cash flow model s and account ed for funds during the project term. Next a Traffic D istribution Instrument was created to vary traffic forecast s to a user specified confidence interval on a normal probability distribution function These new traffic levels were entered into the Simulated Cash Flow Model s and the ir impacts were observed on equity IRR and d ebt default The effect s of

PAGE 19

19 optimism bias in traffic forecasting were inv estigated by creating two different traffic forecasting accuracy distributions, one with no optimism bias (u=1.00, sd=0.34 ) and one with opti mism bias ( u=0.77, sd=0.26) It was found that the risk of default on debt in PPP projects is three times greater than would be anticipated by a projects BBB credit rating The e quity IRR and equity risk premium for the base case scenario s is consistent wi th other investments but returns are at serious risk when accounting for the volatility of traffic forecasting. It was also shown that concessionaires do not accurately price the cost of traffic revenue risk. From these results it was concluded that t here is a general underestimation of risk in direct toll PPP projects by both debt investors and equity participants. It was not possible to conclude that optimism bias is present or the cause of the unusually high default rate in PPP projects; however i t was possible to demonstrate that if present, optimism bias w ould be detrimental to returns on investments in the selected cases.

PAGE 20

20 CHAPTER 1 INTRODUCTION Traditionally the government has built, financed, and maintained roads and highways in the United Sta tes F ollowing the financial crisis of 2008 however, many government infrastructure providers found their funding drastically reduced and their budgets inadequate to meet the growing infrastructure demands of their communities The scarcity of public fu nding, coupled with crumbling infrastructure, which received a D grade (poor) by the American Society of Civil Engineers and need for higher capacity or infrastructure improvements, sent infrastructure providers scurrying for creative solutions (ASCE, 2013 ). Public Private Partnerships (PPPs) are a financing mechanism that has been used around the world to build infrastructure in e conomies where the government is unable to raise fun ding for necessary project s Over the past 10 years PPPs have also emerge d in the U.S. as a significant innovative financing a lternative for funding infrastructure projects. PPPs however have a somewhat mixed track record; an unusually high number of PPPs have failed and defaulted on debt due to lower than expected traffic a nd revenue. Recent research finds that 1) that the ability to forecast traffic is poor and 2) competitive procurement mechanisms can lead to overestimated traffic, i.e. optimism bias which may explain these failures This study examines the potential de fault risk in five surface transportation PPP projects in the U.S Background on Public Private Partnerships private sectors, built on the expertise of each partner that bes t meets clearly defined public needs (Canadian Council for Public Private Partnerships, 2001) PPPs have been applied to procure a wide range of services

PAGE 21

21 including education, health car e, prison operations, and infrastructure. Within infrastructure, the application of PPPs runs an equally broad gamut with PPPs used in telecommunications, water/waste water, railways, port improvements, tunnels and highways. For the purpose of this study we will narrow the definition of a PPP to refer to a surface transportation infrastructure p roject in the U.S. (PPP project ). A long term contract, often called a comprehensive development agreement or comprehensive agreement, is used to outline the resp onsibilities of the public and private sector in the PPP project project; contributes a portion of the financial, managerial and technical resources needed to execute and sometimes operates the project in acco advantage, and; partially shoulders the risks associated with the project and obtains the benefits those expected by each partner (Public Private Infrastructure Advisory Facility, 2009) Often these long term contracts range from twenty five to fifty years and may even extend up to 99 years. PPP Project Delivery Types A brief review of common PPP project types helps provide context for the subsequent discussions. PPP projects ma y be categorized by project delivery type; who will ultimately own or operate the asset, public or private; revenue stream; or the method of compensating the p rivate sector for its services. Table 1 1 lists common PPP arrangements but should by no means b e considered inclusive. Table 1 1. Common PPP d esignations Types of PPP c ontracts esponsibilities DB Design Build O & M Operations Maintenance DBO Design Build Operate DBF Design Build Finance DBFO Design Build Finance Operate BOT Build Operate Transfer

PAGE 22

22 Table 1 1. Continued Types of PPP c ontracts esponsibilities DBFOM Design Build Finance Operate Maintain BOOT Build Own Operate Transfer BOO Build Own Operate The acronym designations in Table 1 1 indicate th e services that are to be delivered by the private sector; hence in a DB (Design Build), the private sector is responsible for the design and construction of a project, while the public sector is responsible for planning, finance and all ongo ing operations and maintenance. A sharp delineation does not exist between all forms of PPP; often the name is a result of the legal framework in which the PPP is procured and lends itself to local colloquialism. M. More detailed definitions of PPP project types are listed on the list of abbreviations page. Figure 1 1 PPP continuum In traditional DBB (Design Bid Build) infrastructure procurement, the public sector procures assets from the private secto r in ord er to deliver services. In PPP projects, the public sector is concerned with procuring services, not assets. When the onus of owning and operating the asset is removed from the public sector, a wide range of PPP types are available. As shown in Figure 1 1, a continuum exists with the private sector gradually taking on more responsibility from the public sector. DB (Design Build) and O & M (Operations Maintenance) projects are the

PAGE 23

23 most common form of PPP projects in the U.S. and have the lowest level of pr ivate sector involvement, with no private sector responsibility for assets. On the opposite end of the spectrum, immediately before privatization are full concession agreements such as BOO (Build Own Operate) where the private sector bears significant ri sk for operations and revenue. Private Finance Another major difference among PPP delivery types concerns which party, public or private, provides project financing. In DB (Design Build) and DBO (Design Build Operate) projects, the private sector does no t provide financing. Private sector project financing such as in a DBFOM (Design Build Finance Operate Maintain), can be a major benefit to the public sector. This method may allow the public sector to reduce the time necessary to complete the project by circumventing the political processes and also decreases the time and difficult process of approving massive budgets. Given the many types of PPPs, this study will only include projects with the following characteristics: the project includes a constructi on phase, the project is privately financed, and the project includes an operation and maintenance period provided by the private sector; essentially a DBFOM. In a DBFOM, the government entity, normally referred to as the a request for pro posals from the private sector to develop, finance, construct, operate and manage a project. Normally companies form teams consisting of contractors, engineers, operators and financiers to develop their proposal. The winning team is referred to as the anywhere from 15 to 99 years. During the term of the concession, the concessionaire will recoup their investment by either collecting user fees (direct tolls) or availability payments from the government. It is up to the concessionaire to assess the potential long term risks and returns from the deal, a lengthy and complex task.

PAGE 24

24 PPP projects are further differentiated by two principal types of revenue streams; direct tolls or availabil ity payments. It should be noted that PPPs can offer alternative financing methods or financing mechanisms, but, in the long term do not provid e new money for infrastructure. Revenues to repay the private investment must come from the same public sources as traditional projects: tolls, fees or taxes. Thus, state officials will consider PPPs in terms of whether they provide better value or overall public service than could otherwise be purchased by those revenue streams (Rall, Reed, & Farber 2010). In a direct toll method, the concessionaire is compensated for providing the service by user fees; this is the traditional concept of a toll road and the mos t common method used worldwide. The concessionaire raises debt and contributes equity to pay for the c onstruction, operations and management of the project during the concession period. The vast majority of traffic revenue risk is borne by the concessionaire who is entitled to keep all cash flows after the debts are serviced. If cash flows are insuffici ent to provide returns on equity or service the debt, investors may lose money. For this reason, when using the direct toll method, accurately assessing traffic demand is a critical concern for the concessionaire. If the projected toll revenues on a proj ect are not sufficient to justify building the facility but the government still desires the project built (for social or political reasons), minimum revenue guarantees can be granted by the state to be used along with the direct toll and make a project vi able. With most highway projects being marginally viable at best, it has become common for governments (outside the U.S.) to grant minimum revenue guarantees to concessionaires. This arrangement has the benefit of being able to share demand risk between t he government and the concessionaire (Endel, Fischer, & Galetovic, 2010).

PAGE 25

25 In the availability payment method, the concessionaire is compensated through direct payments from the state; these payments may be fixed or variable, based on le vel of use (e.g. sha dow tolls). In this method, the concessionaire finances, designs, constructs and operates the facility for a number of years but the concessionaire does not collect tolls. The concessionaire is compensated by the government on metrics established in the contract; this is usually based on the concessionaire maintaining the road in good condit ion and limiting lane closures. The government may or may not toll users of the road but these tolls are not linked to th e As can be see n, in this format the government bears the majority of demand risk. P ublic Private Partnership s in the U.S. In U.S. history, tolls and taxes were the traditional methods for supporting transportation infrastructure. Dating back to the eighteenth century, many of the first roadways were developed by private entities which collect ed user fees to fund the roads. In 1956, the establishment of the Interstate Highway System concluded private management of transportation infrastructure in the U.S. and transferre d the financing, maintenance, and management over to state highway agencies (FHWA, 2007) Over the past twenty years however, transportation infrastructure agencies have been seekin g innovative financing options. As the U.S. transportation system matured the gap between infrastructure requirements ability to fund projects grew. Both state and federal offices have faced rapidly increasing demand, deterioration of facilities, and insufficient resources to maintain the facilities (Zha ng & Kumaraswamy, 2001) (Zhang X., 2004). To meet these challenges, governm ents have experimented with PPPs. The first U.S. PPP project, SR 91 Express Lanes, was a $130 million privately financed 10 mile, four lane toll project located between State Rou te 55 in Orange County CA and the Riverside County line.

PAGE 26

26 The project opened in 1995 and immediately experienced higher than expected traffic. A non compete clause in the concession agreement prevented the public agency from opening a competing faculty to alleviate congestion so in 2003 the Orange County Transportation Authority purchased the toll road for $207.5 million. Since the first projec t in 1993, PPP has grown steadily in the U.S.; t he Public Works Financing newsl etter report U.S. Transportation DBFOM Concessions 1993 2013 summarizes the projects as listed in Table 1 2; t his table does not include rail projects or leases 2 In the firs t ten years, 1993 2003, four highway projects were procured with an average value of $154 million in real dollars. In the following ten years, 2003 2013, eleven projects were procured with an average project value of $1,386 million in real dollars As to second decade the number of PPP deals tripled and the average d eal size increasing nine fold. While PPP is still maturing, it is off to a shaky start in the U.S. Of the seven completed projects, two have entered bankruptcy and one is expected to default on debt due to lower than expected traffic Conversely, one proj ect, SR 91 Express Lanes, had to be purchased by the state due to higher than expected traffic. Table 1 2 U.S. transportation DBFOM c oncessions 1993 2013 Date Project n ame Public s ponsor Risk t ype Cost 2013$ Developer Status In Operation (as of September 2012) 07/1993 91 Express Lanes, CA Caltrans DBFOM (toll) 130 Level 3 /Cofiroute/Granite Jan 2003, issued more bonds 09/1993 Dulles Gree nway, VA Virginia DOT DBFOM (toll) 350 TRIPII ($150m/Brown & Root) 05/1999 Foley Beach Express, AL City of Foley, AL BOO (toll) 44 Baldwin County Bridge Co. 10/2000 Camino Colombia Bypass, TX Texas DOT BOO (toll) 90 Landowners (Granite) Bankrupt TX DO T purchased Jan 2004 2 Source: http://www.pwfinance. net/document/research_reprints/Charts%201013/History.pdf

PAGE 27

27 Table 1 2. Continued Date Project n ame Public s ponsor Risk t ype Cost 2013$ Developer Status In Operation (as of September 2012) 01/2005 SR 125 So. Bay Express, CA* Caltrans DBFOM (toll) 773 PBI/Macquarie ($653m/Fluor Washington) Bankrupt 12/2007 I 495 HOT Lanes, VA* Virginia DOT DBFOM (toll) 1,998 Transurban/Fluor ($1.4bn/Fluor Lane) 03/2008 SH 130 segments 5 6, TX* Texas DOT DBFOM (toll) 1,358 Cintra/Zachary($968 Ferrovial Zachary) Bonds downgraded, restructured?? Under Const ruction (as of September 2012) 02/2009 I 595 Managed Lanes, FL* Florida DOT DBFOM (ap) 1,814 ACS Infrastructure ($1.2 Dragados EarthTech.) 10/2009 Port of Miami Tunnel, FL* Florida DOT DBFOM (ap) 914 Meridiam ($607m Bouygues Jacobs) 12/2009 North Tarr ant Express, TX* Texas DOT DBFOM (toll) 2,047 Cintra/Meridiam($1.6 bn/Ferrovial) 06/2010 I 635 LBJ Managed Lanes, TX* Texas DOT DBFOM (toll) 2,800 Cintra/Meridiam ($2.1 bn/ Ferrovial Agroman) 01/2011 Jordan Bridge, VA Chesapeak e, VA BOO (toll) 140 Figg /Amer. Infra. MLP/Lane ($100m Lane) 04/2012 Midtown Tunnel, VA* Virginia DOT DBFOM (toll) 2,100 Skanska/Macquarie ($1.47 bn Skanska Kiewit Weeks 06/2012 Presidio Parkway, CA* Caltrans DBFOM (ap) 365 Hochtief/Meridiam($24 5m Flatiron Kiewit) 08/2012 I 95 Express Lanes, VA* Virginia DOT 940 Transurban/Fluor ($618m Fluor/Lane) Note: projects used TIFIA funding Increasing Demand The demand for transportation infrastructure is highly correlated wi economic growth. Over the past 30 years, the U.S. experienced rapid growth and increased ownership in automobiles; Americans rely almost exclusively on i ndependent automobile mobility. According to the U.S. Department of Transportation, from 1990 to 2009, private automobile travel accounted for 88% of travel in the U.S., a ir travel accounted for 8%, and public transit accounted for 1% (TRIP, 2012) Accordingly, the U.S. highway system has also experienced exponential growth; from 1990 2010 vehicle travel on U.S. highways increased 38% compared to the 4% increase in ne w road s (TRIP, 2012)

PAGE 28

28 Deteriorating Facilities Numerous reports and analyses of U.S. infrastructure have found the average condition of U.S. transportation infrastructure to be in poor condit ion and in need of maintenance. The Amer ican Society of Civil Engineers report U.S. 2013 Infrastructure Report Card is a comprehensive analysis of national studies, reports and the contributions from thousands of experienced civil engineers working around the country. The report indicated in vestments in surface transportation infrastructur e in the U.S. are insufficient. Furthermore, underfunding maintenance and repairs increases th e costs of future improvements. In the report, ASCE assigns grades to U.S. facilities according to the criteria indicated in Table 1 3 3 Table 1 3 ASCE grading s cale Grade Description Status A Exceptional : f it for the future The infrastructure in the system or network is generally in excellent condition, typically new or recently rehabilitated, and meets capaci ty needs for the future. A few elements show signs of general deterioration that require attention. Facilities meet modern standards for functionality and resilient to withstand most disasters and severe weather events. B Good: adequate for now The infras tructure in the system or network is in good to excellent condition; some elements show signs of general deterioration that require attention. A few elements exhibit significant deficiencies. Safe and reliable with minimal capacity issues and minimal risk. C Mediocre: requires attention The infrastructure in the system or network is in fair to good condition; it shows general signs of deterioration and requires attention. Some elements exhibit significant deficiencies in conditions and functionality, with increasing vulnerability to risk. D Poor: at risk The infrastructure is in poor to fair condition and mostly below standard, with many elements approaching the end of their service life. A large portion of the system exhibits significant deterioration. Co ndition and capacity are of significant concern with strong risk of failure. 3 Source: http://www.infrastructurereportcard.org/a/documents/2013 Report Card.pdf

PAGE 29

29 Table 1 3. Continued Grade Description Status F Failing/critical: unfit for purpose The infrastructure in the system is in unacceptable condition with widespread advanced sig ns of deterioration. Many of the components of the system exhibit signs of imminent failure. ASCE assigned a grade D to 4 million miles of public roads. These public roads provided access to ports, rail terminals, city centers and enabled the movement o f Americans and their goods throughout the country. Bridges received a grade C+; of the 607,380 bridges assessed, one in nine bridges is deemed structurally deficient and the average age is 42 years (ASCE, 2013) TRIP, a national transportation research group, also collected and analyzed data from the U.S. Census, the U.S. Department of Transportation, the Federal Highway Administration, the Bureau of Transportation Statistics, the National Highway Traffic Safety Administration and the Texas Transportatio n Institute, in their 2012 report, Transportation System and Federal Funding This report further emphasized the poor condition of U.S. transportation infrastructure and the impacts on users and the economy 4 : Thirty tw in poor or mediocre condition. Driving on roads in need of repair costs U.S. motorists $67 billion a year in extra vehicle repairs and operating costs $324 per motorist. Traffic Congestion Forty four percent of r urban highways are congested. Traffic congestion costs American motorists $101 billion a year in wasted time and fuel costs. The average U.S. commuter loses 34 hours each year due to traffic congestion. Twenty idges are structurally deficient or functionally obsolete Roadway conditions are a significant factor in approximately o ne third of traffic fatalities. There were 32,885 traffic fatalities in 2010 in the U.S. A total of 187,921 people died on U.S. highwa ys from 2006 through 2010. 4 Source: http://www.tripnet.org/docs/TRIP_National_Fact_Sheet_April_2012.pdf

PAGE 30

30 These reports reveal that many roads, tunnels and bridges, were constructed more than 50 years ago and these systems are simply overwhelmed or crumbling and in need of repair and or expansion (TRIP, 2012) Declining Resources Re searchers and policymakers have identified several specific concerns about obligations resulting in steadily accumulating shortfalls, deterioration of the user fee model as a result of accumulating shortfalls, declining revenues from existing sources, inadequate funding to meet growth, and increased uncertainty and inability to make long term plans as a result of temporary extensions on transportation legislation (Co ngressional Budget Office, 2010) Over the last 50 years, Federal transportation infrastructure spending has steadily declined by 0.5% per year; most of the decline is accounted for by decreased federal spending on transport ation infrastructure projects. As a result, the responsibility for states and local government to provide funding for transportation infrastructure projects has increased from 68%, from 1960 1980, to 75% from 1990 to present day (Congressional Budget Office, 2010). In 2013, f ederal, s tate, and local governments spent $91 billion a nnually on capital investments. Estimates in the ASCE report indicated that maintaining roads at the current grade D level would require $101 billion annually between 2008 and 2028. If the current trend cont inues, the funding gap of 48% estimated in 2010 will increase to 54% by 2040 and roads will continue to deteriorate (ASCE, 2013) Advantages and Disadvantages of PPPs The discussion of the potential advantages and disadvantages in using PPPs to build infr astructure is a hot top ic today in the United States. Research indicates that infrastructure funds provide investment opportunities and help to meet the infrastructure demand needs which

PAGE 31

31 the p ublic sector was unable to do. Analysts and stakeholders have identified potential advantages and disadvantages of the collaborative efforts in PPP projects noted in the lists below (Rall, Reed, & Farber 2010) (Kwak, Chih, & Ibbs, 2009) (Cervero, 2004) (PWC, 2005): Potential Advantages of PPPs Acceleration of finan cing projects without the use of public funding, reducing public sector budgets Transfer risks such as construction, finance, and operation of projects, to the private sector Increased trans parency of project costs Improved quality and efficiency of infras tructure services. With PPPs, risks are allocated to the party best able to manage or absorb each particular risk Increases mobility through tolling, congestion pricing, and more efficient decision making; Monetization of existing assets Lifecycle efficie ncies Reduced traffic congestion (better for environment) Increased ridership and fare revenues Higher land values Potential Disadvantages of PPP s Transportation Infrastructure PPPs are relatively new and few in the USA, public and private sector lack kn owled ge and skills of implementation Limited Competition high cost, high levels of experien ce and security required to bid financial management Monopolization of road usage Potentially higher tolls under private operation Potential loss of control: non compete provisions, and toll rate setting Higher Public Sector Costs: cost of advisors, cost of private finance, and potential tax losses

PAGE 32

32 Public could pay tolls that are higher than tolls based on the costs of the facilities, including a reasonable rate of return should a private concessionaire take advantage of market power gained by control of a road for which there are few alternatives that do not require substant ially more travel time PPPs provide a number of potential advantages and disadvantages for both the public and private sectors making this delivery method complicated and a highly polit ic al matter. However, research indicates i f properly formulated and ma naged, the public and private sectors can realize the benefi ts and avoid the disadvantages (PriceWaterhouseCoopers, 2005) PPP Default Today more than 1,374 PPP projects are located all around the world excluding the U.S. (World Bank, 2011) While PPPs provide a great number of advantages, many have failed. More than 77 projects are currently canceled or under distress, accounting for 5.6% of the total projects around the world; 53.2% are from South America, 29.9% are from East Asia and Pacific, and t he remaining parts of the world, excluding the U.S., account for the final 16.8% (World Bank, 2011) Reasons why PPP projects fail are specific to the individual project but the World Bank presents several recurring themes, specifically highlighting unrealis tic traffic forecasts, and undefined public contribution of funds for a project as the main reasons for PPP failures (The World Bank, 2008) Most PPP failures can be attributed to inadequate or non existent feasibility studies, including unrealistic traff ic forecasts and undefined public contribution of funds. Other common reasons for failure include: p oor legal framework and enforcement; w eak institutional capacity and PPP strategy; u nrealistic revenue and cost estimations; l ack of thorough financial and economic analysis; i nappropriate sharing of risks; l ack of competitive procurement; p ublic resistance (willingness to pay not correctly assessed) Anecdotal evidence from around the world also reflects these reasons for PPP defaults. Major defaults from around the world include the infamous failures of PPP toll roads in Spain,

PAGE 33

33 city Tunnel being financially distressed with less than half of the predicted demand (Chan, Lam Chan & Cheu ng, 2008) Between 1987 and 1995 Mexico embarked on one of the largest PPP toll road programs to date with 52 projects awarded. Unfortunately, the program was plagued with highly ed in the form (World Bank Institute, 2012) In addition, short concession periods led to high tolls and revenues 30% below forecasts. The Mexican government ultimately paid US$7.6 billion to take control of 23 projects (Th e World Bank, 2008). The younger PPP market in U.S. has also exper ienced default on PPP projects. Compared to the worldwide defau lt rate of 5.6%, the U.S. failure rate is 57 %, including SR 91 Express Lanes and SH 130 High profile failures and distresse d or renegotiated failing projects have raised investor concerns about state programs selecting viable projects with qualified partners. In the U.S., five PPP highway projects were completed between 1993 and 2003 Table 1 3; only two of these, Dulles Green way and Foley Beach Ex press, are still in operation. Camino Colombia Bypass and SR 125 South Bay Express both filed bankruptcy due to lower than expected traffic which resulted in revenues that were insufficient to service the debt. More recently, I 495 HOT lanes in Virginia opened in November of 2012 and actual traffic was less than 50% of the forecast for the first two months of operation ; t he long term viability of this project is not yet known. In Texas, SH 130 Segments 5 6 opened in October 2012 but as a result of lower than expected demand the project is now in distress and is undergoing renegotiations and may be purchased by the State. The bonds on the project were recently downgraded to junk bond status by Fitch Ratings ; R ating Servic e is expecting the project to default in 2014.

PAGE 34

34 Purpose of the Study The purpose of this study is to promote a greater understanding of the potential losses to debt and equity due to the inaccuracies of traffic and revenue forecasting and identify the pro bability of default in PPP projects. With only seven PPP highway, toll road, bridge or tunnel projects in operation in the U.S. and eight projects under construction, it is not possible to make a meaningful statistical analysis of investment risk in these projects. For this reason, a case study review of investment risk in five PPP projects was undertaken. The research outline, depicted in Figure 1 2 was designed to meet the objectives of this study. The problem formation and research objective marked t he first phase of the research process. After identifying an objective and establishing the scope of the research, a literature review was conducted to establish the current state of research in the topic. The researcher also familiarized himself with PP P projects around the U.S.; from these projects, case studies were selected based on availability of data, similarity of characteristics, and to ensure a variety of locations. All available data such as financial and legal documents were collected f or eac h of the selected cases. Using E were designed to analyze data f rom each of the selected cases. These models investigate traffic revenue risk and evaluate the likelihood of default base d on varying traffic levels; the results were then compared to each other, other highway PPP projects and the likelihood of default as indicated by the projects credit ratings.

PAGE 35

35 Figure 1 2 Research s tructure Research Questions 1. How accurately do bond rat ings for PPP projects reflect risk based on traffic demand? It is hypothesized that bond ratings accurately reflect the underlyin g default risk based on traffic demand assuming no optimism bias in forecasting. 2. If optimism bias is assumed to be present i n traffic forecasting, how accurately do bond ratings for PPP projects reflect risk based on traffic demand? It is hypothesized that bond ratings do not accurately reflect the underlying default risk based on traffic demand if optimism bias exists in fo recasting. 3. What is the range and average base case risk premium a concessionaire is willing to acc ept to undertake a PPP project? It is hypothesized that all projects will demand a similar risk premium for returns to equity and that the base case equity r isk premium will be 5% 4. How do the case study projects compare in terms of returns and risk? It is hypothesized that the projects will be similar in terms of return and risk. 5. If optimism bias is assumed to be present in traffic forecasting what is the ba se case equity risk premium a concessionaire is willing to accept to undertake a PPP project? It is hypothesized that if optimism bias exi s ts in PPP traffic forecasting the base case equity risk premium will be near 0%.

PAGE 36

36 6. As risk are trans ferred from the public to private partner (availability payment to direct tolling), does the equity risk premium increase? It is hypothesized that the equity risk premium associated with direct toll projects will be larger than the equity risk premium ass ociated with availability payment projects. 7. Do projects with a higher potential for default require a higher equity risk premium? It is hypothesized that as the probability of default increase s the equity risk premium required by the concessionaire will a lso increase.

PAGE 37

37 CHAPTER 2 LITERATURE REVIEW To investigate investment risk based on traffic forecasting accuracy, an extensive literature review was conducted to gain a comprehensive understanding of the procurement, financing, and operation of transportat i on PPPs. To understand how PPP cash flow models are constructed and function required a review of several different disciplines, which at first may appear not to be connected but they in fact rely upon each other to finance a project. The literature revi ew covers five main research threads: infrastructure funding legislation, project finance structure, risk in PPP projects, traffic and revenue forecasting and credit ratings. More specifically, this review introduces the funding legislation that enabled transportation infrastructure to be built using the PPP delivery method and then outlines the basic structure of a PPP deal. Next follows an explanation about how risks are dealt with in PPPs, which can be significantly different than in a typical highwa y project. Traffic demand risk is elaborated on in order to discuss traffic and revenue forecasting which is a method used to quantify and manage traffic revenue risk. Next, the research of Flyvbjerg and Bain is presented and optimism bias is i ntroduced. Finally a discussion of credit ratings follows as an indicator to investors of risk in PPP projects. Funding Legislation for Public Private Partnerships Given the growing funding gap for transportation infrastructure, both f ederal and state governments are challenged with how to fund and manage the tran sportation system in the U.S. The following brief summaries of U S transportation funding legislation are provided to help explain the financing structures of PPP projects in the study. The general tim eline and description of the legislation explained in this section is displayed in Figure 2 1.

PAGE 38

38 Figure 2 1 PPP Legislation t imeline Federal Highway Trust Fund The Federal Highway Trust Fund is the largest federal funding source for highway infrastructure ; the fund was established in 1956 and funded by fuel taxes to pay for highway construct ion, maintenance and expansion. In recent years, the federal trust fund has been unable to meet its obligations. The federal f uel tax was last increased in 1993 to 18 .4 cents per gallon; this tax was not indexed to inflation and has steadily decreased in real purchasing power since 1993 ( Federal H ighway Administration, 2013) As a result, the Congressional Budget Office estimated the Highway Trust Fund fuel tax reven ues will no longer be able to meet the demand beginning in 2015, resulting in steadily accumulating shortfalls (See Figure 2 2 ) (Congressional Budget Office, 2010) Between 2007 and 2010, Congress transferred more than $60 billion, or 30% of the required balance, in additional funds from the general fund in order to sustain the

PAGE 39

39 account (Senate Finance Committee, 2013). The Congressional Budget Office depicted the severity of future Highway Trust Fund shortfalls in Figure 2 2 5 Note: This projection assumes no further appropriations after 2014 from general revenues to the Highway Trust Fund. Figure 2 2 Cash flow of the highway account of the highway trust fund Transportation Infrastructure Finance and Innovation Act ( TIFIA ) In 1998, the Transportati on Infrastructure Finance and Innovation Act (TIFIA), w as established to provide a f ederal credit program for transportation projects. The TIFIA program aimed to attract investment capital for revenue generating transportation projects by offering three fo rms of credit assistance (low cost funding) to qualifying programs direct loans, loan limited seed money in order to attract other private investment and leverage f ederal funds. Currently each dollar of federal funds can provide up to $10 in TIFIA credit assistance and support up to $30 in transport ation infrastructure investment (Federal Highway Administration, 2013) To qualify for TIFIA assistance, a project mus t meet the following criteria: 5 Source: http://www.cbo.gov/publication/43985

PAGE 40

40 Minimum project costs of $50 million Federal funding cannot exceed 33% of project costs (until 2014 changes effective) Senior debt obligations must receive an investment grade rating The project must be able to pledge a dedi cated revenue stream for repayment on the TIFIA loan The TIFIA program has been a critical tool for funding large infrastructure projects and in many cases the benefits of the very forgiving structure of TIFIA loans kept projects alive which would have gon e bankrupt in the real world (Reinhardt, 2012) The U.S. Department of Transportation lists several benefits of the TIFIA financing assistance: TIFIA assistance can provide a funding source wh en there is no market liquidity Improved access to capital mark ets Flexible repayment terms Favorable interest rate s equal to the 30 year U.S. Treasury securities Accelerated financial closing (specialization familiarity with funding for complex projects) Final maturities as long as 35 years after substantial complet ion of the project Rare flexibility in repayment, ability to re negotiate terms when in duress Reduced penalties To date, the use of TIFIA funding has sky rocketed along with a steady increasing budget, jumping from $122 million per year in 2005 2009, to $ 740 million per year in 2013, and $1 billion in 2014 (Kessler & Denton, 2012) (TIFIA, 1998) SAFETEA LU The Safe, Accountable, Flexible, Efficient Transportation Equity Act: A Legacy for Users (SAFETEA LU) law was enacted in 2005 and designed to provide guaranteed funding for transportation infrastructure totaling $244.1 billion, the largest surface transportation investment

PAGE 41

41 in U.S. history. SAFTEA original expiration date of 2009, until Jun e 30, 2012 (TRIP, 2012) SAFETEA played an important role in PPP development, the National Surface Transportation Policy and Revenue Commission created by SAFETEA LU recommended to Congress that PPPs should play an important role in financing and managing our surface transportation system. As a result, the U.S. Treasury authorized the issuance of up to $15 billion in tax exempt bonds for PPP type transportation projects (Federal Highway Administration, 2007). Private activity bonds (PABs) for infrastruct ure projects have contributed to the success of PPPs for surface transportation infrastructure projects acting as a cheap source of funding. Furthermore, the Committee on Finance recently suggested an increase of the current $15 billion cap on transportat ion PAB s to $19 billion in 2014 (Senate Finance Committee, 2013). Moving Ahead for America in the 21 st Century MAP 21 In 2012 Congress enacted Moving Ahead for America in the 21 st Century (MAP 21), which provided $105 billion for surface transportation p rograms in fiscal years 2013 and 2014. MAP 21 was created to a streamline surface transportation programs and built on many of the existing transportation programs (Federal Highway Administration, 2013). This program gave states the right to toll without the previously required federal t olling agreement and streamlined the PPP procurement process which made it easi er for states to undertake PPPs. S ome of the programs highlights are summarized below (Kessler & Denton, 2012). Permitted additional tolling l anes on interstates, as long as non tolled general purpose lanes are not reduced. States were permitted the ability to convert HOV lanes to express toll lanes to fund the reconstruction of the interstate, using the tolls revenues from new lanes, which rep resent only a fraction of the cost of reconstruction. States were permitted to finance new and replacement bridges and tunnels on interstates using toll revenues on all lanes

PAGE 42

42 The list of legislation enacted to provide funding programs for surface transpor tation infrastructure p rojects is extensive. Both federal and state governments are constantly working to find reliable and sustainable sources of revenues for funding transportation infrastructure. A complete u nderstanding of PPP legislation and its imp lications is not necessary to understand this study; however, the reader should be aware that federal legislation is enabling the growth of PPPs around the country. Project Finance Structure for PPP projects Yescombe (2002) clearly and comprehensively de fines the fundamentals of project finance: Project finance is a method of raising long term debt financing for major projects, based on lending against the cash flows generated by the project. The process can take many years and depends on detailed evaluat operating and revenue risks, and their allocation between investors lenders, and other parties through con tractual and other arrangements. In the U.S., the procurement process for surface transportation infrastructure PPP s lasts an average of 18 months. Following the announcement of the project, interested parties part ner together and establish a special purpose company to generate a proposal. This company will function as the borrower and act as legal authority if awa rd ed the project A multitude of factors affect the financing structure for a PPP project such as risk allocation, robustness of the contract, debt to equity ratios, and the use of state guarantees or grants. The general financing str ucture as depicted in Figure 2 3 and is explained in the following sections.

PAGE 43

43 Figure 2 3 Genera l U.S. PPP project financing structure Debt Debt for PPP projects may b e provided by various sources: p ublic funds ( revenue b onds ), PABs, f ederal loans (TIFIA, GARVEE), and privat e banks. The details of these de bt sources will be explained more in the following section. Debt is a cheaper source of funding than equity because it is less risky ; h owever, both debt and equity contributors spend a great deal of time and money to predic t the cash flows of the project and assess financial risks thoroughly. According to researchers, Estache, Trujillo, & Juan (2009) financial risks experienced in transportation infrastructure PPP projects tend to be related to the following factors: R elian ce on optimistic revenue assumptions and on levels of demand from a poorly chosen L ack of attention to financing ne eds in the project feasibility, which leads to lar ger amounts of debt in projects L ong term PPP projects that are financed wi th short term debt, coupled with a sometimes unjustified assumption that the short term debt can be rolled over at the same or even b etter refinancing conditions

PAGE 44

44 R eliance on optimistic revenue assumptions and on levels of demand from a poorly chosen Considering the financial risks involved in financing billion dollar projects with 35+ years long term schedules, creating the loan package for PPP projects may take years to complete; still, it is surpisingly faster than traditional method s bec ause it avoids the lenghty budgeting process for limited funds. Senior debt Senior d ebt receives priority in the repayment schedule ov er all other forms of finance. In the typical PPP financial structure, de bt accounts for 70 80% while e quity accounts fo r 20 30% of financing (European PPP Extertise Centre, 2013) Senior d ebt is less expensive than other forms of financing (except grants) which explains the high proportion of senior debt The two major types of senior debt currently used to finance PPPs are bank loans and PABs PA B s are issued on behalf of state or local governments to provide low cost (tax exempt) financing to projects with private investors. Since May 2013, PAB s have been issued for seven PPP projects totaling $3.8 billion as shown in Table 2 1. PABs have been used in the past to finance infrastructure but until SAFE TEA LU was enacted in 2005 transportation infrastructure was not eligible; this law limits the total amount of PABs to $15 billion. Once the request for the PABs for a P PP project receives approval from the U.S. Dep artment of Transportation and the bonds are sold, the c oncessionaire will be r esponsible for debt service on the PABs (Federal Highway Administration, 2013) The Federal Highway Administration publishes the li st of PAB supported projects in Table 2 1 : PABs Issued for PPP Projects in the U.S. Since May 2013 6 6 Source: www.fhwa.dot.gov/ipd/finance/tools_programs/federal_debt_financing/private_activity_bonds/index.htm#current

PAGE 45

45 Table 2 1 PABs i ssued for PPP p roject s in the U.S. s ince May 2013 Project bonds issued PAB allocations / bonds issued ($ in thousands) Capital Beltway HOT Lanes $589,000 North Tarrant Expressway, TX $400,000 IH 635 (LBJ Freeway), TX $615,000 Denver RTD Eagle Project (East Corridor& Gold Line) $397,835 CenterPoint Intermodal Center, Joliet, Illinois $150,000 CenterPoint Intermodal Center, Joliet, Ill inois $75,000 Downtown Tunnel/Midtown Tunnel, Norfolk, Virginia $675,004 I 95 HOT/HOV Project $252,648 East End Crossing, Ohio River Bridges $676,805 Total $3,831,292 Subordinate debt Subordinate (j unior) debt ranks below senior debt and above equ ity which means that subordinate debt is not repaid until the senior debt obligation has been met. This normally results in s ubordinate debt being riskier than senior debt, thus having a higher credit rating and higher yield (European PPP Extertise Centre 2013) Major types of subordinated debt include mezzanine loans and TIFIA loans. TIFIA loa ns are issued on behalf of the f ederal government and range from a minimum $15 million up to a maximum of 49% of anticipated eligible project costs. Interest ra tes on these loans begin accruing from the date of issuance and are specified to be 1 basis point (or 0.1%) above comparable treasury securities. TIFIA loans offer flexible repayment terms, interest payments can be capitalized up to five years from the da te of issuance; in addition, repayments on the loans are not required until after the substantial completion date to allow time for completing construction and ramp up of traffic (FHWA, 2013) Furthermore, repayment plans are divided into mandatory a nd s cheduled debt repayment, which allow s the concessionaire to commit to mandatory repayments and also allows flexibility in scheduled debt. If the mandatory payments are not made, the project is in default but can be renegotiated by the U.S. Secr etary of T ransportation Scheduled debt can be repaid any time

PAGE 46

46 within the total term of the loan, not to exceed 35 years (FHWA, 2013) These loans are specifically issued to be subordinate debt to fill a funding gap; however in the event of default, the TIFIA loa ns are par with senior debt. subordinated to those of senior lenders to the project except in the event of bankruptcy, insolvency, or liquidation of the obligor. In such an instance, the TIFIA lien would be on p ar (USDOT, 2013) Equity Project equity may come from a variety of sources including the concessionaire, construction contractors, shareholder s or subordinated debt, as well as specialist infrastructure fu nds, for example Macquarie Bank. Equity contributors seek higher returns on their funding contributions than debt holders because they bear higher risk; equity contributors are paid last in the flow of funds and last in the event of default. Therefore, r equired returns on equity are higher making this method of financing more expensive and usually onl y accounts for 20 30% of the project financing In some instances, revenue sharing is used to share windfall profits with the s tate and reduce the return on equity. Credit Enhancement Debt service coverage r atio s (DSCR) and reserve a ccounts provide credit enhancement in the financing structure for PPP projects. Credit enhancements are an important element of project financing because they provide security t o lenders and make the project more attractive to investors. As noted above, project finance lenders rely mainly on cash flows from the project, the revenues generated by the users. The European PPP Expertise Center 2010 report, Capital markets in PPP fi nan cing: Where we were and where are we going summarizes examples of credit enhancements in PPP which enable the following protections for lenders:

PAGE 47

47 Project performance guarantees detailed in the contract, by the spo nsors or other parties involved Abilit y to provide temporary liquidity assets to deal with specific risks i.e letters of credit Insurance against certain project related risks which can include construction risks, loss of revenue, third party liabilit y, environmental liability, etc Governmen t Revenue Guarantees Secured interests over all the project assets (including and especially all contracts) to enable the lenders to take remedial actions if the PPP c ompany has failed Controls over all cash i nflows and outflows of the PPP c ompany, usuall y defined by the required debt service coverage ratios (DSCR). This will ensure that senior de bt service always has priority Cash flow controls in the form of reserve accounts Risk in PPP Critical success f actors (CSFs) have been identified by researcher s to aid in designing PPP projects, and to develop methods to better manage the risk of default. CSFs are defined as competitive performance for the organi (Rockart, 1982) Although CSFs are unique to each project, researchers summarized a list of CSF studies and the findings of those studies from around the world. Kwak Chih & Ibbs (2009) study, Towards a Comprehensive Understanding of Public Priva te Partnerships for Infrastructure Development included a review of researchers and their narrowed lists of CSFs for PPP surface transportation infrastructure projects around the world as listed in Table 2 2 : Critical Success Factors Table 2 2 Critical success f actors Researcher name & purpose of study Critical success factors (CSFs) Akintola Akintoye : i dentified six CSFs 1. Detailed risk analysis 2. A ppropriate risk allocation 3. I ncentives for competition 4. P revention of project cost escalation 5. E ncouragement of innovation in project development 6. A ccurate forecasting of costs

PAGE 48

48 Table 2 2. Continued Researcher name & purpose of study Critical success factors (CSFs) Li et al. : i dentified eight CSFs ( listed in descending order of importance ) 1. A strong private co nsortium 2. A ppropriate risk allocation 3. A vailable financial market 4. C ommitment/responsibility of public/private sectors 5. T horough and realistic cost/benefit assessment 6. T echnical feasibility 7. A well organized public agency 8. G ood governance X.Q. Zhang : i dentified five main CSF aspects 1. E conomic viability 2. A ppropriate risk allocation via reliable contractual arrangements 3. S ound financial package reliable 4. C oncessionaire consortium with strong technical st rength 5. F avorable investment environment The World Bank : i dentif ied nine CSFs 1. Careful planning of PPP project 2. Solid revenue and cost estimations 3. User willingness to pay and communication plan 4. Extensive feasibility study with use of PPP experts 5. Compliance with contractual agreement 6. Strong legal and regulatory f ramewo rk 7. Strong Institutions with appropriate resources 8. Competitive and transparent procurement 9. Mitigation and flexibility in managing macro risks Qiao et. A l : i dentified eight CSFs 1. A ppropriate project identification 2. P olitical and economic situation 3. A ttractive financial package 4. A cceptable toll/tariff levels 5. R easonable risk allocation 6. S election of suitable subcontractors 7. M anagement control 8. Technology transfer As indicated, many researchers identified similar CFSs that affect the success of PPP projects. One f Risk allocation is the measurement and assignment of risks between parties involved in the PPP project, and the public and private sector participants, excluding the end users (Akintoye Edwards, Hardcastle, & Li, 2005) Risk allocation is not unique to PPP but it may be a more important factor in these projects than in typical construction because the roles of the participants are not well identified. If the involved participants bear portions of the risk, the parties are acting

PAGE 49

49 in a shared risk allocation mechanism (Akintoye, Edwards, Hardcastle, & Li, 2005) Although shared risk allocation mechanism is widely used around world, sharing risk does not reduce or eliminate the overall p roject risks as can be seen by the large number of PPP bankruptcies. The success of PPP projects largely depends on effectively allocating risks to the party who is best able to manage them (Loosemore, Raftery, Reilly, & Higgon, 2006) The European PPP E xpertise Centre writes: One of the fundamental trade offs in designing PPPs is therefore to strive for the right balance between risk allocation between the public and private sector, the risk allocation within the private sector consortium and the cost o f funding for the PPP Company. Traffic Revenue Risk Traffic revenue risk, identified by the World Bank as is a critical success factor of surface transportation infrastructure PPP projects. The World Bank goes even furt her to state that the majority of PPP failures can be attributed to inadequate or nonexistent feasibility studies including unrealistic traffic forecast and undefined public funds (The World Bank, 2008) The importance of traffic revenue risk and forecast ing accuracy was also highlighted in a risk perception survey of Indian P PP professionals performed by A.V. Thomas. In this survey traffic revenue risk was identified as the most significant risk factor in PPPs; the respondents however were split on who i s better suited to manage the risk, the public or private sector (Thomas, 2008) Traffic revenue risk is a somewhat new consideration to transportation projects because most transportation projects are financed by taxes and have no investors. While the d emand for the road is forecast in traffic and revenue studies, the volatility in the revenue stream to investors has not been a traditional consideration. For this reason the investment risk associated with demand in transportation projects is often left completely unknown.

PAGE 50

50 Traffic and Revenue Forecasting Forecasting traffic demand is essential for assessing financial feasibility and project approval. This is normally done through traffic and r evenue (T&R) studies which are documents containing the suppo rting traffic forecasts used in establishing surface transportation infrastructure projects. These reports are generated from travel demand models created to estimate the level of traffic demand over long periods of time. It is important to note that the traffic forecasting process is not standardized; there is no universal method. It is the responsibility of the concessionaires bidding on PPP projects to choose from various analytical methods in order to develop their business model and traffic and rev enue forecasts. of the art consensus exists among transportation researchers and (NCHRP, 2006) Figu re 2 3 is a representation of a basic four s tep process used in traffic modeling. This method has been used in the U.S. and around the world for decades; details of the process can be Demand and Revenue (NCHRP, 200 6) 7 The role of the private sector working in traffic forecasting has had both positive and nega tive effects on traffic forecasting Top financial advising firms around the world have produced traffic forecasts and financial models which, compared with t he government models, models. For example, U.S. government offices were not required to account for annual 7 Source: http://onlinepubs.trb.org/onlinepubs/nchrp/nchrp_syn_364.pdf

PAGE 51

51 Figure 2 4 Four steps used in traffic modelin g depreciation costs in their models which can lead to unrealistic traffic demand (NCHRP, 2006) On the negative side, the private sector is a machine built to squeeze every last drop of profit from a project. members and credit analysts are all aware of how frequently feasibility studies miss the mark, most of the projections are showing (Glazier, 2012). Furthermore there is n othing stopping private sector firms fro m inflating revenue estimates ho ping to please clients and win bids (a form of optimism bias) (Glazier, 2012). Unfortunately, identifying the problms or errors in traffic forecasting is difficult because of the nature of the financial confidentiality of t he models.

PAGE 52

52 The Flyvbjerg Distribution Dr. Bent Flyvbjerg was one of the first researchers to find statistically significant results for the comparison of actual traffic on the roads to the forecasted traffic in the proposal s, using a comprehensive data set in his 2005 study How (In)accurate Are Demand Forecasts in Public Works Projects?. This study is unique for the size of its data set; before the 2005 Flyvbjerg study, all comparisons of actual to forecasted road traffic were limited to small data samples such as the JP Morgan study which compared actual to forecast traffic in 14 U S toll projects (Morgan, 1997) rejected because of unclear or insufficient data quality (Flyvb je rg, Holm, & Buhl, Spring 2005 ) ; the remaining data set of 204 projects consisted of a sample from around the world of 183 road projects (170 highways, ten bridges, and three tunnels) and 27 rail projects. The projects originated in 14 countries including the U.S (Flyvbjerg, Holm, & Buhl, Spring 2005 ) In his study, Flyvbjerg found that traffic forecasting ability was poor; furthermore, traffic forecasts are wrong by a large margin. It was found that forecasting erro rs were at least +/ 20% in 50% of the analyzed projects, forecasting errors were at least +/ 40 % in 25% of the analyzed projects. Flyvbjerg also found no differences or patterns in the frequency of overestimating or underestimating traffic, indicating no optimism bias exists in road traffi c forecasting. In fact, traffic was actually 9.5% higher than forecasts predicted (u=1.095, sd= 0 .443). This data applies only to non tolled roads; however Flyvbjerg expressed interest comparing tolled roads, data permitting. In analyzing rail projects, F ted rail 84% of projects have actual traffic that was more than 20% below the forecast traffic and 74% of projects have forecast more than 40% below the forecast traffic. While forecast tr affic was grossly skewed, the predicting accuracy was actually higher than that

PAGE 53

53 of roads as seen by the tighter standard deviation of 0. 281 (u= 0 .486, sd= 0 .281) Flyvbjerg he optimism bias found in rail projects and not in road projects may be explained by funding procedures. The competition for limited funds in rail projects creates incentives to present a project in a favorable light. Thus the winning project and winning bidder is the one who has overestimating benefits Flyvb j years to see if learning had taken place in road traffic forecasting and if fore cast had improved over the time. Contrary to popular belief, Flyvbjerg found that forecasting had gotten worse o ver time. Traffic forecast at the end of the study were off by a larger percentage than forecast at the beginning of the study. The traffic d istribution results are depicted in Tables 2 5 and 2 6 8 Figure 2 5 Flyvbjerg inaccuracy for road projects 8 Source : http://flyvbjerg.plan.aau.dk/Traffic91PRINTJAPA.pdf

PAGE 54

54 Figure 2 6 Flyvbjerg inaccuracy for rail projects Bain Distribution Dr. Robert Bain is another researcher who compiled extensive research on t raffic forecasting accuracy i n his 2009 report Error and Optimism Bias in Toll Road T raf fic F orecasts sts to actual traffic for toll roads, something Flyvbjerg thought would be int four years, the study included 104 privately financed toll ed roads from around the world. resea rch found that t raffic on toll roads was normally distributed and considerably over predicted indicating a high level of error in traffic forecast (u=0.77, sd= 0 .26). Bain concluded that while toll free roads do not appear to have optimism bias, toll roads do in fact hav e a substantial amount of optimism bias. He also concluded that traffic forecasting accuracy for both tolled and un tolled roads was similar in terms of absolute error (Bain R. 2009)

PAGE 55

55 It has been suggested that first year traffic is difficult to predict due to the effects of the ramp up period therefore the following forecast periods may be more accurate for observation purposes. Bai n investigate d whether or not traffic forecasts improved over time; in other words, ive year mark more accurate? Results showed that for the first five years the mean of actual traffic as compared to the forecasted traffic varied from 0.79 to 0.80, while the standard deviation varied from 0.26 to 0.22. From this study he concluded that there is no evidence that traffic forecast improve after the first year (Bain R. 2009). Figure 2 7 Bain normal traffic distribution Bain also provided a personal anecdote which effectively highlights the difficulties in traffic forecasting. While exam ining base case traffic forecasts from four different internationally recognized traffic consultants all forecasting traffic for the same toll road, the following differences in predictions were observed. In the short run (first 5 years) the difference

PAGE 56

56 Figure 2 8 Bain traffic distribution adjusting for optimism bias between the highest and lowest forecasts was 26%, the medium term (first 15 years) was 106% and the long term (the first 35 years) was 266%. In the short run (first 5 years) the differenc e between the highest and lowest forecasts was 26%, the medium term (first 15 years) was 106% and the long term (the first 35 years) was 266%. This anecdotal evidence highlights the high variance in traffic forecasts even when modeling for the same projec t (Bain R., 2009 ). Credit Ratings on PPP Project B onds likelihood of default. Credit ratings are not governed by strict mathematical formulas but are an opinion by the age ncy based on both qualitative and quantitative factors. Credit ratings are used by investors to compare risk across industries and between different credit instruments. The majority of rated debt in the United States is rated by one of three major credit rating agencies,

PAGE 57

57 and rating scale; however, ratings are comparable amongst the three agencies according to the following Table 2 3. Table 2 3 Credit ratings d efin ed Moody's Fitch S&P Heading Aaa AAA AAA Prime Investment Grade Aa AA+ AAA High Investment Grade A1 A+ A+ Upper Medium Investment Grade A2 A A A3 A A Medium Investment Grade Baa1 BBB+ BBB+ Baa2 BBB BBB Lower Medium Investment Grade Baa3 BBB BBB Ba BB BB Non Investment Grade B B B Highly Speculative Caa CCC CCC Extremely Speculative C D D In default Credit ratings are important in PPP projects because private debt is used to finance the projects and investors require an ind ication of the riskiness of that debt. In addition, many projects receive TIFIA loans from the USDOT and an investment grade credit rating is a receive TIFIA cr edit assistance unless the senior debt obligations funding the project, i.e., those obligations having a lien senior to that of the TIFIA credit instrument on the pledged security, receive investment grade ratings from at least two Credit Rati (US DOT, 2013). at S&P and Fitch. edit ratings for municipal bonds on state and local government owned toll facilities in th e U.S. After each factor has been ratings. Considerations for the additi onal weights include: varying levels of risk factors such as

PAGE 58

58 the stability of the business model for the project, corporate governance, management strength, financial transparency, debt capacity, level of supervision by government, and the extent of likely government support. The final rating is also adjusted to incorporate any credit enhancements such as event risk protection clauses, debt structure and liquidity protection, and control afforded to creditors (Moody's, 2006) 9 Table 2 4 key rating factors for state and local government owned toll facilities in the United States Factors Sub factors Market position Scope of operations Competition Service area competition Demand Governance and maintenance Governance Regulatory framework Management Financial position and performance Operating performance Debt service coverage Revenue diversity Budgetary flexibility Financial reserves Debt and capital p lan Capital needs Capital planning and funding Covenants and legal framework Security pledge and flow of funds Rate co venant Additional bonds test Debt service and other reserves following list indicates several general assumpti ons about operational toll roads which issued both municipal and corporate bonds. In general, the credit rating agencies expect to see more PPPs for new toll facilities in the U.S., as the need for infrastructure continues to grow, and Sourc e: http://www.ibtta.org/files/pdfs/Moodys%20Toll%20Methodology 2006.pdf

PAGE 59

59 federal resources c ontinue to stretch thin (Moody's, 2006) Overall, operational toll roads have a low credit risk; toll roads have qualities similar to other sectors with low business risk. Investors Service 2006 report Moody's Rating Methodology for State and Loc al Government Owned Toll Facilities in the United States indicated that toll road operators for these projects receive high investment grade ratings as a result of (Moody's Investor's Service, 2006) : 1. Monopoly type activities, supported by long term conces sion agreements 2. Typically strong visibility in revenues and profit generation, due to (i) low demand elasticity and general resilience to economic fluctuations, and (ii) generally clear and predictable mechanisms for tariff increases, which will sust ain re venues over the long term 3. Straightforward business models, characterized by high profit and cash flow conversion, often with upside for cost efficiencies, and limited scope for cash calls resulting fro m off balance sheet liabilities However, financing for toll road operators has become increasingly leveraged as competition has increased and a wider deterioration in credit quality across the whole industry has occurred (Glazier, 2012) Although operational toll road investments offer lower risk than other s ectors, research suggests that traffic and revenue forecasting remains one of the primary concerns. toll road traf of project characteristics that present particular challenges to forecasting practitioners, the sensitivity of the forecasting process to input data quality and detail, and general trends and concerns within the toll road sector. In short a toll road off ering can incorporate more or less uncertainty depending on its specific attributes (Standard & Poor's, 2002). The credit rating agencies suggest that traffic and revenue studies s hould include a systematic approach to the analysis of traffic and revenu e risk and have developed tools such as the S&P Traffic Risk Index, to supplement for this inconsistency in project finance (Standard & Poor's, 2002)

PAGE 60

60 Municipal Bond Default The mos t common bond used to fund PPP projects in recent years is the Private Activity Bond (PAB), categorized as a type of municipal debt. Municipal bonds are normally considered a safe investment and have a lower risk of default than corporate bonds for the sa me credit rating. Moody's Investors Service in its annual report US Municipal Bond Defaults and Recoveries, 1970 2012 states that w hile municipal bond defaults have increased in number since the start of the financial crisis, they still remain extremely infrequent. In the period from 1970 2007, defaults of Moody's rated credits averaged 1.3 per year. For the 2008 12 period following five new defaults in 2012, the average rose to 4.6 defaults per year (Moody's, 2012) Although downside pressure in the municipal sector may persist for some time, Moody's believes municipal defaults will remain few in number. Historically, the majority of municipal defaults 70% have been in the healthcare and housing project finance sectors; between 1970 and 2011, 29 def aults occurred in housing bonds, 22 in hospitals bonds and only four in infrastructure bonds (Moody's, 2013) average 10 year cumulative default rate on municipal debt ra ted Baa between 1970 2006 was debt of 2.4507% to account for the period before 1970 and the fact that future periods may not enjoy the level of economic growt h exp erience d between 1970 2006 (Moody's, 2007) Figure 2 9 provides proposed idealized cumulativ e probability of default rates for municipal debt obligations, by rating category 10 10 Source: https://www.moodys.com/sites/products/DefaultResearch/102249_RM.pdf

PAGE 61

61 Note: Figure2 9 provides proposed idealized cumulative probability of default (PD) rates for municipal debt obligations, by rating category. For instance, municipal obligations rated in the Ba1 category have an idealized PD of 3.6761% within ten years of the initial rating date Figure 2 9 municipal idealized 10 year cumula tive probability of default rates Summary of Literature Review The information in this literature review is presented to provide a comprehensive projects and the d evelopment of the research questions. The following list summarizes the most important observations gained from the literature review : 1. Infrastructure funding shortfalls have generated legislation for promoting the growth of the PPP mechanism as an alte rnative delivery method for roads by providing low cost funding and assistance for PPP projects. 2. The majority of PPP failures can be attributed to inadequate or nonexistent feasibility studies including unrealistic traffic forecast and undefined public fu nds. 3. Flyvbjerg found that traffic forecasting ability was poor; furthermore, traffic forecasts are prediction of traffic explained by incentives in funding procedures. Finally, he found that traffic forecasting ability has not improved over time. 4. Bain found that traffic on tolled roads was normally distributed and considerably over predicted. He also found that traffic forecast do not improve after the first year.

PAGE 62

62 These observations led the researcher to investigate potential losses to debt and equity due to the inaccuracies of traffic and revenue forecasting and identify the probability of default in PPP projects. In order to accomplish this specific research question s were developed

PAGE 63

63 CHAPTER 3 METHODOLOGY A reward is the goal of taking a risk but the ever present question remains, whether the reward is worth the risk. In business the reward is most commonly known as profit, while the penalty is known as loss. Compa nies spend millions of dollars each year estimating risk and reward, developing complex and layered financial models, conducting sensitivity analyses, hiring rating agencies and consultants to perform analyses on projects, checking the numbers, and checkin g the numbers again, all in an effort to maximize profit and minimize loss. The same process occurs when developing a proposal for a PPP project; however, the PPP industry is relatively new in the U.S. and projects are few in number. This has led to a po or understanding of traffic revenue risk and rating and the risk based on traffic forecasting accuracy. In short, PPP projects may be riskier than previously thought which leads to the following questions for this study. Research Questions and Hypotheses This study is designed to answer the following questions and test the following hypotheses: 1. How accurately do bond ratings for PPP projects reflect risk, based on traffic demand? It is hypothesized that bond ratings accurately reflect the underlying default risk based on traffic demand, assuming no optimism bias in forecasting. 2. If optimism bias is assumed to be present in traffic forecasting, how accurately do bond ratings for PPP pro jects reflect risk, based on traffic demand? It is hypothesized that bond ratings do not accurately reflect the underlying default risk based on traffic demand, if optimism bias exists in forecasting. 3. What is the range and average base case risk premium a concessionaire is willing to accept to undertake a PPP project? It is hypothesized that all projects will demand a similar risk premium for returns to equity and that the base case equity risk premium will be 5%. 4. How do the case study projects compare in terms of returns and risk? It is hypothesized that the projects will be similar in terms of return and risk.

PAGE 64

64 5. If optimism bias is assumed to be present in traffic forecasting, what is the base case equity risk premium a concessionaire is willing to accept to undertake a PPP project? It is case equity risk premium will be near 0%. 6. As risk are transferred from the public to private partner (availability payment t o direct tolling), does the equity risk premium increase? It is hypothesized that the equity risk premium associated with direct toll projects will be larger than the equity risk premium associated with availability payment projects. 7. Do projects with a hi gher potential for default require a higher equity risk premium? It is hypothesized that as the probability of default increases the equity risk premium required by the concessionaire will also increase. Methodological Steps to Answer Each Question/Hypoth eses To answer question 1, the researcher will compare the historical default risk for municipal debt rated BBB, the same rating given to most PPP projects, with the probability that forecast traffic will not generate sufficient cash flows to service the debt as predicted by the model developed. If the bond financing accurately reflects the underlying risk of default from low traffic, they will have a similar likelihood of default. To answer question 2, the researcher will compare the historical default risk for municipal debt rated BBB, t he same rating given to most PPP projects, with the probability that forecast traffic will not generate sufficient cash flows to service the debt as predicted by the model developed. This time however, the distribution of traffic will be adjusted to reflect optimism bias in the forecast. If the bond financing accurately reflects the underlying risk of default from low traffic given optimism bias, they will have a similar likelihood of default. To answer question 3, the researcher will calculate the base case equity internal rate of return (IRR) for each project as predicted by the model developed. The risk free rate at the time of financing will be subtracted from the equity IRR to yield the equity risk premium (Rp) An average equity Rp will be calculated for all projects. To answer question 4, the researcher will graph the equity IRR versus probability of exceedance for each project on the same graph. Comparisons will be made between projects about the rate

PAGE 65

65 and vola tility of returns. To answer question 5, the researcher will calculate the base case equity IRR for each project as predicted by the model using a traffic distribution to reflect optimism bias in the forecast. The risk free rate at the time of financing will be subtracted from the equity IRR to yield the equity Rp. An average equity Rp will be calculated for all projects. To answer question 6, the researcher will compare the mean equity Rp on the five direct toll projects with the mean equity Rp from to the mean equity Rp from t w o availability payment projects that are not part of the case studies The AP projects should require a lower equity Rp because they do not bear traffic revenue risk. To answer question 7, the researcher will compare the base ca se equity IRR with the probability of cash flows not being sufficient to allow the concessionaire to maintain the TIFIA scheduled debt service. Projects with a higher probability of missing a TIFIA scheduled debt payment are riskier and therefore should r equire a higher equity IRR. Case Selection Types of Cases Cases for this study were selected based on project delivery types, location, and revenue stream. First, as previously indicated, this study will evaluate U.S. public private p artnerships for surf ace transportation infrastructure projects; more specifically, the projects mus t be a road, bridge, or tunnel. Because road and road links are similar to each other and the most common element of any state DOT, this study will focus on highway road projec ts. Although railways are another form of transportation that DOTs may undertake, railways are no t as common as roads and have a much different risk profile. Within the realm of U.S. highway PPP project s, a variety of differe nt project types exist. Is an operations and m aintenance contract on an existi ng road a PPP? Operations and m aintenance PPPs are now com mon throughout the U.S. Is an operating l ease a PPP? The sale, actually an operating lease, of the Chicago Skyway raised the public visibility f or all PPP

PAGE 66

66 projects. With varying interpretations of what project delivery types are considered PPPs, the cases were restricted to the following specifications: 1. T he project delivery type must be a DBFOM 2. T he project must have an element new construction 3. P rivate financing must be arranged by the concessionaire 4. T he project must have an operating period in which the concessionaire is responsible for the operations, maintenance and capital expenditures; and the concessionaire receives returns for providing the project. 5. T he projects must have a similar underlying revenue structure based on traffic volume, such as direct tolling, shadow tolling or availability payments 6. T he concessionaries must bear some level of design, construction, operating and demand ris k 7. The projects must be g reenfield PPP infrastructure projects as defined by the US Department of Transportation. The Federal Highway Administration websi te provides the definition for g reenfield PPPs ng and/or buildi ng a new asset. In the case of a g reenfield PPP project, a public agency transfers all or part of the responsibility for project development, construction and operation to a private sector entity. Greenfield projects generally present higher risks to bot h parties than do brownfield projects because of the greater uncertainty surrounding provide a variety of locations because each state uses different methods t o procure PPP projects ; selected cases are listed in Table 3 1. The four states with the greatest number of PPP projects matching the selected case descriptions are Virginia, Texas, California and Florida.

PAGE 67

67 Table 3 1. Selected c ases Notice to p roceed Pr oject n ame Project t ype Public s ponsor Risk Revenue s tructure 12/2007 I 495 HOT Lanes, VA Highway Virginia DOT DBFOM Direct Toll 12/2009 North Tarrant Express, TX Highway Texas DOT DBFOM Direct Toll 06/2010 I 635 LBJ Managed Lanes, TX Highway Texas DOT DBFOM Direct Toll 04/2012 Elizabeth River Crossing Tunnel, VA Tunnel Virginia DOT DBFOM Direct Toll 08/2012 I 95 Express Lanes, VA Highway Virginia DOT DBFOM Direct Toll While all of the projects listed are similar, each state developed a unique approa ch to distribute the demand risk. For instance, Texas had the most traditional PPP contract structure, with the concessionaire bearing the majority of the demand risk. In Texas, the concessionaires set the toll rates and collect the revenues. Florida, o n the other hand, used a maximum availability payment method (MAP), in which the state bore all of the demand risk and only compensated the concessionaire for keeping the facility open for use; this method placed all the demand risk on the government. Som e projects received large up front capital con tributions from the government while some projects received relatively small government contributions Number of Cases Ideally, analyzing a large number of similar cases would allow researchers to draw me aningf ul statistical results; h owever, this is just not possible with PPP in the U.S. The complexity and newness of PPP contracting in the U.S. means a limited number of cases are available for analysis. At the beginning of this research, 22 PPP projects for s urface transportation infrastructure existed; only 12 projects meeting the criteria for this study. With this limited number of cases it was impossible to analyze the data and perform any statistical comparison. Instead, this study focused on individual case studies; therefore it was determined to obtain and analyze as many case studies in depth as possible.

PAGE 68

68 A good amount of literature an d documentation was available about existing PPP projects in the U.S., including relevant project websites, annual rep orts, and local knowledge. However, obtaining sensitive data such as the forecasting models, traffic and revenue studies and legal contracts, limi ted this study to fo cus on just five PPP projects. The information presented in this study is based on the be st available public information on each PPP project. While this number may not be statistically significant in a quantitative analysis, it will contribute qualitative insight to the U.S. PPP industry. Selecting Conditions Cases were selected based on avai lability of financial data, and the criteria stated above. Financial data in PPP projects was extremely difficult to obtain. States provided public information such as overviews of the projects, promotional materials extolling the virtues of the project and copies of the comprehens ive development agreements but financia l models were considered proprietary and sensitive. Discussions were held with individual state DOTs and Federal Office of Information Acts (FOIA) were filed for both the USDOT and TXDOT. After exhausting efforts in trying to obtain f inancial data financial information was obtained through the public records for the issuance o f PAB transcripts. These transcripts were published by the underwriter and distributed in order to sell bonds to finance the PPP projects. The bond transcripts were published on the Electronic Municipal Market Access ( EMMA ) as requ ired by SEC regulations. The details of each selected case are displayed in Appendix A, which inc ludes the FHWA Project Profiles Data Collection Collection and organization of project data was a rigorous process which involved dissecting complex legal and financial documents for each of the selected cases. The financial data used in this study was collected from two primary sources; the EMMA website, established

PAGE 69

69 to provide key information to the public abou t municipal securities, and government online resources, such as the FDOT and FHWA websites. In addition, project information was obtained cluded up to date info rmation about the status of the project for the driving public. The data that will be used in the models includes the traffic forecasts, T&R studies, debt and equity repayment schedules, and rules and regulations governing the cash f lows throughout the duration of the project Instrumentation Two Instruments were created in Excel to analyze the data. The first is a set of two traffic distributions that when applied to the traffic fore cast in the T&R studies adjust s the forecast traf Simulated Cash Flow Model that uses the rules and relationships as outlined in the PAB transcript s to regulate how cash is generated and spent in the case studies. Both instruments are discussed in detail in t he following section. Traffic Distributions Data for comparing the ratio of actual traffic to forecasted traffic for U.S. PPP highway projects is not available because of the limited n umber of operational PPP projects. Existing projects have only been in operation for a short time period; as a result traffic data has not been recorded, or has not been published by the private owners. The Flyvbjerg and Bain studies are two of the most significant works on the accuracy of traffic forecasting and extremely i mportant to this study. Since U.S. PPP highway projects are under the same financial pressure as both rail and toll roads, it is assumed that U S PPP highway traffic forecast are no more accurate than the traffic forecast in tolled roads that Bain observ predicted traffic (u= 0 .77, sd= 0 .26) is used as the basis to model the predictive ac curacy of the T&R studies in each case study

PAGE 70

70 r does not set out to create a prediction with bias, instead he sets out to create and unbiased estimate with u=1.00. It would be hard for a concessionaire or financier to undertake a project while simultaneous believing that the forecast is over predicte d by as much as 25%. The procurement process can only work if everyone has some amount of faith that the predictions of the traffic will be re centered to u= 1.00. To maintain the same predictive accuracy that the model previously had but eliminate the bias requires adjusting the standard deviation. This was accomplished by assuming that the covariance of both distributions would remain the same Cv = sd/u As a result a new distribution (u=1.00, sd= 0 .34) is produced assuming that optimism bias does not exist in U S PPP highway projects. Figure 3 1. Bain probability distribution function forecasting accuracy The predictive accu racy of these two distributions will be used to vary yearly traffic and stress test the cash flow models. A confidence i nterval will be selected along the curve and t he ratio of actual to predicted traffic is computed based on the normal distribution Th e ratios of

PAGE 71

71 actual to predicted traffic for each probability of exceedance is shown in Table 3 2. It can be seen in a normal distribution that 60% of the time actual traffic will exceed 0. 9 1 multiplied by the predicted traffic, assuming no optimism bias. Table 3 2. Ratio of actual to forecast traffic Probability of exceedance (Actual / forecast) traffic Tolled roads w/ no optimism bias Tolled roads w/ optimism bias (u = 1.00, sd = 0.34) (u = 0.77, sd = 0.26) 0.01 1.79 1.37 0.1 0 1.44 1.10 0.2 0 1.29 0.99 0.3 0 1.18 0.91 0.4 0 1.09 0.84 0.50 1.00 0.77 0.6 0 0.91 0.70 0.7 0 0.82 0.63 0.8 0 0.71 0.55 0.9 0 0.56 0.44 0.99 0.21 0.17 These distributions show the actual to forecast traffic only in the first year of operation. The cash flow m odel fo r this study however will apply this ratio to every year of the forecast traffic for the entire term of the concession, in some case up to eighty years. The dual r prediction is wrong, and that traffic forecast do not get better with time. Bain has demonstrated that traffic forecast do not get better with time and are no more accurate five years after opening than in the first year U nfortunately no studies have been conducted on the accuracy of traffic forecast from year five through eighty (Bain R. 2009) The Cash Flow Model Concessionaires for each project developed a cash flow model to analyze financial data for the project, track revenues, expenses, debt rep ayment and profit for each year of the concession. These models are private intellectual property that contributes to each

PAGE 72

72 For this reason, a cash flow model was constructed to emu for each project These reproduced cash flow models will be referred to in this study as the Simulated Cash Flow Model s. type and contract style, each project was unique and a separate Simulated Cash Flow M odel was constructed in Excel for each project. The following sections explain the methodology used in constructing the Simulated Cash Flow Models used to perform this study. The Simulated Cash Flow Model for North Tarrant Express (NTE) will be used as a sample throughout the methodology to explain the process of building the models but calculations for all models can be found in Appendix B. The following documents and worksheets taken from the PAB transcripts were incorporated into the Simulated Cash Flow Model: Bond P roceeds Sources and Uses of Funds PAB Debt S ervice Flow of Funds Lender orecast overage Revenue Sharing/ Compensation Terms The Simulated Cash Flow Model was constructed to analyze financial data on each of the selected cases and determine an equity IRR and probability of default based on varying traffic distribution. Because cash flow models are used in finance to predict f uture revenues a nd expenses, a cash flow model relies heavily on net present value (NPV) analysis which is the concept that a Therefore, it is possible to

PAGE 73

73 convert present dollars to a future (or past) val ue using an inflation rate. The general equation to calculate NPV is a geometric growth rate: Future Value = Pre sent Value*(1+ inflation)^ time In PPP infrastructure finance, the goal in developing a cash flow model is to accurately predict, for each ye ar of the concession the inflows of cash, i.e. toll revenues ; and the outflows of cash, i.e. O&M, CAPEX, debt repayment, and payments to equity. Using the Simulated Cash Flow Model it is possible to which i s the discount rate where th e NPV of the cash flows is zero. The Simulated Cash Flow Model also predict s the In order to construct the Simulated Cash Flow Model the PAB transcripts were reviewed to establish conditions, cla uses, and rules governing the project. The flow of funds for the Simulated Cash Flow Model are depicted in Figure 3 2 and explained in detail in the following sections. The Simulated Cash Flow Model detail ed in each contract, with the most senior obligations p aid first and equity paid last; h owever, the models were simplified to analyze relevant data and some of the steps where eliminated as show in Figure 3 2. These models therefore should be thought of a s a simplified analysis an d similar to the concessionaire s models.

PAGE 74

74 Figure 3 2. Simulated cash flow model flow of funds Cash flow models constructed by the concessionaires are developed after months of planning and observa tions taken by teams of exper ts; they consist of extremely complex calculations, and are subject to external review by consultants and government officials. It is not the purpose of this study to second guess the cash flow models produced by the concessionaires. The Simulated Cash Fl ow Models were constructed in order to apply a different set of inputs to detailed explanation of the assumptions made in these Simulated Cash Flow M odels can be found later in this section. base case scenari traffic and revenue forecast which is used to populate the Simulated Cash Flow Model but the base case scenario represents just one level of traffic. In order to explore the risk and return at higher or lower traffic levels it is necessary to constr uct a cash flow model that can function using varying levels of traffic inputs.

PAGE 75

75 Step 1: Convert Nominal to Real D ollars The first step in construction the Simulated Cash Flow Model s is to convert nominal to real dollars. Project revenues, operations & maintenance costs, and major maintenance costs as t ranscripts. These numbers are given in n ominal dollars or the value of a dollar in the year it is transacted. In order to perform the regression analysis and make meaningful comparisons in later steps, the effects of inflation must be removed. Each PAB transcript presented a base year for fina ncial calculations and an assumed rate of inflation that was used to convert real to nominal dollars. Using this information, the procedure was reversed to calculate real dollars in the base financial year Some of the concessionaire s cash flow models v aried inflation rates over time instead of using a steady growth rate; however, to simplify the Simulated Cash Flow Model a single constant inflation rate was used in each model In the case of NTE revenues were already given in the base case financial year of 2008 so it was unnecessary to convert revenues to real dollars; only o perating and m aintenance cost and major m aintenance needed to model so 2.5% was us ed to convert nominal to real dollars. Table 3 3: NTE Convert Nominal to Real Dollars depicts the first 15 years of project information, complete information and graphs for each project can be found in Appendix B. Table 3 3. NTE convert nominal to real do llars Year O&M e xpenses (year of $) O&M e xpenses (2008$) M ajor maintenance c osts (year of $) M ajor maintenance costs (2008$) 2015 7,463,443 6,278,735 182,625 153,636 2016 17,360,575 14,248,632 356,717 292,774 2017 19,495,894 15,610,915 584,175 467,765 2018 20,854,252 16,291,308 2,966,781 2,317,645 2019 22,305,707 17,000,178 2,827,915 2,155,281 2020 24,011,378 17,853,801 1,360,308 1,011,465

PAGE 76

76 Table 3 3. Continued Year O&M e xpenses (year of $) O&M e xpenses (2008$) M ajor maintenance c osts (year of $ ) Major maintenance costs (2008$) 2021 23,540,161 17,076,512 1,226,520 889,743 2022 23,645,306 16,734,426 2,027,880 1,435,186 2023 26,046,114 17,983,945 1,404,791 969,960 2024 25,182,203 16,963,360 16,356,968 11,018,461 2025 26,742,834 17,575,258 10, 207,916 6,708,592 2026 27,088,127 17,367,984 18,408,521 11,802,916 2027 28,750,186 17,984,038 18,001,670 11,260,544 2028 30,770,831 18,778,544 20,163,854 12,305,414 2029 31,189,767 18,569,960 54,359,508 32,364,906 2030 31,788,945 18,465,075 56,567,300 32,857,946 Step 2: Determine Length of Concession to be Modeled Each of the cases in this study had different concession terms and it was necessary to decide what years of the concession should be modeled for each project. The concession terms general ly begin immediately following the signing of the comprehensive development agreement and have a fixed expiration date; terms range from 30 years to 85 years The first five years of most concession contracts was pla nned as a construction period; du ring t his period the concessionaire collects no or limited revenue. Construction delays may even prevent the concessioner from collecting revenues for greater than five years if construction is not complete. Concessionaires must however make debt service payme nts on the PABs during construction. Because they have no income during the construction pe riod it is assumed that the PAB pa yments during this period come from the construction financing package and are included in the total cost to construct the project For this reason the Simulated Cash Flow Model s only model the operating phase of the concession when revenues, not debt, must pay for expenses and debt repayment. Even if sections of the project are opened early and the concessionaire collects some rev enue before the project is complete, this was still considered to be part of the construction financing and not modeled in the Simulated Cash Flow Model For example, NTE

PAGE 77

77 had a 52 year concession term beginning in 2009 and ending in 2061. Construction wi ll take place from 2009 until the end of 2014; during this time the concessionaire will collect no revenues. The project will open in 2015 and this will be the first year the concessionaire collects revenue. Cash flows are therefore modeled from 2015 unt il 2061. Step 3: Select Traffic & Revenue Forecasts to be M odeled For each project, traffic and revenue (T&R) studies were prepared by two independent transcript, i Traffic Forecast in each case, revealed that the was the more conservative. For this reaso used in constructing the Simulated Cash Flow Model Most T&R studies did not forecast for the full term of the concession because it is difficult to predict traffic greater than 35 years in the future. Many of the T&R studies capped the traffic at a fixed level after a specified number of years. For example, traffic would increase at 2% per year for the first 30 years and then remain constant for the last 10 years of the forecast. This is because there is a finite num ber of cars that can fit on a road at the same time; traffic cannot grow infinitely. It is generally assumed that the capacity of highway lane is 2,000 cars per hour. The Simulated Cash Flow Model maintained the traffic cap if one was provided so that tr affic could not grow infinitely. If a T&R study did not specify a traffic cap, it was assumed that the road had reached capacity by the end of the T&R study and a fixed level of traffic was used for the remainder of the concession term in Simulated Cash F low Model

PAGE 78

78 Some of the T&R studies forecasted traffic in terms of annual t raffic, or the total number of vehicles to use the facility over the course of a year. Other T&R studies forecasted traffic in terms of average annual daily t raffic (AADT) or the number of vehicles that can be expected to use the facility on any given weekday. Annual traffic was used to model traffic in the Simulated Cash Flow Models so when traffic was given in AADT a traffic annualization f act or from the T&R study was used to c onvert AADT to annual t raffic. This factor relates traffic on weekends and holidays to traffic on weekdays. This is necessary because traffic on weekends is often less than on the weekday and simply multiply ing AADT by 365 days/year would over estimate t raffic. A similar annualization factor was used to convert daily revenue Table 3 4 depicts the first 15 years of forecast traffic for NTE; a traffic annualization factor 337 was used to convert daily to annual traffic. All traffic forecast can be found in Appendix B. Table 3 4. NTE traffic forecast with annualization Year Lenders base t ransactions (average y ear ) 2014 2015 12,954,280 2016 22,237,619 2017 27,071,210 2018 30,394,367 2019 28,747,111 2020 29,314,619 2021 29,667,458 2022 30,021, 308 2023 30,375,832 2024 30,731,367 2025 31,346,729 2026 31,453,221 2027 31,605,208 2028 31,760,902 2029 31,920,303 2030 32,017,696 *Note: traffic annualization f actor 337 applied

PAGE 79

79 Step 4: Perform Regression Analyses o n Revenues A regression analy sis was performed on traffic and revenues from the T&R study because revenues fluctuate as traffic levels fluctuate; traffic is the source of revenues and the independent variable. The T&R study provides a revenue leve l for each future traffic level but w e can also think of these revenues as corresponding to a particular traffic level in any given year. Because revenues are given in real dollars we do not need to worry about the effects of inflation. Revenue is a dependent variable correlated to traffic; however, the relationship is not necessarily linear for reasons such as varying levels of elasticity o f demand, varying profit obj ectives the effects of competing facilities, and the effects of traffic congestion and economic growth. For example High Oc cup ancy Toll (HOT) lanes are toll lanes which provide a guaranteed travel time from a starting point to a destination. In all cases of HOT lanes users always have the alternative to use the free lanes. To regulate the travel time in the HOT lanes, conge stion pricing is used where the travel time, not the toll, is constant and the toll rate varies with demand. As traffic volume rises in the free lanes travel time also increases and hence the demand for the HOT lane increases. To regulate the number of vehicles in the HOT lanes toll price s also rise; l ikewise, as traffic volume falls, so does the demand and the price of the to ll on HOT lanes. This results in traffic volume having an exponential effect on revenue. It should be noted that the concession varied depending on the year. Revenue leakage is potential revenue that is never actually s models also include revenue from activates other than tolling, like investing. For simplicity leakage and investing activity were not included in the Simulated Cash Flow Model.

PAGE 80

80 In the case of NTE the regression analysis resulted in choosing an exponent ial relationship (REVENUE = 14,827,478e0.00000006.19*TRAFFIC) to model revenues. The R 2 value was found to be 0.97 indicating a very good fit between revenue and traffic. This relationship allows the traffic volume to be varied and the revenue to be pred icted with a high level of certainty. Table 3 5 depicts NTE Traffic & Revenue Forecast worksheet for the first 15 years of the project, complete information and graphs for each project are located in Appendix B. Table 3 5 NTE traffic & revenue forecast Year Lenders base transactions (average year ) Lenders base annual revenue (2008 $) 2014 2015 12,954,280 38,548,000 2016 22,237,619 59,289,000 2017 27,071,210 72,975,000 2018 30,394,367 84,006,000 2019 28,747,111 78,2 74,000 2020 29,314,619 81,197,000 2021 29,667,458 83,641,000 2022 30,021,308 86,106,000 2023 30,375,832 88,592,000 2024 30,731,367 91,099,000 2025 31,346,729 95,777,000 2026 31,453,221 100,391,000 2027 31,605,208 105,062,000 2028 31,760,902 109,79 0,000 2029 31,920,303 114,576,000 2030 32,017,696 119,176,000

PAGE 81

81 Figure 3 3 NTE revenue vs. traffic graph Step 5: Perform Regression Analyses on Operations & Maintenance Costs Operations & m aintenance (O&M) costs are forecasted in the PAB transcripts and receive the highest level of payment priority because these costs are necessary to operate the road. O&M costs generally include minor maintenance, paying employees, and the cost associated with collecting tolls. Because these functions are necessar y to keep the road in operation and generating revenues t hey are non deferrable. Major m aintenance costs, such as repaving the road, are allocated as capital expenditures. O&M schedules for each case were rvice Coverage for the Bo nds During These figures were forecasted only during the debt repayment peri od. For all case studies a linear relationship was found between O&M and revenue. Table 3 first 15 years of project information, complete information and graphs for each project can be found in Appendi x B.

PAGE 82

82 Ta ble 3 6. NTE expenses Year Revenues (2004$) O&M e xpense (2004$) 2013 41,894,852 24,011,096 2014 60,136,582 21,816,054 2015 71,684,560 22,104,229 2016 75,001,550 22,475,051 2017 75,993,350 22,580,278 2018 76,998,450 22 ,686,742 2019 78,016,850 22,794,462 2020 79,784,800 22,978,426 2021 80,769,000 23,086,828 2022 81,765,550 23,196,464 2023 82,774,450 23,307,356 2024 83,796,650 23,419,545 2025 84,830,250 23,532,981 2026 85,877,150 23,647,756 2027 86,936,400 23,763 ,857 2028 88,009,900 23,881,347 2029 89,095,750 24,000,205 2030 90,194,900 24,120,459 Figure 3 4. NTE O&M vs traffic graph The relationship between O&M Expenses and Traffic for NTE was found to be (O&M EXPENSES = 1.373.409 + 0.52 TRAFFIC). The re lationship had an R 2 of 0.93 ind icating a

PAGE 83

83 good fit between O&M e xpenses a nd traffic. The complete regression analysis can be found in Appendix B. Step 6: Perform Regression Analyses on Major Maintenance Costs Major maintenance costs estimates are based on accounted for as capital expenditures on the road; this does not include minor maintenance items such as repai rs. After O&M expenses, major maintenance c osts claim the next highest priority on revenues because the conces sionaire is required by contract to keep the road in good operating condition for eventual turn are contractually obligated to pay major m aintenance before debt, in other words, funds cannot be diverted from major m aintenance to prevent default. Major maintenance schedules were projected only d uring the loan repayment period (not and Debt Service Cov erage for the Bo of the PAB transcripts. In addition, major m aintenance is not perf ormed annually but on a cycle when items are expected to maintenance fund prior to the time of use, i.e., funds may be deposited into the account for 15 years and upon repaving, funds are completely exhausted. Because each project had unique major maintenance requirements, no relationship was calculated betwee n the funds deposited into the account, or spent from the account. The cyclical nature of major maintenance made modeling difficult. Fig ure 3 chedule.

PAGE 84

84 Figure 3 5 NTE major maintenance vs time graph It was d ecide that because the total traffic, not the annual traffic, is responsible for deterioration of the road surface that the relationship between accumulated traffic and accumulated major m aintenance would be modeled. Table 3 7. NTE Accumulated Major Maint enance worksheet depicts the first 15 years of project information, complete information and graphs for each project can be found in Appendix B. Table 3 7. NTE accumulated major maintenance Year Traffic Accumulated t raffic Ma jor maintenance c osts (2004$) Accum ulated major maintenance c ost (2004$) 2013 20,786,400 20,786,400 10,281,624 10,281,624 2014 30,003,000 50,789,400 10,281,624 2015 35,963,700 86,753,100 10,281,624 2016 36,724,200 123,477,300 10,281,624 2017 37,333,500 160,810,800 10,281,6 24 2018 38,037,900 198,848,700 2,930,761 13,212,385 2019 38,755,800 237,604,500 16,521,626 29,734,011 2020 40,662,900 278,267,400 9,869,806 39,603,817 2021 41,394,900 319,662,300 6,950,654 46,554,471 2022 42,140,100 361,802,400 14,148,122 60,702,592 2023 42,898,800 404,701,200 5,178,275 65,880,867 2024 43,671,300 448,372,500 505,983 66,386,850

PAGE 85

85 Table 3 7. Continued Year Traffic Accumulated t raffic Major maintenance c osts (2004$) Accumulated major maintenance c ost (2004$) 2025 44,457,300 492,829,800 8,051,310 74,438,159 2026 45,257,700 538,087,500 9,591,818 84,029,978 2027 46,072,500 584,160,000 15,707,701 99,737,679 2028 46,902,000 631,062,000 6,922,780 106,660,458 2029 47,746,500 678,808,500 7,271,392 113,931,850 2030 48,606,000 727,414,500 17 ,792,793 131,724,643 Figure 3 6 NTE accumulated cost vs accumulated traffic graph Figure 3 7. NTE Accumulated Cost vs Accumulated Traffic Graph demonstrates a strong linear relationship was found between these with (ACCUM MAJOR MAINTENANCE = 33, 296 ,704 + 0.29 ACCUM TRAFFIC). The R 2 value was 0.94 indicated a good fit. Step 7: Estimate Revenue The first quanti ty to be calculated by the Simulated Cash Flow Model is r evenue Revenues are estimated for each year of the concessionaire based upon the relationship between traffic and revenue that was calculated in Step 4 of the model formulation. The revenues are first calculated in real dollars and then converted to nominal dollars for each year of the concession.

PAGE 86

86 For the NTE base case, traffic on th e HOT lanes in the first year of operation, 2015, is predicted to be 12,954,280 cars, resulting in $33,056,696 in 2008 dollars which is equivalent to $39,294,024 in 2015 dollars. Step 8 : Estimate O&M Expense After calculating revenues, the deductions from revenues are calculated in order of priority according to the flow of funds. The h ighest priority expense is O&M e xpense for reasons discussed ea rlier. O&M e xpense is based on the relationship between traffic and O&M e xpense that was calculated in Step 5 of the model formulation. O&M e xpense is calculated directly in nominal dollars so it is easy to subtract from r evenues. For the NTE base case, traffic on the HOT lanes in the first year of operation, 2015, is predicted to be 12, 954,280 cars, resulting in O&M e xpenses of $9,601,216 in 2015 dollars. Step 9 : Estimate Major Maintenance Expense M aintenance expense is calculated after revenues and O&M expense. Major maintenance e xpenses would normally reflect a very c main tenance e xpense having little to do with the following year. Because t his is difficult to model, the major m aintenance e xpense s we re modeled to grow linearly and then deducted from each year regardless if the funds were actually spent that ye ar or in a future year. Major maintenance e xpenses are based upon the re lationship between traffic and major maintenance e xpense that was calculated in Step 6 of the model formulation. Major maintenance e xpense is calculated directly in nominal dollars so it is easy to subtract from r evenues. For the NTE base case, traffic on the HOT lanes in the first year of operation, 2015, is predicted to be 12,954,280 cars, resulting in $0 bei ng spent on major m aintenance. Spend ing $0 on major maintenance is unusual but this is an anomaly caused by using a linear relationship on a very low accumulated traffic count, and occurs in the first year only The model begins deducting for

PAGE 87

87 major maintenance e xpense in the second year of operation, 2016, with $7,875,610 in 20 16 dollars. Step 10 : Calculate Net Cash F low A vailable for Debt Service. Net cash flows available for debt s ervice is the amount of money left over to serv ice debt obligations after O&M expense and major maintenance e xp ense have been subtracted from r evenu es. This category also includes funds le ft in the revenue a ccount from previous years that were no t distributed as profit to the c oncessionaire either because the DSCR was not high enough to allow distribution or in the case of NTE because distributions t o equity are not permitted in the first five years of ope rations. After satisfying O&M expense and major maintenance e xpense, NTE has $29,692,808 to service debt in 2015. Step 11 : Determine Debt Service on Private Activity Bonds (PABs) The private activit y b onds (PABs) receive highest priority in debt repayment on each project and therefore have the first claim on revenues after O&M expense and m ajor maintenance expense have been paid. Interest payments on the PABs began immediately upon financi al close a nd are paid from the debt reserve a ccount during the construction period; after the construction period ends interest on the bonds is paid from general revenues. Pa yments on the debt principal are structured to begin later in the project timeline when the project generates higher cash flows. Failure to make a PAB debt payment results in default. Each PAB transcript provided an amortization schedule for repayment of the bonds which included details of the bond interest rates and credit ratings on the bonds This table was pts. For 2015, NTE has a $27,874,000 debt service payment on the PAB.

PAGE 88

88 Step 1 2 : Determine Debt Service on TIFIA Loans TIFIA loans are junior debt and TIFIA debt is not paid until all senior debt is paid each year. Flow and Debt Service Coverage for The TIFIA loans are normally issue d for 35 year repayment periods; however repayment is deferred and interest ca pitalized for the first five years following substantial completion of the project, to allow revenues to increase to sufficient levels to make payments. TIFIA repayment schedules are separated into two separate categories : mandatory debt repayment and sch eduled debt repayment. Each of these debt schedules are unique and require f urther explanation. The TIFIA mandatory d ebt is the minimum that t he concessionaire is requir ed to repay to the TIFIA loan each year. Failure to meet a TIFIA mandatory debt paym ent is conside red to be default; however the Secretary of Transportation has the ability to renegotiate the debt. The Simulated Cash Flow Model does not automatically assume default from a missed TIFIA mandatory d ebt payment but instead assumes as long a s the debt will be repaid in the life of the concession that the missed payment will be recapitalized on the loan with a 1% pena lty. For NTE, the first TIFIA mandatory d ebt payment is $1,920,915 and paid in 2020. The TIFIA scheduled d ebt is a negotiate d payment plan between the concessionaire and TIFIA program based on the expected cash flow s This repayment normally star t s later in the concession term than m andatory debt r epayment allowin g revenues to grow. The TIFIA s cheduled r e payment can begin at any time and does not have to follow the r epayment schedule in the PAB t ranscript. If revenues are not sufficient to meet the scheduled repayment, TIFIA scheduled payments are pushed further in the future to allow traffic and revenue to increase. The amo unts not paid are capitalized at the TIFIA interest rate.

PAGE 89

89 Step 1 3 : An Explanation of Reserve Accounts Every project is required to have reserve accounts, with require d minimum balances over specified time frames as a form of credit enhancement to protect the investors. If revenues would be insufficient to cover any payment, the necessary monies will be taken from a reserve account, if available, to prevent default. Each project had different reserve accounts for different aspects of the project as well a s different requirements for those accounts. NTE has a major maintenance reserve a ccount and a debt service reserve a ccount The debt service reserve a ccount is never drawn upon in the base case NTE Simulated Cash Flow M odel The funds initial deposited into the account are transferred for distribution to equity after all debts have been repaid. The major maintenance reserve a ccount is not actually held in reserve in the Simulated Cash Flow Model, but used to pay for m ajor m aintenance e xpenses immediate ly Step 1 4 : Calculate Cash F low A vailable after Debt Service. Cash f low a vailable after d ebt s ervice is the amount of money left over for payments to e quity after all debt service obligations have been met. A fter satisfying all PAB, TIFIA m andatory and TIFIA s cheduled debt service NTE has $1,818,808 available to e quity in 2015. Step 1 5 : Calculate Revenue Sharing All projects contained some element of revenue sharing. Revenue sharing is designed to prevent the concessionaire from reaping windfall prof its and occurs when revenues are greater than a pre established threshold set in the comprehensive development agreement. In the case studies, two separate methods were used to calculate revenue sharing. In North Tarrant Express, I 635 LBJ Managed Lanes and Elizabeth River Crossings a table was provided that created multiple bands with allowable minimum and maximum revenue. As these bands are exceeded a percentage of the total accumulated revenue less previous revenue sharing payments is shared

PAGE 90

90 with the state. I 95 and I 495 have a similar revenue sharing structure that is based upon the IRR earned by the concessionaire. According to the comprehensive development agreement, payments to the state from revenue sharing have the highest priority even above O&M. The Simulated Cash Flow Model deducts these amounts last immediately before distribution to equity. While this is not the flow of funds specified in the comprehensive development agreement, there is little chance of missed payments on any other it em if revenues are sufficiently high to activate revenue sharing. In the base case, NTE had one instance of revenue sharing in the first year of operation with $1,611,811 being paid to the state. Step 1 6 : Calculate Payments to Equity Each year following t he repayment of all debt obligations, the concessionaire is entitled to all remaining cash flows if all conditions and debt service coverage ratios (DSCR) are met. In the case of NTE, the concessionaire cannot make payments to equity for the first five ye ars but in 2020 the concessionaire is allowed to pay himself the profits from the previous five years $250,973,560. Step 17 : Calculate Equity IRR Finally an equity IRR is calculated based upon the payments to equity for each year of the concession term. The entire equity contribution made by the concessionaire is assumed to occur in the first year of the concession agreement and each year following the start of operations the concessionaire receives payments to equity. Data Analysis The completed Simula ted Cash Flow Model functions autonomously after receiving input from the Traffic Distribution Instrument. The Traffic Distribution Instrument provides the Simulated Cash Flow Model with an array containing one traffic prediction for each year of the

PAGE 91

91 conc ession term. There are two separate traffic distributions in the Traffic Distributions Instrument and the user must select either the traffic distribution with or without optimism bias. After the desired distribution has been selected, the user must spec ify the point along a normally distributed probability curve in which he is interested. The Traffic Distribution Instrument then uses data from the base case traffic forecast and the PDF curve to calculate the new forecast traffic. After the model was c onstructed, the traffic distribution without optimism bias (u=1.00, sd= 0.34 ) and a base case probability = 0.5 was applied to each case study. This is reproduces the base case traffic forecast as given in the PAB transcripts. To verify the model was accu rate the outputs were compared to the base case data in the PAB transcripts. Next, the effects of lower than expected traffic were investigated in each case study. A probability of 0.4, 0.3, 0.2, 0.1 and 0.01 was applied to the traffic distribution in e ach case. The Simulated Cash Flow Model was examined for the follow ing occurrences: default on PAB debt, default on TIFIA mandatory r epayment, and failure to maintain the TIFIA s cheduled r epayment plan An iterative trial and error process was used to fi nd the probability of the occurrences above and they were recorded. The model was then run for varying levels of traffic 0.99, 0.9, 0.8, 0.7, 0.5, 0.4, 0.3, 0.2, 0.1, 0.01 and the equity IRR was record at each probability. If the proj ect defaulted before reaching one of the lower probabilities an iterative process was used to find the exact probabil ity of default. This procedure was then repeated for the traffic distribution with optimism bias (u=0.77, sd= 0.26 ). The final result of both the equity IRR a nd probability of default can be found in Chapter 4 Results.

PAGE 92

92 Limitations and Delimitations Limitations There are only seven projects completed and eight projects under construction. This results in a small sample set that does not allow the results to b e generalized to the larger financial models were not provided for this study so the study was conducted using the best data available. Delimitations The ac tual traffic model and methodology used by the traffic engineer is not examined. It is assumed that the traffic model s are state of the practice and it is not the goal of this study to question the methodology used in the traffic forecas ting. It is assumed however that all traffic studies fit into the accuracy distribution found by Bain. No particular traffic study The forecast from lender s, not the spo base case scenario. Interpretation and Refinement of Model Several methods were used to ass ess the validity o f the Simulated Cash Flow Model. The f irst method was to compare the revenue, O&M expense and major maintenance expense as estimated by the Simulated Cash Flow Model at t he base case scenario with the revenue, O&M expense and major maintenance Expense as provided in the PAB transcripts. The Simulated Cash Flow Model seeks to reproduce these values so this is an excellent comparison for verifying the Simulated Cash Flow Model Overall the base case estimates by the Simulated Cash Flow Model where found to be very close to values provided in the PABs; h owever, in

PAGE 93

93 some instances values were slightly under estimated in the early y ears of the concession and over estimated in the later years, or vice versa. While small, this is concerning because the timing of cash flows is critical It was not possible using this method to compare any values beyond the term of the loan because these values were not provided in the PAB transcripts. Another excellent comparison was e r evenue s haring worksheet from the Simulated Cash Flow Model. A permit fee is used to predetermined level, a portion of the revenue is shared with the state. This pre determined level is therefore designed to be above the base case scenario. Revenue sharing was a good indicator of the goodness of fit for the Simulated Cash Flow Model for NTE, LBJ and ERC because the revenue bands are given as an absolute value each yea r. They are not modeled or dependent on any other factor; they are also given for the full term of the concession. For the Simulated Cash Flow Model for NTE, revenue sharing was found to begin at 0.58 on the normal probability distribution function (base case = 0.50). For the Simulated Cash Flow Model for ERC, revenue sharing was found to begin at 0.51 on the normal probability distribution function. For the Simulated Cash Flow Model for LBJ, revenue sharing was found to begin immediately in the first y ear at 0.50, the base case. Revenue sharing in the base case would not be expected and is probably the result of a model simplification concerning the actual finish date of construction and the number of months the concessionaire earns revenue in the firs t year of operations. The next instance of revenue sharing in LBJ occurs at 0.58 on the probability distribution function. Overall these results indicate a good fit to the Simulated Cash Flow Model because it would be expected that revenue sharing begins somewhere between 0.50 and 0.60 probability

PAGE 94

94 I 495 and I 95 had a more complicated revenue sharing structure based on the the Simulated Cash Flow Model accordi ng to the terms of each comprehensive development agreement. For the Simulated Cash Flow Model for I 95, revenue sharing was found to begin at 0.51 on the normal probability distribution function. This once again indicates a good fit for the Simulated Ca sh Flow Model. For the Simulated Cash Flow Model for I 495 however, revenue sharing never occurred no matter the location selected along the normal probability distribution function. This makes the model for I 495 somewhat suspect. Summary of Methodology The methodology contains a detailed and involved description of each step required in the formation of the Simulated Cash Flow Model s o a summary is provided as a general overview and recap of the investigation process. 1. After reviewing literature on traff ic forecasting and PPP projects in the U.S., the researcher is investigating the potential losses to debt and equity due to the inaccuracies of traffic and revenue forecasting and identifying the probability of default in major PPP projects. In order to a ccomplish this, seven specific research questions and hypothesis were developed. 2. Five U.S. cases were selected for the study based on project type, availability of data and project location. The cases are I 495 HOT Lanes, VA; North Tarrant Express, TX; I 635 LBJ Managed Lanes, TX; Elizab eth River Crossing, VA; and I 95 Express Lanes, VA. 3. A T raffic Distribution Instrumen t was created to forecast traffic in each year of the con cession based on two inputs: the l base case traffic and revenue forecast and a confidence interval supplied by the user. Two separate traffic distributions were created, one assuming no optimism bias and one assuming optimism bias. 4. The Simulated Cash Flow Model model and account for funds as they are earned, spent and distributed each year. 5. A confidence interval is entered by the user into the Traffic Distribution Instrument, and the traffic forecast from each year is automatically entered into the Simulated Cash Flow Mod el The Simulated Cash Flow Model then returns the equity IRR and results of default for each confidence interval The results obtained from the cash flow model will be discussed in the following section.

PAGE 95

95 CHAPTER 4 RESULTS This chapter presents the res ults from the Simulated Cash Flow Model for each project. These results include two parts, 1) an equity IRR for a given p robability of exceedance and 2) the probability of a specified event occurring during project term, these events will be defined in th e following sections. Next, the equity IRRs and probabilit y of events for each project are used to answer each research question. Results of the Cash Flow Model Analysis of Selected Case D ata The first step of analyzing the results of the Simulated C ash F low M odel was to select and apply various confidence intervals to the Traffic Distribution Instrument and observe the output RR outputs were recorded for both traffic distributio ns for each project. The process described above was performed twice for each project, once for the traffic distribution assuming no optimism bias, and once for the traffic distribution assuming optimism bias. The results are recorded in Table 4 1 and Ta ble 4 2. Table 4 1 Equity IRR given probability of exceedance without optimism bias Probability of exceedance 0.01 0.10 0.20 0.30 0.40 0.50 0.60 0.70 0.80 0.90 0.99 I 495 11.57 11.54 11.31 11.00 10.63 10.69 9.70 9.06 8.07 6.74 LBJ 29.42 22.92 18. 08 14.90 11.95 9.15 6.51 3.49 NTE 20.34 17.81 15.44 13.63 12.02 10.44 8.71 6.91 4.94 0.16 ERC 18.16 18.16 17.67 16.57 15.09 12.42 9.20 6.48 I 95 32.70 28.24 23.37 20.26 17.63 14.63 11.63 8.82 5.96 Table 4 2. Equity IRR given probability o f exceedance with optimism bias Probability of exceedance 0.01 0.10 0.20 0.30 0.40 0.50 0.60 0.70 0.80 0.90 0.99 I 495 11.47 10.71 10.15 9.65 9.16 8.59 7.98 7.35 6.60 5.27 5.03 LBJ 20.98 12.51 8.69 6.28 4.04 NTE 16.90 12.33 10.23 8.56 7.17 5.82 4. 73 3.11 4.81 ERC 18.16 15.51 12.02 8.95 6.78 6.10 5.67 I 95 26.17 18.07 14.11 11.29 9.16 7.27 5.85

PAGE 96

96 The probability of an event occurring refers to the following possibilities: 1. The probability of default on PABs 2. The probability of a missed p ayment on the TIFIA mandatory repayment 3. The probabilit y of a missed payment on the TIFIA scheduled repayment ( P robability of E xceedance)) because default occurs when t he threshold traffic level is NOT exceeded. While this is not the strict definition of probability, a simplification was necessary to alleviat e confusion in the discussion. Table 4 3. Probability of default without optimism bias Debt r isk N(1.00, 0.34) P roject PAB probability of default TIFIA probability of default Restructure TIFIA mandatory TIFIA scheduled change I 495 4.00% 4.00% 10.00% 17.00% LBJ 28.00% 28.00% 35.00% 45.00% NTE 7.00% 9.00% 17.00% 22.00% ERC 26.00% 26.00% 26.00% 33.00% I 95 19.00% 19.00% 19.00% 21.00% Average 16.80% 17.20% 21.40% 27.60% Table 4 4. Probability of default with optimism bias Debt r isk N(0.77, 0.26) Project PAB Probability Of Default TIFIA Probability Of Default Restructure TIFIA Mandatory TIFIA Scheduled Cha nge I 495 8% 8% 23.00% 32.00% LBJ 54% 54% 65.00% 76.00% NTE 14% 19% 38.00% 45.00% ERC 52% 52% 26.00% 62.00% I 95 36% 36% 36.00% 48.00% Average 32.80% 33.80% 37.60% 52.60% Findings to Answer Each Research Question/Hypotheses Research Question 1 How accurately do bond ratings for PPP projects reflect risk, based on traffic demand? It is hypothesized that bond ratings accurately reflect the underlying default risk based on traffic demand, assuming no optimism bias in forecasting.

PAGE 97

97 Answering th is question required two steps; first, a probability of default was associated with the bond ratings to understand the risk of default implied by a Baa3 crediting rating. The details of the calculation of probability of default can be found in Appendix D and the results are summarized below in Table 4 5. Most of the bonds were rated low investment grade, Baa, with a ten year default rate of 2.4507%. I 495 was the lo n e exception and rated higher in the Aa range. Table 4 5 Selected cases bond ratings Pro jects PAB Ten year probability of default I 495 Aa3/AA /A+ 0.1230% LBJ NR/BBB /Baa3 2.4507% NTE NR/BBB /Baa2 2.4507% ERC NR/BBB /Baa3 2.4507% I 95 NR/BBB /NR 2.4507% *Note NR, not rated Next the probability of default on the PABs and TIFIA loan s was obtained from the results of the Simulated Cash Flow Model A trial and error procedure was used to identify the confidence interval that corresponds with the probability of default. As indicated in Table 4 6, the probability of defa ult for the projects varies greatly. For the PABs, the average probability of default based upon traffic forecasting accuracy was 16.80% and the default rate implied by the credit rating was 5.88%. The hypothesis that bond ratings accurately reflect the underlying default risk based on traffic demand is false since the default risk based on traffic forecasting accuracy is nearly three times greater than would be anticipated by investors. Table 4 6. Proba bility of default PAB s without optimism bias PAB n ( 1.00, 0.34) Project Debt t erm (years) Recommended BBB default per year Recommended BBB default risk Traffic f orecast probability of default I 495 40 0.012% 0.49% 4.00% LBJ 28 0.245% 6.86% 28.00% NTE 30 0.245% 7.35% 7.00% ERC 30 0.245% 7.35% 26.0 0%

PAGE 98

98 Table 4 6. Continued PAB n (1.00, 0.34) Project Debt t erm (years) Recommended BBB default per year Recommended BBB default risk Traffic forecast probability of default I 95 30 0.245% 7.35% 19.00% Average 5.88% 16.80% Table 4 7. Probability of de fault on TIFIA loans without optimism bias TIFIA Loan n (1.00, 0.34) Project Debt term (years) Recommended BBB default per year Recommended BBB default risk Traffic forecast probability of default I 495 40 0.012% 0.49% 4.00% LBJ 38 0.245% 9.31% 28 .00% NTE 40 0.245% 9.80% 9.00% ERC 40 0.245% 9.80% 26.00% I 95 35 0.245% 8.58% 19.00% Average 7.60% 17.20% This situation is similar with the TIFIA loan s as depicted in Table 4 7. The probability of default based on t he ability to predict traffic was more than double the default rate implied by a Baa3 rating. Interestingly, I 495 had a higher credit rating than the other projects and the Simulated Cash Flow Model concurred that this project had a much lower probability of default than the other pr ojects. Research Question 2 If optimism bias is assumed to be present in traffic forecasting, how accurately do bond ratings for PPP projects reflect risk, based on traffic demand? It is hypothesized that bond ratings do not accurately reflect the underlying default risk based on traffic demand, if optimism bias exists in forecasting. The exact same procedure used to answer research question 1 is also used to answer research question 2. This time howe ver, the traffic distribution was adjusted to traffic forecast distribution. Instead of assuming the tr unbiased (u=1 .00 sd= 0 .34), it was assumed that the results are skewed (u= 0 .77, sd= 0 .26). A trial and error method was run in the same manner as before and the res ults are summarized in Table 4 8 and 4 9

PAGE 99

99 Table 4 8 Probability of d efault on PABs with optimism bias PAB N(0.77, 0.26) Project Debt term (years) Recommended BBB default per year Recommended BBB default risk Traffic forecast probability of d efault I 495 40 0.012% 0.49% 8.00% LBJ 28 0.245% 6.86% 54.00% NTE 30 0.245% 7.35% 14.00% ERC 30 0.245% 7.35% 52.00% I 95 30 0.245% 7.35% 36.00% Average d 5.88 % 32.80% Table 4 9. Probability of d efault on TIFIA loans with optimism bias TIFIA Loan N(0.77, 0.26) Project Debt term (years) Recommended BBB default per year Recommended BBB default risk Traffic forecast probability of default I 495 40 0.012% 0.49% 8.00% LBJ 38 0.245% 9.31% 54.00% NTE 40 0.245% 9.80% 19.00% ERC 40 0.245% 9 .80% 52.00% I 95 35 0.245% 8.58% 36.00% Average 7.60 % 33.80% As shown in the results, if optimism bias exists in traffic and revenue forecasting in PPP projects, the probability of default based o n traffic forecasting accuracy was much higher than th suggested probability of default. The hypothesis that bond ratings do not accurately reflect the underlying default risk based on traffic demand if optimism bias exists in PPP traffic forecasting is true. PABs were more than four times riskier than would be expected based on their credit rating and TIFIA loans were more than four times riskier than would be expected based on the senior debt credit rating. I n addition when optimism bias was assumed, the default rate for both the PA Bs and TIFIA loans double d from roughly 17% to 33%. Research Question 3 What is the range and average base case risk premium a concessionaire is willing to accept to undertake a PPP project? It is hypothesized that all projects will demand a similar r isk premium for returns to equity and that the base case equity risk premium will be 5%.

PAGE 100

100 To answer this research question a n equity risk premium (Rp) was calculated for each project and summarized in Table 4 10 A risk premium is the additional compens ation an investor requires in order to accept a risk. Comparing equity risk premiums is better than comparing equity IRRs, because equity IRRs fail to account for the different investing climate in different years. The equity risk p remium is calculated b y subtracting the risk free rate from the equity IRR. The equity risk premiums ranged from LBJ with a low of 5.03% to I 95 with a high of 12.07%. This demonstrates that either I 95 was deemed a riskier project than LBJ, or the procurement process enabled the concessionaire to earn a higher profit for I 95. The average equity risk premium for all projects was 7.66% which was higher than the hypothesized 5.00% equity risk premium. Table 4 10. Equity risk premium for selected cases Project E(Ri)% Rrf% Rp% I 495 10.69 4.76 5.93 LBJ 9.15 4.12 5.03 NTE 10.44 4.45 5.99 ERC 12.42 3.16 9.26 I 95 14.63 2.56 12.07 Average 11.47 3.81 7.66 Research Question 4 How do the case study projects compare in terms of returns and risk? It is hypothesized that the proj ects will be similar in terms of return and risk. To answer this question, the equity risk premium versus probability of exceedance for each project was graphed on the same chart, Figure 4 1. An equity risk premium was calculated from the equity IRR to a ccount for different investing climates at different points in time. In general, it was observed that as the probability of exceedance increases, the equity risk premium decrease d ; simply stated this means a project is unlikely to receive a high equity ri sk premium with a high level of certainty. It was observed that some projects, such as I 95, clearly receive a

PAGE 101

101 high equity risk premium for every probability of exceedance; this is fairly obvious because the I 95 line dominates all others. Other than I 9 5 it was difficult to determine if one project received a higher equity risk premium than another project because the equity risk premiums varied depending on the probability of exceedance so the graphed lines cross over the confidence interval. Figure 4 1. Equity risk premium for selected cases The volatility associate d with the equity risk premium is indicated by the slope of the lines in Figure 4 1; volatility was easier to identify. While I 495 has one of the lowest equity risk premium s, it also h as the flattest slope indicating lower volatility. It can be seen that I 495 has the lowest volatility followed by NTE and LBJ the highest volatility. Research Question 5 If optimism bias is assumed to be present in traffic forecasting, what is the base c ase equity risk premium a concessionaire is willing to accept to undertake a PPP project? It is hypothesized that premium will be near 0%.

PAGE 102

102 To answer this qu estion a n equity risk premium was calculated for each project assuming optimism bias, and summarized in Table 4 11. Table 4 11 Equity risk premiums for selected cases assuming optimism bias Project E(Ri)% Rrf% Rp% I 495 8.59 4.76 3.83 LBJ 2.41 4.12 1.71 NTE 5.82 4.45 1.37 ERC 6.10 3.16 2.94 I 95 7.27 2.56 4.71 Average 6.04 3.81 2.23 If optimism bias is assumed in PPP traffic forecasting, the average equity risk premium drop ped 5.43% from 7.66% to 2.23%. An equity risk premium of 2.23% i s highe r than the hypothesized equity risk premium of 0.00%. Optimism bias did not have as large of an effect on the equity risk premium as the researcher predicted. The effects of optimism bias on the base case equity ris k premiums are summarized in Table 4 12 Table 4 12 Effects of optimism bias on base case Project Rp without optimism bias % Rp with optimism bias % Effect of optimism bias on Rp % I 495 5.93 3.83 2.10 LBJ 5.03 1.71 6.74 NTE 5.99 1.37 4.62 ERC 9.26 2.94 6.32 I 95 12.07 4.71 7.36 A verage 7.66 2.23 5.43 Optimism bias had the effect of lowering the equity IRR and equity risk premium across the entire confidence interval. The effect of this on I 495 can be seen in Figure 4 2. Graphs for each of the selected cases can be found in App endix E

PAGE 103

103 Figure 4 2 Effects of optimism bias on I 495 Research Question 6 As risk are transferred from the public to private partner (availability payment to direct tolling), does the equity risk premium increase? It is hypothesiz ed that the equity ris k premium associated with direct toll projects will be larger than the equi ty risk premium associated with availability payment projects. To answer this question it was necessary obtain the equity IRR for availability payment (AP) PPP projects. This in formation was obtained from the Value for Money Analysis on Port of Miami Tunnel and I 595 Express both reports prepared by Jeffery A. Parker and Associates for the FDOT, and displayed in Table 4 13 Both Florida projects are AP type and were procured aro und the same time as the case studies. Table 4 1 3. Availability p ayment PPP projects equity risk premium, risk free rates, and risk premium AP pr oject E(Ri)% Rrf% Rp% POMT 11.30 4.31 6.99 I 595 11.54 3.64 7.90 Average 11.42 3.98 7.45

PAGE 104

104 The equity risk premium was calculated for these two projects in the same manner as the case study projects. The average equity risk premium for AP projects was found to be 7.54%. This is very similar to the equity risk premium found in direct toll pr ojects without opt imism bias, 7.66% and significantly higher than the equity ris k premium with optimism bias, 2.23%. When graphed with the case study projects, it was observed that the AP project risk premiums match closely with the other projects at a probabilit y of exceedance of 0.5. Figure 4 3. Risk premium comparison of direct toll and AP p rojects It was hypothesized that the equity risk premium s for AP projects would be lower than the direct toll risk premiums because concessioners would demand a higher r isk premium for accepting traffic demand risk. The equity risk premium earned for accepting traffic revenue risk is 0.21%, Rp(Tolls) Rp(MAP) or 7.66% 7.45% = 0.21% Research Question 7 Do projects with a higher potential for default require a higher equity risk premium? It is hypothesized that as the probability of default increases the equity risk premium required by the concessionaire will also increase.

PAGE 105

105 A graph was made to comp are the probability of a TIFIA s chedule change with the equity risk premium. A TIFIA schedule change is a good indication of risk to equity but not necessarily debt. The graph shows that no clear relationship exists between the probability of a TIFIA Schedule change and the equity risk premium. The only clear relationsh ip indicated by the graph is that the equity risk premium is lower if optimism bias exists. Table: 4 14 Equity risk premium and TIFIA scheduled change on selected c ases Debt Risk N(1.00, 0.34) Debt Risk N(0.77, 0.26) Project TIFIA schedule change Rp TIF IA schedule change Rp I 495 17.00% 5.93 32.00% 3.83 LBJ 45.00% 5.03 76.00% 0.00 NTE 22.00% 5.99 45.00% 1.37 ERC 33.00% 9.26 62.00% 2.94 I 95 21.00% 12.07 48.00% 4.71 Figure 4 4. Equity risk premium and scheduled change on selected cases

PAGE 106

106 CH APTER 5 CONCLUSION The following summary is a brief recap of the entire study. The scarcity of public funding, coupled with crumbling infrastructure, and increasing demand for higher capacity and infrastructure improvements throughout the U.S. led to inc reased usage of PPP as an alternative method of financing roads. PPPs however have a somewhat mixed track record; an unusually high number of PPPs have defaulted on debt due to lower than expected traffic and revenue. This has not slowed the growth of PP P as a delivery method; more projects with a higher value p er project are under procurement now than ever before. These observations le d the researcher to investigate potential losses to debt and equity due to the inaccuracies of traffic and revenue forec asting and identify the pr obability of default in PPP projects. Next, a literature review was conducted to investigate the problem of lo ss and default in PPP projects. The literature review covers five main research threads, which at first may appear not to be connected but they in fact rely upon each other to finance a project: infrastructure funding legislation, project finance structure, risk in PPP projects, traffic and revenue forecasting, and credit ratings. The following main observations were mad e : 1. Infrastructure funding shortfalls have generated legislation for promoting the growth of the PPP mechanism as an alternative delivery method for roads by providing low cost funding and assistance for PPP projects. 2. The majority of PPP failures can be at tributed to inadequate or nonexistent feasibility studies including unrealistic traffic forecast and undefined public funds. 3. Flyvbjerg found that traffic forecasting ability was poor; furthermore, traffic forecasts are wrong by a large margin. Additionall prediction of traffic explained by incentives in funding procedures. Finally, he found that traffic forecasting ability has not improved over time. 4. Bain found that traffic on tolled roads was normally dist ributed and considerably over predicted. He also found that traffic forecast do not improve after the first year.

PAGE 107

107 After conducting a compre hensive literature review the purpose of the study, hypotheses and research questions were developed. The purpose of this study is to promote a greater understanding of the potential losses to debt and equity due to the inaccuracies of traffic and revenue forecasting and identify th e probability of default in PPP projects. The following hypotheses and research questi ons were addressed in this study: 1. How accurately do bond ratings for PPP projects reflect risk, based on traffic demand? It is hypothesized that bond ratings accurately reflect the underlying default risk based on traffic demand, assuming no optimism bia s in forecasting. 2. If optimism bias is assumed to be present in traffic forecasting, how accurately do bond ratings for PPP projects reflect risk, based on traffic demand? It is hypothesized that bond ratings do not accurately reflect the underlying defau lt risk based on traffic demand, if optimism bias exists in forecasting. 3. What is the range and average base case risk premium a concessionaire is willing to accept to undertake a PPP project? It is hypothesized that all projects will demand a similar risk premium for returns to equity and that the base case equity risk premium will be 5%. 4. How do the case study projects compare in terms of returns and risk? It is hypothesized that the projects will be similar in terms of return and risk. 5. If optimism bias i s assumed to be present in traffic forecasting, what is the base case equity risk premium a concessionaire is willing to accept to undertake a PPP project? It is se case equity risk premium will be near 0%. 6. As risk are transferred from the public to private partner (availability payment to direct tolling), does the equity risk premium increase? It is hypothesized that the equity risk premium associated with direct toll projects will be larger than the equity risk premium associated with availability payment projects. 7. Do projects with a higher potential for default require a higher equity risk premium? It is hypothesized that as the probability of default increases the equity risk premium required by the concessionaire will also increase. Next a methodology was d evis ed to answer the research questions. Five U.S. PPP highway projects were selected for the study based on project type, availability of data and project location. The cases are I 495 HOT Lanes, VA; North Tarrant Express, TX; I 635 LBJ Managed Lanes, TX; Elizabeth River Crossing, VA; and I 95 Express Lanes, VA. Based on

PAGE 108

108 Traffic Distribution Instrument was created to predict the forecast ed traffic base case forecast and a confidence interval supplied by the user. Two separate tra ffic distributions were created, one assuming no optimism bias (u=1.00, sd=0.34) and one assuming optimis m bias (u=0.77, sd=0.26) A large portion of the investigation was creating the Simulated Cash Flow Model for each case study to and distributed each year. When a confidence interval is entered by the user into the Traffic Distribution Instrument, the traffic forecast s from each year are adjusted to fit the normal probability distribution function. These are then automatically entered into the Simulated Cash Flow Model The Simulated Cash Flow Model then returns the equity IRR for the selected confidence interval The probability of default was identified through a trial and error process and manually looking for the project to default on debt. The results were presented in Chapter 4 and are used to draw the following conclusions. Summary of Conclusions Default Risk Research questions 1 and 2 investigated the probability of default on project debt. The results of these two questions clearly demonstrate that cre dit ratings do not accurately reflect traf fic revenue risk in PPP projects and it appears that there is a general underestimation of default risk in PPP projects by the credit rating agencies. Based solely on the ability to forecast traffic, it was found that PPP projects are two to four times riskier than an investor would expect Because the credit ratings did not a ccurately reflect the risk of default based on traffic forec asting accuracy without optimism bias it w as not possible to attribute the discrepancy between credit ratings and traffic forecasting risk to optimism bias.

PAGE 109

109 Traffic Revenue Risk Effects on Equity Research questions 3, 4 and 5 investigated the risk to equity. It was found that c oncessionaires earned an average equity IRR of 11.76%, and average equity risk premium of 7.66% in the base case scenario. These numbers appear to be in line with other investments. When the accuracy of traffic forecasting was accounted for however, returns diverged wi ldly and in the event of lower than expected traffic, profits eroded quickly. Because of this vast diversion it wa s difficult to identify projects with superior returns ; it was however possible to identify less volatile projects. Ability of C oncessionaire s to Identify R isks Research questions 6 and 7 investigate the pricing of risk. It does not ap pear that concessionaires are able to identify or effectively price traffic revenue risk but instead relay heavily upon the base case traffic forecast. No corre lation was found to exist between the equity risk premium and the probability that traffic rev enue would necessitate a TIFIA s cheduled debt change. It would generally be assumed that projects with a higher risk of failing to meet the debt repayment schedu le would demand a higher equity risk premium. In addition the equity risk premium charged on direct toll projects was comparable to the equity risk premium charged on availability paymen t projects It would be assumed that when a concessionaire takes on a dditional risk in the form of traffic revenue risk, a premi um would be charged to cover the risk. Since the premium charged to bear this risk is only 0.21%, it would indicate that concessionaires are confident in their traffic forecast and do not regard t raffic revenue risk as a large risk. Effect of Optimism Bias Optimis m bias was inv estigated with two pairs of research questions, research questions 1 and 2 and research questions 3 and 5. It was found that if the procurement process results in

PAGE 110

110 traff ic forecast with optimism bias, than the risk of default on debt is doubled and the equity risk premium is reduced by over 5 % Because of the limited scope of the study and high volatility in traffic forecasting, it was not possible to determine if optimi sm bias is responsible for default in PPP projects. When looking at both debt and equity the results of volatility in traffic forecasting were large enough to explain the high rate of default without optimism bias. It was possible to determine however that if optimism bias is in fact present in the traffic forecast, t he effects are large enough to have a serious negative impact on both debt and equity. It was not possible to determine if optimism bias or a general underestimation of risk wa s the leadin g cause of default in major U.S. highway PPP projects. Discussion Debt in PPP Projects It appears that there is a general underestimation of risk in PPP projects by the credit rating agencies. Even if it is assumed that the procurement process does not cr eate optimism bias, the cases analyzed are several times riskier than would be implied by their credit ratings. With three out of seven PPP projects in default, or on the verge of default, it seems very unlikely that these projects have a default rate of 0.245% per year as suggested by a BBB credit rating The default rate o f 16.88% without optimism bias, or 32.80%, with optimism bias, as predicted by the Simulated Cash Flow Model based on traffic foreca sting accuracy seems to fit the, albeit small, real world sample better. Credit ratings are not based on traffic demand alone but include a host of other factors such as governance and maintenance and legal framework; a credit rating r factors but was based solely on the traffic demand as analyzed in the Simulated Cash Flow Model It is possible the other factors considered by the credit rating agency could lessen the risk to project but it is highly doubtful that any factor could ove rcome such a significant lack of revenue.

PAGE 111

111 This underestimation of risk has significant implications because the problem will not be fully vetted for years to come until construction is finished and the projects begin collecting tolls. The industry is ju st beginning to understand this problem; since this research began, there has been a slow and steady march away from direct tolls to an availability payment model. At this point in time there are no direct toll projects under procurement because there are no private investors willing to accept a direct toll. The demise of the direct toll model has negative impacts upon the PPP industry; a major benefit in PPP was believed to be the ability to better ass ess risk, including traffic revenue risk. Over the years the government has frequently developed projects years before the demand necessitated it. Because the se projects were developed by the government the losses were spread over a large number of people and went unnoticed ; this is someti prevent this from happening because the private sector would refuse projects which would result in a loss. This has turned out not to be the case and the private sector has been little bet ter at identifying traffic revenue risk than the public sector. The switch to an availability payment method places traffic revenue risk firmly with the public partner. This is akin to admitting that the private sec tor cannot manage traffic revenue risk and removes any incentives to pr Equity in PPP Projects The overall equity IRR and equity risk premium are in line with other industries but the volatility from traffic forecasting associated with these projects is alarming. Just as risk was under estimated in debt, it is also under estimated in e quity. The fact that the AP projects and direct tolls have about the same equity risk premium indicates over confidence in the traffic forecast on the part of the concession aire. In reality traffic revenue risk is a substantial risk and

PAGE 112

112 the concessionaire should earn a substantial premium over AP projects for taking on this risk; this is clearly not happening. Risk may be underestimated in each of these projects as a result of compet ition. The concessionaire that wins the bid may be the concessionaire most willing to underestimate the his bid must be lower than the competition even if the competition continu ally underestimates risk. With the cost to develop a proposal at over a million dollars, a concession never win a bid. I t only requires one concessionaire to underestimate risk and all concessionaires are stuck hav ing to underestimate risk. This may be why we see a movement away from direct tolling to availability payments instead of a significant increase in equity risk premiums by all of the concessionaires It would be difficult to get all bidders to raise thei r equity risk premium significantly because they may lose the bid. with traffic forecasting means each project may ha ve a wide range of returns. It i s also difficult to identify the most profitable project because it is difficult to tell if the differences in equity risk premium are due to one project earning higher profits than another project or a real additional risk associated with a particular project. The PPP m arket has high barriers to entry and there are relatively few bidders so it is possible for a project to earn economic profits. This may explain why default risk did not correlate with equity risk premium; some projects did not have more risk, they just e arned more profit. Optimism Bias in PPP Projects It was not possible to determine if optimism bias is the cause of default in PPP projects because the case study sample was small and the volatility in traffic forecasting could also explain the defaults T he traffic forecasting distributions from Bain and Flyvb j erg are normally

PAGE 113

113 distributed however so in real life we should observe as many projects resulting in windfall profits as failing, if optimism bias does not occur. To my knowledge this has not happen ed. The study could not conclude that optimism bias was the cause of default, but it has shown that optimism bias has the potential for severe financial impacts on PPP projects. This should come as no great surprise when examining Table 3 2 ; the reductio n in traffic is obvious even before it enters the Simulated Cash Flow Model It takes little imagination to predict the impact of lowered traffic on cash flows. Recommendations for Practice Since a general underestimation of risk based on traffic forecast ing accuracy has been found in U.S. PPP h ighway projects it is recommended that the direct toll projects currently under construction apply a top down model approach as used in this study to examine the investment risk before the projects enter ope ration s. It is further recommend ed that the TIFIA program employ t he same approach to analyze their investment risk exposure in the ir portfolio of infrastructure projects Recommendations for Further Study 1. The traffic forecasting accuracy distribution used in this st udy was developed by Dr. Robert Bain. In order to conduct this study it was assumed that the projects analyzed are good fits with the projects used to develop the distribution. The st udy would be more accurate if o the projects most closely matching U.S. PPP highway projects. 2. The Simulated Cash Flow Model for this study was created in Excel. The study would benefit by being transferred from a static model in Excel to a dynamic model in a risk modeling software su ch as Crystal Ball. 3. Several PPP projects that have default have been considered lossless given default. It would be beneficial to examine not only the probability of default, as done in this study, but also the losses to investors given default.

PAGE 114

114 APPEN DIX A TIFIA PROJECT PROFILES The following project profiles are provided by the FHWA website for quick reference. S.R. 125 So. Bay Express (California) 11 Project Description The South Bay Expressway (SBX) toll road (the SBX Project) is a 9.2 mile, privat ely developed southern extension of SR 125, extending from San Miguel Road in Bonita, CA near the Sweetwater Reservoir to SR 905 in Otay Mesa, near the International Border. The SBX Project connects the only commercial port of entry in San Diego to the reg ional freeway network. This project, made possible through an innovative public private partnership, completes the missing link in San Diego's third north south freeway corridor. The SBX Project connects Otay Mesa, the largest area of industrial zoned land remaining in San Diego County, with eastern Chula Vista and points north and east, reducing commute times and providing convenient access to downtown San Diego, Sorrento Valley, Santee, I 8 and I 15, and Mexico. The SBX Project was developed pursuant to California's AB 680 legislation passed in 1989. Under the original franchise agreement, the private developer raised capital for the Project and constructed the road in exchange for a 35 year toll concession. Caltrans owns the highway, but leases the road back to the franchisee. Currently, the San Diego Association of Governments (SANDAG) has the franchise, under an amended agreement executed when the toll road was sold to SANDAG in December 2011 (see discussion below for more information).Control will reve rt back to Caltrans in 2042. In conjunction with the construction of the toll road, two local government funded projects at the northern end of the toll road known as the "Gap and Connector" were built to link the SBX Project to the existing San Diego fre eway network. The SBX Project offers cash and credit card payment as well as electronic toll collection through the FasTrak system. Construction of the toll road and the "Gap and Connector" projects was performed under design build contracts. Location : Sa n Diego County, California Project Sponsor / Borrower : Caltrans, South Bay Expressway, LLC (formerly South Bay Expressway, L.P.) Fiscal Year Approved: Fiscal Year 2003, amended and restated in 2011 (see below for further information) Mode: Toll Highway C ost: $658 million 11 Source: http://www.fhwa.dot.gov/ipd/project_profiles /

PAGE 115

115 Funding Sources Construction Period Financing: Bank debt $340 million (backed by toll revenues) TIFIA loan $140 million (backed by toll revenues) Donated right of way $48 million Investor equity $130 million Project Delivery / C ontract Method: 35 year Build Transfer Operate franchise with the State of California that allows the franchisee to set market rate tolls Private Partner: South Bay Expressway, L.P. (SBX LP), formerly owned by Macquarie 125 Holdings, Inc. and Macquarie In frastructure Partners. (These entities were the original developer and equity holders, respectively. See below for current ownership.) Project Advisors / Consultants Nossaman, Guthner, Knox & Elliott, LLP Special Counsel to SBX Milbank, Tweed, Hadley & McCloy, LLC Legal Counsel to SBX Salomon Smith Barney Financial Advisor to SBX Parsons Construction Manager Orrick, Herrington & Sutcliffe LLP Lenders' Legal Counsel Wilbur Smith Associates Borrower's Traffic Consultant AECOM/Maunsell Tolling Advisor to SBX Louis Berger Lenders' Traffic Advisor ARUP Lenders' Technical Advisor USDOT TIFIA JPO Advisors TIFIA Financial Advisor: Montague DeRose and Associates, LLC/TransTech Management, Inc. TIFIA Legal Counsel: Nixon Peabody LLP Lenders: Bank lenders (syndicated group of 10 banks); USDOT TIFIA Duration / Status: Opened to traffic in November 2007 TIFIA Credit Assistance Direct Loan: $140 million The TIFIA loan is secured by a priority security interest in all project collateral, including, but not limited to: (a) all income, tolls, revenues, rates, fees, charges, rentals, or other receipts derived by or related to the operation or ownership of the project including all amounts from joint development or leasing of air space lease rights; (b) any revenues assigned to the Borrower and proceeds of the sale or other disposition of all or any part of the project; and (c) all income derived from permitted investments. The TIFIA loan is also secured by a mortgage on the Borrower's leasehold interest in the real estate underlying the toll road right of way.

PAGE 116

116 Innovations $140 million TIFIA loan was the first ever provided to a private toll road development and the first with bank debt and private equity. The original TIFIA debt service repayment structur e was sculpted with mandatory and scheduled components. Related Links / Articles http://wwwcf.fhwa.dot.gov/exit.cfm?link=http://www.southbayexpressway.com/ Contact s South Bay Expressway info@southbayexpressway.com Tel: (619) 710 4021 Samuel Johnson Director of Operations 619 710 4021 samuel.johnson@sandag.org Pr oject Financial Status / Financial Performance Financial close May 22, 2003 TIFIA credit agreement signed May 22, 2003 On March 22, 2010, the privately owned toll road operator and TIFIA borrower, SBX LP, applied for reorganization under Chapter 11 of th e U.S. Bankruptcy Code. With accrued interest, the outstanding balance of the TIFIA loan at the time of the bankruptcy filing was $172 million and, pursuant to TIFIA statutory requirements, TIFIA's debt became on par with that of the Lenders. The senior li en of TIFIA and the Lenders was confirmed by the Court during the bankruptcy process. The filing was primarily the result of the burden of claims by the contractor that built the SBX Project, particularly the ongoing litigation costs. The SBX Project's fin ancial condition was also a factor as the financial prospects were being impacted by lower than anticipated revenues due to the economic downturn. On December 30, 2010, SBX LP filed a Plan of Reorganization (Plan) with the Bankruptcy Court, pursuant to whi ch SBX LP was converted to a Delaware limited liability company, South Bay Expressway, LLC (SBX LLC), and the debt of the Lenders and TIFIA was restructured. The Bankruptcy Court confirmed the Plan on April 14, 2011, which included the settlement of all li tigation matters with the contractor, Caltrans, and certain other parties. Under the Plan, TIFIA's secured claim was $99 million, of which approximately $93 million represented debt (the new loan amount) and $6 million was equity. TIFIA's unsecured claim w as $73 million, or 42 percent of the $172 million outstanding balance. All future toll revenues were to be shared pro rata between TIFIA (32 percent) and the Lenders (68 percent). The Lenders and TIFIA held 100 percent of the restructured debt and owned al l of the equity in the reorganized company. Although DOT wrote down a portion of the principal balance, TIFIA was scheduled to recapture more than 90 percent of the original loan by the final maturity date of

PAGE 117

117 2042. The reorganized company, SBX LLC, emerged from bankruptcy on April 28, 2011, concurrent with the financial close of the restructured loans. Soon after emergence, San Diego Association of Governments (SANDAG) approached TIFIA and the Lenders with respect to a possible purchase of the SBX Project b y SANDAG. On July 22, 2011, SANDAG, the Lenders and TIFIA reached an agreement in principal for the purchase of the SBX Project for $344.5 million in cash and debt (excluding cash on hand and non core assets). On December 21, 2011, SANDAG purchased the SBX Project from TIFIA and the Lenders, with TIFIA issuing a note to SANDAG for a restated loan in the amount of $94.1 million. In addition, as consideration for the sale of the project, TIFIA received a cash distribution of $15.4 million and holds a subordin ated note from SANDAG in the amount of $1.4 million. The basis for allocations between the Lenders (68 percent) and TIFIA (32 percent) was the pro rata share of the outstanding debt as of the bankruptcy filing. The TIFIA note has a senior lien on the SBX P roject Revenues and is structured into three tranches that bear interest at the same rates as in the Plan, which rates are higher than the rate for TIFIA's original loan for the SBX Project. The DOT also has a separate subordinate note, which compensates T IFIA in part for its equity portion under the Plan. Fitch Ratings has assigned an investment grade rating to the TIFIA debt. Now that substantially all of the assets (i.e., the SBX Project) of SBX LLC have been sold to SANDAG, TIFIA and the Lenders are in the process of liquidating and winding down SBX LLC. The ultimate recoveries of the TIFIA loan for this Project depend on ongoing performance of the Toll Road. However, the credit quality of the cash flow stream has been improved significantly through the sale of the Toll Road to SANDAG. Although the principal amount of the original loan was reduced, based on the credit attributes of the restructured loan and the higher interest rates (compared to the 4.46 percent rate in the original loan), the TIFIA progr am is positioned to realize 100 percent of the original loan balance.

PAGE 118

118 SH 130 segments 5 6 (Texas) Project Description SH 130 is a four lane, 91 mile toll road east and south of Austin designed to relieve congestion on the heavily traveled I 35, the prima ry north south route through Central Texas. Segments 1 4 of SH 130 (which are part of the Central Texas Turnpike System that includes SH 45 North and Loop 1) were constructed as a separate project and opened in stages between November 2006 and April 2008. On March 22, 2007, TxDOT signed a Comprehensive Development Agreement (CDA) with the SH 130 Concession Company to design, construct, operate, and maintain a 40 mile extension of SH 130 (Segments 5 and 6) under a 50 year concession from the date of opening (October 2012). The concession company is also financing the project and will share toll revenues with the state. The extension will follow the current US 183 alignment from north of Mustang Ridge to north of Lockhart and extend southwest to I 10 northeas t of Seguin. Location: Austin, Texas Metropolitan Area Project Sponsor / Borrower: Texas Department of Transportation (TxDOT) Fiscal Year Approved : Fiscal Year 2007 Mode: Toll Highway Cost: $1,327.9 million Funding Sources Senior bank loans $685.8 mill ion TIFIA loan $430 million Private equity $209.8 million Interest income $2.3 million Project Delivery / Contract Method: DBFOM (design, build, finance, operate and maintain) Private Partner: SH 130 Concession Company, LLC (joint venture of Cintr a Concesiones de Infraestructuras de Transporte, S.A. and Zachry American Infrastructure) Project Advisors / Consultants: AECOM/Maunsell Project's Traffic Consultant Capita Symonds Lenders' Traffic Auditor Atkins International Ltd. l Advisor To USDOT TIFIA JPO: TIFIA Legal Advisor: Nixon Peabody, LLP TIFIA Financial Advisor: Montague DeRose and Associates, LLC/High Street Consulting Group, LLC Lenders: Bank Syndicate, USDOT TIFIA

PAGE 119

119 Duration / Status: Construction began April 2009 Ope n to motorists October 2012 TIFIA Credit Assistance: Direct Loan: $430 million The TIFIA loan will be secured by a lien on Project Revenues subordinate to the lien securing Senior Lien Obligations, which will be bank loans, and will be senior to the equit y to be provided by investors Innovations: First, privately developed and operated open toll road facility in Texas Contacts: Cesar Diaz Plaza Chief Financial Officer SH 130 Concession Company (877) 741 3089 Project Financial Status / Financial Performan ce Financial close March 7, 2008 TIFIA loan agreement signed March 7, 2008 The first TIFIA interest payment is scheduled for June 2017. Principal repayments are scheduled to begin in 2018. The final maturity of the TIFIA loan is June 2047. A bank liquid ity facility and contingent equity will be available to meet senior and TIFIA debt service obligations in the first five years of operation. In addition a 12 month debt service reserve account will be established beginning in year six of operations and wil l be in place through the final maturity of the TIFIA loan.

PAGE 120

120 IH 635 Managed Lanes (Texas) Project Description The IH 635 Managed Lanes Project will relieve congestion north of Dallas on 13 miles of IH 635 (LBJ Freeway) from just west of I 35E (near Luna R oad) to just east of US 75 (near Greenville Ave.), and south on I 35E from I 635 to Loop 12. The project will involve: Reconstruction of the main lanes and frontage roads along IH 635 Addition of six managed lanes (mostly subsurface) along IH 635 from I 35 E to US 75 and four managed lanes west and east of that stretch Addition of six elevated managed lanes along I 35E from Loop 12 to the I 35E/IH 635 interchange The project is being built under a public private partnership (Comprehensive Development Agreeme nt [CDA]) between TxDOT and LBJ Infrastructure Group, which will operate and maintain the facility for 52 years. Construction is expected to take five years. The managed lanes will be dynamically priced after six months of an introductory fixed price sched ule. HOV2+ users will receive a 50 percent discount during peak operating periods. Tolls will be collected by the North Texas Tollway Authority. Location: Dallas Fort Worth Metroplex, Texas Project Sponsor / Borrower: Texas Department of Transportation ( TxDOT) LBJ Infrastructure Group, LLC (the Concession Company and TIFIA borrower) Fiscal Year Approved: Approved in Fiscal Year 2008. Conditional term sheet executed with the LBJ Infrastructure Group in January 2009. Mode: Highway / Managed Lanes Cost: $ 2,615 million (TIFIA eligible project costs) Funding Sources: Private Activity Bonds (PABs) $615 million TIFIA loan $850 million Equity contribution $664 million Toll Revenues $17 million Public Funds $490 million Project Delivery / Contract Me thod: DBFOM (design, build, finance, operate and maintain)

PAGE 121

121 Private Partner: LBJ Infrastructure Group, LLC Cintra Concesiones de Infraestructuras de Transporte, S.A. Meridiam Infrastructure Finance Dallas Police and Fire Pension System Other Private Part ners: Ferrovial Agroman, S.A. W.W. Webber, Inc. Bridgefarmer & Associates, Inc. Project Advisors / Consultants: Macquarie Capital (USA) Inc. Financial Advisor White & Case LLP Counsel to the Concession Company Bracewell & Giuliani LLP Counsel to the Concession Company Arup Sponsor's Traffic Consultant Hatch Mott MacDonald Lenders' Traffic Advisor To USDOT TIFIA JPO: TIFIA Legal Advisor: Hawkins Delafield & Wood, LLP TIFIA Financial Advisor: Montague DeRose and Associates, LLP/High Street Consultin g Group, LLC Lenders: Bondholders, USDOT TIFIA Duration / Status: Commercial close (CDA execution) September 4, 2009 Construction began January 18, 2011 Expected completion in early 2016 TIFIA Credit Assistance: Direct loan: $850.0 million. The TIFIA loa n will be repaid with project revenues, which include all income, tolls, revenues, rates, fees, charges, rentals, or other receipts derived by or related to the operation of the project. Innovations: When completed, this project will have one of the most comprehensive managed HOV lane systems in the country, deploying Automatic Vehicle Identification (AVI) technology capable of reading the transponders of passing vehicles Innovative financing package includes PABs and TIFIA credit assistance Gary Moonshowe r, P.E. Project Manager TxDOT P.O. Box 133067 Dallas, TX 75313 3067 Tel. (214) 320 4489 Andy Rittler Director of Corporate Affairs LBJ Infrastructure Group Tel. (214) 592 3144 info@lbjexpress.com Proje ct Financial Status / Financial Performance TIFIA credit agreement executed on June 21, 2010. Financial close for the PABs occurred on June 22, 2010.

PAGE 122

122 Downtown Tunnel/Midtown Tunnel/MLK Extension (Virginia) Project Description The Downtown Tunnel / Midtown Tunnel / MLK Extension consists of five components of construction involving three facilities in the Hampton Roads region of Virginia. The Midtown Tunnel portion consists of a new two lane tolled tunnel under the Elizabeth River parallel to the existing M idtown Tunnel connecting the Cities of Norfolk and Portsmouth as well as modifications to the existing tunnel to provide increased capacity for east west travel linking Route 58 and I 264 in Portsmouth to the interchange at Brambleton Avenue/Hampton Boulev ard in Norfolk. Modifications to the interchange are also planned. The planned improvements to the Downtown Tunnel will bring it into compliance with current fire and life safety standards. The MLK Extension portion of the project consists of extending U.S Route 58 south from London Boulevard, approximately 0.8 mile to I 264 with an interchange at High Street. The $2.1 billion project will be built on a design, build, finance, operate, and maintain (DBFOM) concession basis by Elizabeth River Crossings Opco LLC (ERC) comprised of Skanska Infrastructure Development and Macquarie Group. ERC will operate the concession for 58 years. Tolling of the existing Midtown and Downtown Tunnels to help finance the project will start in January 2014 Location: Cities of Norfolk and Portsmouth, Virginia Project Sponsor / Borrower: Project Sponsor: Virginia Department of Transportation (VDOT) Borrower: Elizabeth River Crossings Opco, LLC Fiscal Year Approved: Fiscal Year 2012 Mode: Highway / Tunnel Cost: Est. Total Cost: $ 2,089 million Funding Sources: Senior Debt (Private Activity Bonds) $675 million TIFIA loan $422 million Equity Contributions $272 million Public Funds $408 million Toll Revenues $268 million TIFIA Capitalized Interest $43 million Project Del ivery / Contract Method: DBFOM (design, build, finance, operate, and maintain) Project Partner: Elizabeth River Crossings Opco, LLC Skanska Infrastructure Development. Inc. Macquarie Financial Holding Limited Construction Joint Venture (design build ember s): Skanska USA Civil Southeast, Inc. Kiewit Construction Company Weeks Marine, Inc

PAGE 123

123 Project Advisors / Consultants: Project design: Parsons Brinckerhoff Volkert & Associates, Inc. COWI Systems Support Transdyn Federal Signal ETC design, installation, O&M To the Borrower: Macquarie Capital Funds Inc. Finance Steer Davies Gleave Traffic & Revenue McGuireWoods Consulting Public Affairs Orrick, Herrington & Sutcliffe Borrower's Legal Counsel Hunton & Williams Borrower's Local Counsel PBS&J Inte lligent Transportation Systems OR Colan Associates Right of Way Arup Equity TA/Asset Condition Survey Atkins Lender's TA and Traffic Advisor Marsh Insurance To USDOT TIFIA JPO: TIFIA Legal Advisor Katten Muchin Rosenman LLP TIFIA Financial Adviso r Scully Capital Lenders: Bondholders, USDOT TIFIA Duration / Status: Comprehensive Agreement with ERC signed December 2011 (commercial close). Financial close: April 12, 2012 Construction start: mid 2012 Expected completion: 2017 Duration of concessio n: 58 years TIFIA Credit Assistance: Direct Loan: $422 million The TIFIA loan will be repaid with toll revenues. TIFIA is further secured by a fully funded debt service reserve fund. Innovations: Virginia's Commonwealth Transportation Board issued its fi rst GARVEE bonds under the Commonwealth of Virginia Federal Transportation Grant Anticipation Revenue Notes Act of 2011 to provide the public subsidy to support the project's private financing.

PAGE 124

124 Contacts: P3 project development and procurement: Ryan Pedra za Program Manager, Office of Transportation Public Private Partnerships 600 East Main, Suite 2120 Richmond, VA 23219 Tel: (804) 371 9870 ryan.pedraza@vdot.virginia.gov Project construction: Frank Fabian, PE 152 Tunnel Facility Drive Portsmouth, VA 23707 Tel: (757) 396 6800 frank.fabian@vdot.virginia.gov Project Financial Status / Financial Performance: TIFIA credit agreement executed on April 12, 2012

PAGE 125

125 Dulles Greenway (Virginia) Project Description The Dulles Greenway is a 14 mile, limited access highway extending from the State owned Dulles Toll Road which carries traffic between Washington's Capital Beltway and Dulles Airport to Le esburg. The two roads connect at a toll plaza. Drivers pay one toll, which the operators of the two facilities divide. The Greenway was privately financed and constructed from 1993 to 1995 as a DBFO and had an initial agreement to have operational responsi bilities revert to the Commonwealth of Virginia in 2036. To finance the Greenway, the limited private partnership, TRIP II put up $40 million in equity, and secured $310 million in privately placed taxable debt. Ten institutional investors led by CIGNA Inv estments, Prudential Power Funding Associates, and John Hancock Mutual Life Insurance Company provided $258 million in long term, fixed rate notes (due in 2022 and 2026). Three banks (Barclays, NationsBank, and Deutsche Bank AG) agreed to provide part of t he construction funding and $40 million in revolving credit. Loans were to be repaid with toll revenues, and the financing was secured by a first mortgage and security interest in the developer's right, title, and interest in the facility. When the Greenwa y opened to traffic in September 1995, traffic fell short of projected levels, and tolls were reduced. Users but not revenues increased. Tolls were increased in July 1997 and the Virginia General Assembly allowed the speed limit on the facility to be incre ased from 55 to 65 miles per hour. Still facing financial challenges, TRIP II restructured its debt in 1999 and agreed to an extension of the project. In 2001 the Virginia State Corporation Commission (SCC) extended TRIP II's concession period for an addit ional 20 years to 2056. In September 2004 variable peak and discounted off peak point to point rates were introduced to better manage peak period congestion. In August and September 2005 Macquarie Infrastructure Group (MIG, now Macquarie Atlas Roads) agree d to purchase TRIP II for $617.5 million. This included a payment of $84.5 million to Kellogg Brown & Root for its 13.3% share of the company, and $533 million to the Shenandoah Group, the family held company that held the remaining 86.7% of the company af ter having bought out Autostrade International's former 30% share in 2003. In December 2006, MIG completed the sale of 50% of its economic interest in the Dulles Greenway to Macquarie Infrastructure Partners (MIP) and subsequently MIG holds a 50% economic interest in the Greenway. The maximum toll schedule has been set by the SCC through to the end of 2012. From 2013 through to 2020 tolls can escalate annually at the higher of CPI plus 1%, real GDP, or 2.8% per annum. Post 2020 tolls are set by the SCC on a pplication. Location: Loudoun County, Northern Virginia Project Sponsor / Borrower: Toll Road Investors Partnership II (TRIP II) Mode: Toll Road ]Cost: $350 million

PAGE 126

126 Funding Sources: MIG's Purchase Macquarie raised financing for its investment in TRIP II through the placement of private stock in Australia. Macquarie also raised funds in New York through the float of $425 million worth of shares in the closed end Macquarie Global Infrastructure Total Return Fund. Macquarie used the monies generated from the se sales to make multiple purchases. TRIP II's 1999 Refinancing $332 million in AAA Bonds replacing all outstanding agreements were insured by MBIA and included: $35 million of current pay interest only bonds $297 million zero coupon bonds maturing in 20 03, 2005 with blended interest rate of approximately 7% TRIP II's original financing is discussed above under Description. Project Delivery / Contract Method: DBFOM (design, build, finance, operate, and maintain) Private Partner: Owner: TRIP II, a fully o wned subsidiary of Macquarie Atlas Roads and Macquarie Infrastructure Partners Operator: Autostrade International of Virginia O&M, Inc., a subsidiary of Italian based Autostrade S.p.A, the largest toll road operator in the world Private Investor Partner: TRIP II, a fully owned subsidiary of Macquarie Infrastructure Group and Macquarie Infrastructure Partners Project Advisors / Consultants: N/A Lenders: Bondholders Duration / Status: Opened September 1995 and has since been expanded from four to six lanes Innovations: One of first U.S. projects to embody the basic concepts of project revenue financing. Enabled by 1988 action of Virginia's General Assembly, authorizing private development of toll roads. The Greenway is the first toll road in greater Washingt on, D.C. to feature variably priced tolls. Contacts: Toll Road Investors Partnership II, L.P. (TRIP II) 45305 Catalina Court Suite 102 Sterling, VA 20166 Tel: (703) 668 0021 Project Financial Status Greenway owners began to default in 1996. Large ref inancing package completed in 1999. TRIP II Purchased by MIG in September 2005 for $617.5 million

PAGE 127

127 I 95 Express Lanes (Virginia) Project Description The I 95 Express Lanes will be the second major step in creating a regional network of tolled managed lanes in Northern Virginia. The project consists of the development, design, finance, construction, maintenance and operation of 29.4 miles of High Occupancy Vehicle (HOV)/High Occupancy Toll (HOT) Lanes along I 95 and I 395 corridor in Northern Virginia, from Garrisonville Rd. in Stafford County to Edsall Rd. in Fairfax County over a 76 year concession period. The project is divided into four segments: 8.3 miles of new construction two lane reversible (includes 7 new brides) 7.0 miles of two lane HOV conversi on two lane reversible 11.9 miles of two lane HOV conversion three lane reversible 2.2 miles of two lane HOV conversion three lane reversible (including connection to 495 Express Lanes at the Springfield Interchange) The new managed lanes will provide congestion relief and connectivity to users travelling to and from major employment centers in Northern Virginia, such as Tysons Corner and Washington, D.C., and five maj or military sites, including Ft. Belvoir, Quantico Marine Corps Base, and the Pentagon, while providing a reliable pathway for transit vehicles and carpools to travel throughout the region. In many areas, the project will provide first time, direct HOV and transit access to these destinations. Location: Fairfax, Prince William, and Stafford Counties, Virginia Project Sponsor / Borrower: Virginia Department of Transportation (VDOT) 95 Express Lanes LLC (the Concession Company and TIFIA borrower) Fiscal Year Approved: Fiscal Year 2012 Mode: Highway / Express Lanes Cost: Total Cost: $922.6 million (excluding $25.4 million in early development costs already incurred by VDOT) Funding Sources: TIFIA loan $300.0 million Private Activity Bonds $252.6 million C ommonwealth of Virginia Grant $82.6 million Private Equity $280.4 million TIFIA Capitalized Interest $6.5 million Interest Earnings $0.6 million A TIGER III (Transportation Investment Generating Economic Recovery) grant will be used to pay the su bsidy cost of the loan to the federal government. Project Delivery / Contract Method: DBFOM (design, build, finance, operate, and maintain) Project Partner: 95 Express Lanes LLC Fluor Enterprises, Inc. Transurban DRIVe

PAGE 128

128 Project Advisors / Consultants: Virg inia Department of Rail and Public Transportation Fluor Lane 95, LLC Design builder To the Borrower: Scotiabank Financial Advisor Orrick, Herrington & Sutcliffe LLP Legal Advisor To USDOT TIFIA JPO: Bryant Miller Olive, P.A. TIFIA Legal Advisor Tay lor De Jongh TIFIA Financial Advisor To Public Sector Sponsor: KPMG Financial Advisor Ballard Spahr LLP Legal Advisor ATCS Technical Advisor Lenders: Bondholders, USDOT TIFIA Duration / Status: FHWA FONSI December 2011 Private Activity Bonds Finan cial close July 2012 Construction began August 2012 Revenue service estimated to begin early 2015 Conclusion of concession 2087 TIFIA Credit Assistance: Direct Loan: $300.0 million Innovations: The Sponsors, together with VDOT, are also partners in the 495 Express Lanes project. The I 95 Express will be linked directly into 495 Express Lanes at the Springfield Interchange. The two projects have common traffic and tolling ma nagement systems and will share the same operations center/operator. Contacts: Larry O. Cloyed, PMP Senior Project Manager Virginia Department of Transportation Virginia Megaprojects Office 6363 Walker Lane, Suite 500 Alexandria, VA 22310 T el: (571 ) 483 2584 Larry.Cloyed@VDOT.Virginia.gov Project Financial Status / Financial Performance Private Activity Bonds sold July 2012 TIFIA credit agreement executed on November 20, 2012

PAGE 129

129 Port Of Miami Tun nel (Florida) Project Description The Port of Miami Tunnel will improve access to and from the Port of Miami, serving as a dedicated roadway connector linking the Port (located on an island in Biscayne Bay) with the MacArthur Causeway (State Road A1A which connects Miami to Miami Beach) and I 395 on the mainland. Currently the Port is linked to the mainland only by the Port Bridge. The tunnel will: (i) improve access to the Port helping to keep it competitive and efficient, (ii) improve traffic safety in downtown Miami by removing cargo trucks and cruise line buses from congested city streets, and (iii) facilitate ongoing and future development plans in and around downtown Miami. The project includes a tunnel under the Main Channel (the shipping channel between Dodge and Watson Islands), roadway work on Dodge Island and Watson Island/MacArthur Causeway and widening the MacArthur Causeway Bridge. Twin tubes, each 3,900 feet long and 41 feet in diameter, will reach a depth of 120 feet below the wat er. The project is being developed as a public private partnership with Miami Access Tunnel, LLC (MAT). The state has agreed to pay for approximately 50 percent of the capital costs (design and construction) and all operations and maintenance, while the r emaining 50 percent of the capital costs will be provided by the local governments. Under the concession agreement, FDOT will provide MAT a total of $100 million in milestone payments during the construction period between 2010 and 2013 and a $350 million final acceptance payment upon construction completion. This will be followed by 30 years of availability payments during the operating period. The annual payment will be $32.479 million (2009$), with adjustments for inflation. Deductions will be made from meet prescribed performance standards. Location : Miami, Florida Project Sponsor / Borrower : Florida Department of Transportation (FDOT) Miami Access Tunnel (MAT) Miami Dade County City of Miami Fisc al Year Approved: Fiscal Year 2010 Mode: Highway / Tunnel Cost: Total Project Cost: $1,113 million ($1,072.9 in eligible project costs) Design and Construction $607 million Financing and other capital costs $195.1 million SPV Costs/Insurance/O&M durin g construction $59.6 million Reserves $41.2 million State development cost $209.8 Funding Sources: Total Eligible Project Costs: $1,072.9 million Senior bank debt $341.5 million TIFIA loan $341 million Equity contribution $80.3 million FDOT mi lestone payments during construction $100 million

PAGE 130

130 FDOT development funds $209.8 million TIFIA capitalized interest during construction is not included in total eligible costs in the amount of $40.1 million Project Delivery / Contract Method: DBFOM (de sign, build, finance, operate, and maintain) Private Partner: Miami Access Tunnel, LLC (MAT) Meridiam Infrastructure Finance, S.a.r.l. (90% equity partner) Bouygues Travaux Publics, S.A. (10% equity partner) Project Advisors / Consultants: Barclays Capita l Macquarie Infrastructure To USDOT TIFIA JPO: TIFIA Legal Advisor: Hawkins, Delafield & Wood, LLP TIFIA Financial Advisor: Scully Capital Services, Inc. Lenders : USDOT TIFIA 10 bank club (senior bank debt): BNP Paribas Banco Bilbao Bizcaya Argentina RBS Citizens Banco Santander Bayerische Hypo und Vereinsbank, AG Calyon Dexia ING Capital Societe Generale WestLB Duration / Status : Commercial close June 2, 2009 Construction began May 2010; expected completion May 2014 TIFIA Credit Assistance : Direct loan: $341.5 million The TIFIA loan holds a second priority security interest in project revenues after senior obligations. The project's senior debt obligations will be fully amortized prior to commencement of TIFIA payments, providing TIFIA with a sole c laim on project cash flows available for debt service. Innovations: Second U.S. application of availability payments to finance a transportation project (the first also being in Florida the I 595 Corridor Roadway Improvements ) Contacts: Teresa Alvarez, P.E. District Consultant Management Engineer FDOT Tel: (305) 470 5142 Teresita.Alvarez@dot.state.fl.us Project Financial S tatus / Financial Performance Financial close and TIFIA credit agreement executed October 15, 2009 Availability payments are pledged to secure the TIFIA loan.

PAGE 131

131 I 595 Managed Lanes (Florida) Project Description The I 595 corridor was opened to traffic in 1 989, coordinating the movement of high traffic volumes between the developable areas in the western parts of the Southeast Florida region with the established north south freeway and principal roadways to the east, including I 75, Florida's Turnpike, SR 7, I 95 and US 1. However, travel demand within the corridor has increased at a pace where the long range traffic forecasts for the current highway would be reached in the short term. The I 595 Corridor Roadway Improvements project consists of the reconstruc tion and widening of the I 595 mainline and all associated improvements to frontage roads and ramps from the I 75/Sawgrass Expressway interchange to the I 595/I 95 interchange, for a total project length of approximately 10.5 miles. The project passes thro ugh, or lies immediately adjacent to, six jurisdictions: City of Sunrise; Town of Davie; City of Plantation; City of Fort Lauderdale; Town of Dania; and unincorporated areas of Broward County. A major component of the project is the construction of three a t grade reversible express toll lanes to be known as 595Express, serving express traffic to/from the I 75/Sawgrass Expressway from/to east of SR 7, with a direct connection to the median of Florida's Turnpike. These lanes will be operated as managed lanes with variable tolls to optimize traffic flow, and will reverse directions in peak travel times (eastbound in the AM and westbound in the PM). The project is being implemented as a public private partnership between FDOT and a private concessionaire to desi gn, build, finance, operate, and maintain the roadway for a 35 year term. FDOT will provide management oversight of the contract; will install, test, operate and maintain all tolling equipment for the express lanes; and will set the toll rates and retain t he toll revenue. Location: Broward County, Florida Project Sponsor / Borrower: Florida Department of Transportation (FDOT) Fiscal Year Approved: Fiscal Year 2008 Mode: Highway / Managed Lanes Cost: $1,833.6 million (present value in 2009 dollars, given a 5 % discount rate) total final acceptance and availability payments over the 35 year contract to design, build, finance, operate and maintain the roadway Funding Sources: State and federal resources Support FDOT's final acceptance payments ($686 million YO E) and availability payments ($65.9 million annual Maximum Availability Payment [MAP] in 2009 dollars) made to concessionaire (Federal aid receipts, state motor fuel tax receipts, toll receipts) Concessionaire's financing sources for repayment: Senior bank debt $781.1 million (backed by final acceptance/availability payments) TIFIA loan $603 million + capitalized interest during construction (backed by final acceptance/availability payments) Equity $207.7 million Revenues $10.0 million FDOT qualifyi ng development funds $232 million

PAGE 132

132 Project Delivery / Contract Method: DBFOM (Design, build, finance, operate, and maintain) Private Partner: I 595 Express, LLC (ACS Infrastructure Development and TIAA CREF (50/50 split of the equity portion on loan)) as Concessionaire Subcontractors/Subconsultants: Dragados USA Inc. Design build contractor AECOM Technical Services, Inc. Lead engineering firm HNTB Corp Construction engineering and inspection Roy Jorgensen Associates, Inc. Operations & Maintenance Project Advisors / Consultants: To Sponsor: Dewey & LeBoeuf LLP legal Macquarie Capital (USA) Inc. financial Scott Wilson, Plc. technical To Lender: Simpson Thacher & Bartlett LLP legal To Authority: Nossaman LLP legal Jeffrey A. Parker & Associ ates, Inc. financial Reynolds, Smith and Hills, Inc. technical The Corradino Group construction oversight To USDOT TIFIA JPO: TIFIA Legal Advisor: Hawkins Delafield & Wood LLP TIFIA Financial Advisor: Taylor DeJongh Lenders: 12 bank club (senior ban k debt) USDOT TIFIA Duration / Status: Construction began June 2009; expected completion summer 2014 TIFIA Credit Assistance: Direct loan: $603 million USDOT has a subordinate lien on availability payments made by FDOT to I 595 Express, LLC. Innovations : First U.S. application of availability payments to a transportation project. I 595 Express, LLC will receive no compensation from FDOT until the facility is fully operational. Upon FDOT's final acceptance of the project construction, I 595 Express, LLC w ill be eligible to receive a series of annual lump sum final acceptance payments, including potential incentive bonuses for completing a series of interim milestones (related to major construction activities) within established contractual deadlines.

PAGE 133

133 Perfo rmance based availability payments will be made monthly during the operating period of the project. A maximum availability payment of $65.9 million (in 2009 dollars) begins in 2014 and escalates annually. If quality and performance requirements stipulated in the contract as well as availability of the roadways to traffic are not met, then the availability payments will be subject to downward adjustment in accordance with the contract. Contacts: Paul A. Lampley, P.E. I 595 Construction Project Manager Flor ida Department of Transportation Tel: (954) 845 9550 paul.lampley@dot.state.fl.us Project Financial Status / Financial Performance TIFIA loan agreement executed March 2, 2009 Financial close reached on March 3, 2009 The first interest payment is sched uled for June 2014. Principal repayments are scheduled to begin in 2031. The final maturity of the TIFIA loan is June 2042. A six month debt service reserve based on senior and TIFIA interest and principal will be available until the final maturity of the TIFIA loan. A $9 million contingency reserve will be available until six months after scheduled substantial completion to cover construction cost overruns and help maintain target minimum DSCR.

PAGE 134

134 North Tarrant Express (Texas) Project Description On June 23 2009, TxDOT awarded two Comprehensive Development Agreements (CDAs equivalent to public private partnerships) for the North Tarrant Express project to NTE Mobility Partners. The Concession CDA for Phase 1 includes the design, development, construction, finance, maintenance, and operation of 13 miles along Interstate (IH) 820 and State Highway (SH) 121/SH 183 from IH 35W to SH 121, from north of Fort Worth to just southwest of Dallas Fort Worth International Airport. The duration of the concession is 52 years. The existing highway includes two general purpose lanes in each direction. Proposed improvements include three general purpose lanes in each direction with two managed lanes in each direction for a total of ten lanes with frontage roads for future t raffic volumes. The CDA for Segments 2 4 includes developing master plans for the remainder of the corridors along SH 183 from SH 121 to SH 161, IH 820 east from SH 121/SH 183 south to Randol Mill Road, and along IH 35W from IH 30 to SH 170 in Tarrant and Dallas counties, as well as other facilities for connectivity, safety, and financing. When all phases are completed, the Project will comprise 36 miles of managed lanes. Location: Dallas Fort Worth Metroplex Project Sponsor / Borrower: Texas Department of Transportation (TxDOT) NTE Mobility Partners, LLC (the Concession Company and TIFIA borrower) Fiscal Year Approved: Fiscal Year 2009 Mode: Highway / Managed Lanes Cost: $2,047 million Funding Sources: Phase 1: Private Activity Bond Proceeds $398 million TIFIA Loan $650 million Public Funds $573 million Equity Contribution $426 million Total does not include TIFIA capitalized interest of $54 million Project Delivery / Contract Method: DBFOM (design, build, finance, operate, and maintain) Private Pa rtner: NTE Mobility Partners, LLC (the Concession Company) Cintra Concesiones de Infraestructuras de Transporte, S.A. Meridiam Infrastructure Dallas Police and Fire Pension System Other Private Partners Ferrovial Agromn S.A. W.W. Webber, LLC Earth Tech, I nc.

PAGE 135

135 Maunsell Australia Proprietary Limited Aguirre & Fields, LP Ross Communications CSJ Engineering Assoc. Project Advisors / Consultants: J.P. Morgan Securities, Inc. Financial Advisor Macquarie Capital (USA) Inc. Financial Advisor White & Case LLP Counsel to the Concession Company Bracewell & Giuliani LLP Counsel to the Concession Company AECOM Enterprises Sponsor's Traffic Consultant Hatch Mott MacDonald Lenders' Traffic Advisor CH2M Hill, Inc. Independent Engineer To USDOT TIFIA JPO TIFIA Legal Advisor: Hawkins Delafield & Wood, LLP TIFIA Financial Advisor: Montague DeRose and Associates, LLP/High Street Consulting Group, LLC Lenders: Bondholders, USDOT TIFIA Duration / Status: Commercial close (CDA execution) June 23, 2009 Construction beg an October 2010; expected completion in 2015 TIFIA Credit Assistance: Direct Loan: $650.0 million The TIFIA loan will be repaid with project revenues, which include all income, tolls, revenues, rates, fees, charges, rentals, or other receipts derived by or related to the operation of the Project. Financial Status / Financial Performance: TIFIA credit agreement executed on December 16, 2009 Financial close occurred on December 17, 2009 Innovations: When completed, this project will have a state of the art electronic toll collection system with open architecture, ensuring a seamless, free flow operation of the managed lanes. Innovative financing package including PABs and TIFIA credit assistance. Only the second PABs issuance ever under the $15 billion of au thority provided to DOT by SAFETEA LU. The first transportation infrastructure project in the US to reach financial close with direct investment by a pension fund. Contacts: Mohammad Al Hweil, P.E. TxDOT North Tarrant Express Project Manager Tel: (817) 370 3512 Belen Marcos Chief Executive Officer, NTE Mobility Partners LLC Tel: (817) 710 0503

PAGE 136

136 Presidio Parkway (California) Project Description The Presidio Parkway project is a replacement of Doyle Drive, a 1.6 mile segment of Route 101 in San Francisco tha t is the southern access to the Golden Gate Bridge, connecting Marin and San Francisco counties and providing a major regional traffic link between the peninsula and North Bay Area counties. The current structure, built in 1936, does not meet current highw ay standards and is seismically deficient. The Presidio Parkway project area extends from the Golden Gate Bridge Toll Plaza on the west to Broderick Street on the east, and includes Richardson Avenue, Gorgas Avenue and Marina Boulevard. The Presidio Parkwa y will be a six lane facility with a southbound auxiliary lane between the Park Presidio Interchange and the new Presidio access at Girard Road. The roadway will consist of various sections (from the toll plaza east to Richardson Avenue) with a landscaped median: a high viaduct between the Park Presidio Interchange and the San Francisco National Cemetery (Presidio Viaduct) shallow cut and cover tunnels past the cemetery to Battery Blaney (Battery Tunnels) at grade roadways to the Main Post cut and cover tun nels from the Main Post to east of Halleck Street (Main Post Tunnels) a low causeway from Halleck Street to Girard Road at grade connection to Richardson Avenue The project is being developed in two phases. Caltrans is responsible for the design, financin g and construction of Phase I, currently nearing completion. Phase I, delivered through a traditional design bid build process, consists of a replacement bridge at the Park Presidio Interchange, the new southbound Presidio Viaduct, the southbound Battery T unnel, and a temporary bypass east of the Main Post to allow construction of the Main Post Tunnels and roadway to Richardson Avenue. Through a competitive procurement process, Caltrans selected a private consortium, the Golden Link Concessionaire, to deliv er Phase II as a design, build, finance, operate, and maintain (DBFOM) availability pay concession. The P3 Project Agreement with GLC was executed on January 3, 2011. GLC will receive milestone payments following substantial completion and quarterly availa bility payments through the concession period, based on performance. Major construction will begin in late 2012 to complete the remaining elements of the Presidio Parkway, including the northbound Presidio Viaduct and Battery Tunnel, the Main Post Tunnels, and the new Girard Road Interchange with a direct connection to the Presidio. This phase will also include final landscaping. Location: San Francisco, California Project Sponsor / Borrower: Project Sponsor: California Department of Transportation (Caltra ns), San Francisco County Transportation Authority Borrower: Golden Link Concessionaire, LLC (GLC) Fiscal Year Approved: Fiscal Year 2012 Mode: Highway Cost: Estimated Total Cost: $851.6 million

PAGE 137

137 Funding Sources: Phase I: $486.9 million Federal funds $70 .8 million ARRA grant $83.3 million State funds $229.0 million Local funds $103.9 million Phase II: $364.7million Bank Loan $166.6 million TIFIA Tranche A Loan $89.8 million TIFIA Tranche B Loan $60.2 million Parent Company Contribution $2.6 million Private Equity $43.0 million TIFIA Capitalized Interest $2.5 million Project Delivery / Contract Method: Phase I: Design bid build Phase II: DBFOM (design, build, finance, operate, and maintain) Project Partner: Phase II Golden Link Partners, LLC HOCHTIEF PPP Solutions North America Meridiam Infrastructure Construction Joint Venture (design build members): Flatiron West, Inc. Kiewit Infrastructure West, Co. Project Advisors / Consultants: Phase I Construction: C.C. Myers R&L Brosamer Phase II Project Design: HNTB Corporation To the Borrower: Scotia Capital, Inc. Financial advisor Milbank, Tweed, Hadley & McCloy LLP Legal Counsel Moore McNeil, LLC Insurance Consultant To USDOT TIFIA JPO: TIFIA Legal Advisor Bryant Miller Olive TIFIA Fin ancial Advisor Montague DeRose and Associates, LLP/High Street Consulting Group, LLC To Public Sector Sponsor: Sperry Capital, Inc. Financial Advisor KPMG Corporate Finance LLC Financial Advisor Nossaman, LLP Legal Advisor

PAGE 138

138 Lenders: Banks, USDOT TIFI A Duration / Status: Phase I: Construction to be completed Summer 2012 Phase II: P3 Agreement with Golden Link Concessionaire signed January 3, 2011 (commercial close). Pre construction began mid 2011, Financial close: June 14, 2012, Major construction st art: late 2012, Expected completion: September 2015, Duration of concession: 30 years TIFIA Credit Assistance: Direct Loan: $150.0 million TIFIA credit assistance totals $150 million and is structured into two tranches, to reflect the two distinct sources of repayment and state and local funding limitations. The $89.8 million short term loan (Tranche A) is to be repaid fully following substantial completion in the form of a milestone payment. The $60.2 million long term loan (Tranche B) is to be repaid us ing the non Federal portion of the quarterly availability payments to GLC over a 28 year period. The TIFIA loans will be secured by a lien on project collateral. Innovations: legislation and since the development of the South Bay Expressway and SR 91 Express Lanes irst availability payment contract for transportation infrastructure First project with direct Federal aid participation in availability payments. First TIFIA loan to be repaid in part with a milestone payment following substantial completion. Incorporat ion of numerous Context Sensitive Design features to minimize traffic impacts and to protect and enhance environmental and cultural resources Contacts: Kome Ajise, P3 Program Manager California Department of Transportation 1120 N Street, MS 49, Sacramento CA 95814 Tel: (916) 654 4227 Bob Kuo, CEO Golden Link Concessionaire, LLC Tel: (925) 899 6421 bob.kuo@GLC presidioparkway.com Leroy L. Saage, PE, Deputy Director for Capital Projects San Francisco County Transportation Authority 1455 Market Street, 22nd Floor, San Francisco, CA 94103 Tel: (415) 522 4800 Fax: (415) 522 4829 www.sfcta.org lee.saage@sfcta.org Project Financial Status / Financial Performance TIFIA credit agreement executed on June 14, 2012 The first TIFIA interest payment is scheduled for June 2016. Principal repayments are scheduled to begin in December 2018. Level debt service payments commence in 2019. The final maturity of the TIFIA loan is December 2045.

PAGE 139

139 APPENDIX B SIMULATED CASH FLOW MODEL CONSTRUCTION I 495 Table B 1. I 495 s ummary Financial base year 2004 Term 85 years Start d ate 2008 End d ate 2098 Construction end date 2013 TIFIA l ength 40 Total c ost 2,093,100,000 Inflation r ate 3.0% TIFIA r ate 4.76% Bond r ating BBB Baa3 Risk free rate 4.66% PAB l ength 30 Formulation of the Simulated Cash Flow Model The following information explains each step in the formulation of the cash flow model for each PPP project. Table B 2. I 495 convert nominal to real dollars Year O&M e xpenses (year of $) O&M e xpenses (2004$) M ajor maintenance c osts (year of $) Major maintenance c osts (2004$) 2013 31,329,034 24,011,096 13,415,187 10,281,624 2014 29,318,952 21,816,054 2015 30,597,423 22,104,229 2016 32,044,048 22,475,051 2017 33,159,900 22,580,278 2018 34,315,733 22,686,742 4,433,039 2,930,761 2019 35,513,029 22,794,462 25 ,740,155 16,521,626 2020 36,873,628 22,978,426 15,838,141 9,869,806 2021 38,159,009 23,086,828 11,488,372 6,950,654 2022 39,490,427 23,196,464 24,086,230 14,148,122 2023 40,869,590 23,307,356 9,080,136 5,178,275 2024 42,298,304 23,419,545 913,861 505, 983 2025 43,778,276 23,532,981 14,977,808 8,051,310 2026 45,311,546 23,647,756 18,378,916 9,591,818 2027 46,900,028 23,763,857 31,000,506 15,707,701 2028 48,545,861 23,881,347 14,072,586 6,922,780

PAGE 140

140 Table B 2. Continued Year O&M e xpenses (year of $ ) O&M e xpenses (2004$) M ajor maintenance c osts (year of $) Major maintenance c osts (2004$) 2029 50,251,099 24,000,205 15,224,680 7,271,392 2030 52,017,971 24,120,459 38,371,782 17,792,793 2031 53,847,072 24,241,362 40,848,974 18,389,761 2032 55,740,56 7 24,362,906 48,161,945 21,050,467 2033 57,700,757 24,485,106 3,943,647 1,673,472 2034 59,730,027 24,607,980 18,881,833 7,779,065 2035 61,830,678 24,731,476 42,137,037 16,854,273 2036 63,888,082 24,810,108 80,179,781 31,136,778 2037 66,012,069 24,888, 283 107,009,159 40,345,262 2038 68,204,756 24,966,003 30,737,175 11,251,186 2039 70,468,331 25,043,275 8,855,674 3,147,160 2040 72,805,051 25,120,103 13,751,172 4,744,600 Table B 3. I 495 traffic f orecast Year Lenders base transactions ( avera ge year ) 12 2013 69,288 2014 100,010 2015 119,879 2016 122,414 2017 124,445 2018 126,793 2019 129,186 2020 135,543 2021 137,983 2022 140,467 2023 142,996 2024 145,571 2025 148,191 2026 150,859 2027 153,575 2028 156,340 2029 159,155 2030 1 62,020 2031 164,886 12 Traffic annualization factor of 300

PAGE 141

141 Table B 3. Continued Year Lenders base transactions ( average year ) 13 2032 167,751 2033 170,616 2034 173,482 2035 176,347 2036 176,347 2037 176,347 2038 176,347 2039 176,347 2040 176,347 Table B 4. I 495 traffic & r evenue forecast Year Lenders base transactions (average year ) Lenders base annual revenue (2004 $) 2013 20,786,400 45,538,000 2014 30,003,000 65,366,000 2015 35,963,700 77,918,000 2016 36,724,200 78,949,000 2017 37,333,500 79,993,000 2018 38,037,900 81,051,000 2019 38,755,800 82,123,000 2020 40,662,900 83,984,000 2021 41,394,900 85,020,000 2022 42,140,100 86,069,000 2023 42,898,800 87,131,000 2024 43,671,300 88,207,000 2025 44,457,300 89,295,000 2026 45,257,700 90,3 97,000 2027 46,072,500 91,512,000 2028 46,902,000 92,642,000 2029 47,746,500 93,785,000 2030 48,606,000 94,942,000 2031 49,465,800 96,113,000 2032 50,325,300 97,299,000 2033 51,184,800 98,500,000 2034 52,044,600 99,716,000 2035 52,904,100 100,946, 000 13 Traffic annualization factor of 300

PAGE 142

142 Table B 4. Continued Year Lenders base transactions (average year ) Lenders base annual revenue (2004$) 2036 52,904,100 102,148,000 2037 52,904,100 103,322,000 2038 52,904,100 104,468,000 2039 52,904,100 105,586,000 2 040 52,904,100 106,678,000 Figure B 1. I 495 revenue vs. traffic

PAGE 143

143 Table B 5. I 495 T&R regression analysis summary output Regression s tatistics Multiple R 0.998695092 R s quare 0.997391886 Adjusted r s quare 0.997131075 Standard e rror 656456.7968 Observations 23 ANOVA df SS MS F Significance F Regression 2 3.29596E+15 1.64798E+15 3824.188244 1.45635E 26 Residual 20 8.61871E+12 4.30936E+11 Total 22 3.30458E+15 VARIABLES Coefficients Standard e rror t Stat P value Intercept 11629046. 33 2580084.577 4.507234544 0.000215115 Avg annual trips 3.197990623 0.133775852 23.90558974 3.49703E 16 Avg annual trips ^2 2.05698E 08 1.70615E 09 12.05627526 1.25162E 10 VARIABLES CONTINUED Lower 95% Upper 95% Lower 95.0% Upper 95.0% Intercept 17011008.43 6247084.228 17011008 6247084 Avg annual trips* 2.918939088 3.477042159 2.9189391 3.477042 Avg annual trips ^2 2.41287E 08 1.70108E 08 2.41E 08 1.7E 08 RESIDUAL OUTPUT Observation Predicted real annual revenue (2004$) Residuals 1 45957994.65 419994.6482 2 65803771.08 437771.0799 3 76777841.87 1140158.13 4 78072834.41 876165.5899 5 79093193.05 899806.9458 6 80253775.39 797224.6143 7 81415597.41 707402.5932 8 84399000.33 415000.327 9 85504379.32 484379.3224 10 86607047.9 9 538047.9854 11 87706222.24 575222.2392 12 88801058.39 594058.3889

PAGE 144

144 Table B 5. Continued Regression s tatistics RESIDUAL OUTPUT Observation Predicted real annual revenue (2004$) Residuals 13 89889830.02 594830.0207 14 90972430.06 575430.0552 15 92047436 535435.9989 16 93113780.31 471780.3075 17 94170328.18 385328.1759 18 95215516.2 273516.1988 19 96230661.69 117661.6873 20 97215056.24 83943.76416 21 98169059.35 330940.6489 22 99092988.1 623011.8952 23 99986197.75 959802.2543 Tab le B 6. I 495 O&M expenses Year Revenues (2004$) O&M e xpense (2004$) 2013 41,894,852 24,011,096 2014 60,136,582 21,816,054 2015 71,684,560 22,104,229 2016 75,001,550 22,475,051 2017 75,993,350 22,580,278 2018 76,998,450 22,686,742 2019 78,016,850 2 2,794,462 2020 79,784,800 22,978,426 2021 80,769,000 23,086,828 2022 81,765,550 23,196,464 2023 82,774,450 23,307,356 2024 83,796,650 23,419,545 2025 84,830,250 23,532,981 2026 85,877,150 23,647,756 2027 86,936,400 23,763,857 2028 88,009,900 23,88 1,347 2029 89,095,750 24,000,205 2030 90,194,900 24,120,459 2031 91,307,350 24,241,362 2032 92,434,050 24,362,906 2033 93,575,000 24,485,106 2034 94,730,200 24,607,980

PAGE 145

145 Table B 6. Continued Year Revenues (2004$) O&M e xpense (2004$) 2035 95,898,700 24,731,476 2036 97,040,486 24,810,108 2037 98,155,697 24,888,283 2038 99,244,505 24,966,003 2039 100,307,115 25,043,275 2040 101,343,763 25,120,103 Figure B 2. I 495 O&M vs. traffic

PAGE 146

146 Table B 7. I 495 O&M regression analysis summary output Reg ression s tatistics Multiple R 0.997194949 R s quare 0.994397766 Adjusted r s quare 0.994102912 Standard e rror 58109.25978 Observations 21 ANOVA df SS MS F Significance F Regression 1 1.13879E+13 1.13879E+13 3372.504216 7.36863E 23 Residual 19 6415 7035367 3376686072 Total 20 1.1452E+13 VARIABLES Coefficients Standard e rror t Stat P value Intercept 17188609.74 109828.2133 156.5045011 5.08106E 31 Traffic (Lender) 0.142666941 0.002456672 58.07326593 7.36863E 23 VARIABLES CONTINUED Lower 95% Upper 95% Lower 95.0% Upper 95.0% Intercept 627338.0741 3374156.619 627338.0741 3374156.619 Traffic (l ender) 0.454510388 0.580477326 0.454510388 0.580477326 RESIDUAL OUTPUT Observation Predicted O&M e xpenses (2004$) Residuals 1 22319440.8 21521 1.4122 2 22427939 47111.5422 3 22514865.97 65412.36712 4 22615360.56 71381.54105 5 22717781.16 76680.79018 6 22989861.28 11435.38567 7 23094293.48 7465.482848 8 23200608.89 4145.016952 9 23308850.3 1494.198613 10 23419060.51 484.8591463 11 23 531196.72 1783.84008 12 23645387.34 2368.717171 13 23761632.37 2224.821793 14 23879974.59 1372.191287 15 24000456.83 252.081786 16 24123079.06 2620.17651 17 24245744.1 4381.597477 18 24368366.33 5460.288212

PAGE 147

147 Table B 7. Continued Regression s tat istics RESIDUAL OUTPUT Observation Predicted O&M Expenses (2004$) Residuals 19 24490988.57 5882.233019 20 24613653.6 5673.334114 21 24736275.84 4799.462602 Table B 8. I 495 a ccumulated major maintenance Year Traffic Accumulated t raffic Major mai ntenance costs (2004$) Accumulated major maintenance cost (2004$) 2013 20,786,400 20,786,400 10,281,624 10,281,624 2014 30,003,000 50,789,400 10,281,624 2015 35,963,700 86,753,100 10,281,624 2016 36,724,200 123,477,300 10,281,624 2017 37,333,500 160,810,800 10,281,624 2018 38,037,900 198,848,700 2,930,761 13,212,385 2019 38,755,800 237,604,500 16,521,626 29,734,011 2020 40,662,900 278,267,400 9,869,806 39,603,817 2021 41,394,900 319,662,300 6,950,654 46,554,471 2022 42,140,100 361,802,400 14,148,122 60,702,592 2023 42,898,800 404,701,200 5,178,275 65,880,867 2024 43,671,300 448,372,500 505,983 66,386,850 2025 44,457,300 492,829,800 8,051,310 74,438,159 2026 45,257,700 538,087,500 9,591,818 84,029,978 2027 46,072,500 584,160,000 15,707, 701 99,737,679 2028 46,902,000 631,062,000 6,922,780 106,660,458 2029 47,746,500 678,808,500 7,271,392 113,931,850 2030 48,606,000 727,414,500 17,792,793 131,724,643 2031 49,465,800 776,880,300 18,389,761 150,114,404 2032 50,325,300 827,205,600 21,050 ,467 171,164,871 2033 51,184,800 878,390,400 1,673,472 172,838,343

PAGE 148

148 Table B 8. Continued Year Traffic Accumulated t raffic Major maintenance costs (2004$) Accumulated major maintenance cost (2004$) 2034 52,044,600 930,435,000 7,779,065 180,617,408 2035 52,904,100 983,339,100 16,854,273 197,471,681 2036 52,904,100 1,036,243,200 31,136,778 228,608,460 2037 52,904,100 1,089,147,300 40,345,262 268,953,721 2038 52,904,100 1,142,051,400 11,251,186 280,204,908 2039 52,904,100 1,194,955,500 3,147,160 283,35 2,067 2040 52,904,100 1,247,859,600 4,744,600 288,096,667 2041 52,904,100 1,300,763,700 11,153,582 299,250,250 2042 52,904,100 1,353,667,800 10,547,138 309,797,388 2043 52,904,100 1,406,571,900 4,895,492 314,692,880 2044 52,904,100 1,459,476,000 648,7 31 315,341,611 2045 52,904,100 1,512,380,100 315,341,611 2046 52,904,100 1,565,284,200 315,341,611 2047 52,904,100 1,618,188,300 6,857,050 322,198,661 Figure B 3 I 495 cost vs. accumulated traffic

PAGE 149

149 Table B 9. I 495 accumulated major maintenan ce & accumulated traffic regression analysis summary output Regression stat istics Multiple R 0.988023472 R s quare 0.976190382 Adjusted r s quare 0.975468879 Standard e rror 18140261.55 Observations 35 ANOVA df SS MS F Significance F Regression 1 4 .45229E+17 4.45229E+17 1352.994522 2.29477E 28 Residual 33 1.08593E+16 3.29069E+14 Total 34 4.56088E+17 VARIABLES Coefficients Standard e rror t Stat P value Intercept 23359461.96 5718988.659 4.084544201 0.000264501 Accum tr affic 0.2330579 17 0.00633601 36.78307385 2.29477E 28 VARIABLES CONTINUED Lower 95% Upper 95% Lower 95.0% Upper 95.0% Intercept 34994831.82 11724092.11 34994831.82 11724092.11 Accum t raffic 0.220167208 0.245948626 0.220167208 0.245948626 RESIDUAL OUTPUT Observa tion Predicted accum cost (2004$) Residuals 1 18515026.87 28796650.65 2 11522590.18 21804213.96 3 3140965.157 13422588.94 4 5417900.41 4863723.374 5 14118768.17 3837144.381 6 22983801.92 9771417.117 7 32016147.95 2282137.134 8 41492958.74 18 89142.07 9 51140367.92 4585897.255 10 60961451.86 258859.5761 11 70959356.84 5078489.877 12 81137299.06 14750449.42 13 91498424.81 17060265.32 14 102046090.1 18016112.19 15 112783651 13045972.5 16 123714533.4 17054075.13

PAGE 150

150 Table B 9. Conti nued Regression s tatistics RESIDUAL OUTPUT Observation Predicted accum cost (2004$) Residuals 17 134842233.3 20910383.02 18 146170246.4 14445603.15 19 157698642.8 7584238.443 20 169427352.4 1737518.499 21 181356375.2 8518032.128 22 193485781.3 12868373.02 23 205815500.7 8343819.248 24 218145220.1 10463239.74 25 230474939.4 38478781.96 26 242804658.8 37400248.74 27 255134378.1 28217688.89 28 267464097.5 20632569.75 29 279793816.9 19456432.88 30 292123536.2 17673851.95 31 304453255.6 102 39624.19 32 316782975 1441364.184 33 329112694.3 13771083.55 34 341442413.7 26100802.91 35 353772133 31573471.91 Table B 10. I 495 r evenue s haring Base case level internal rate of retur n (IRR) revenue sharing per centage First Tier 7.940% to 8.496% 5% Second Tier 8.497% to 8.965% 15% Third Tier 8.966% to 12.980% 30% Amended and Restated Comprehensive Agreement Relating to the Route 495 HOT Lanes in Virginia Project, Dated December 19, 2007 by and among the Virginia Dep artment of Transportation and Capital Beltway Express LLC

PAGE 151

151 LBJ Express Table B 11. LBJ s ummary Financial base year 2008 Term 52 Start d ate 2009 End d ate 2061 Construction e nd d ate 2016 Bond r ating BBB /Baa3 Total c ost 2,794,582,000 Inflation r a te 2.5% TIFIA r ate 4.12% TIFIA l ength 40 Risk f ree r ate 4.11% PAB l ength 30 Formulation of the Simulated Cash Flow Model The following information explains each step in the formulation of the cash flow model for each PPP project. Table B 12. LBJ convert nominal to real dollars Year O&M e xpenses (year of $) O&M e xpenses (2008$) M ajor maintenance costs (year of $) Ma jor maintenance costs (2008$) 2013 21,485,521 17,634,168 2014 22,929,305 18,360,145 2015 23,129,864 18,069,013 2016 23,937 ,582 18,243,903 793,544 651,299 2017 26,648,277 19,814,483 208,170 166,688 2018 25,431,045 18,448,198 1,849,442 1,444,781 2019 27,044,141 19,139,874 316,468 241,194 2020 27,595,912 19,054,027 1,018,200 757,089 2021 27,886,754 18,785,213 3,832,579 2,78 0,231 2022 30,258,587 19,885,794 3,371,583 2,386,161 2023 29,928,365 19,189,047 8,139,161 5,619,810 2024 31,294,881 19,575,815 1,736,723 1,169,900 2025 32,396,408 19,770,586 4,006,849 2,633,281 2026 33,992,108 20,238,435 4,156,170 2,664,795 2027 37,7 54,419 21,930,208 5,688,757 3,558,475 2028 36,262,286 20,549,737 7,884,953 4,811,958 2029 38,643,450 21,365,011 7,656,296 4,558,454 2030 40,293,829 21,734,112 6,642,743 3,858,535 2031 41,744,143 21,967,217 13,213,465 7,488,034 2032 46,026,401 23,629,9 42 5,804,118 3,208,954 2033 44,527,582 22,302,877 8,949,579 4,827,319 2034 45,686,426 22,325,186 2,638,321 1,388,376 2035 47,792,832 22,784,883 5,304,558 2,723,359 2036 51,443,744 23,927,247 5,345,689 2,677,537 2037 57,796,867 26,226,516 9,041,101 4,4 18,036

PAGE 152

152 Table B 13. LBJ traffic forecast Year Lenders base transactions (average year ) 2013 100,000 2014 1,700,000 2015 13,800,000 2016 71,200,000 2017 78,700,000 2018 79,800,000 2019 76,100,000 2020 76,800,000 2021 77,500,000 2022 78 ,200,000 2023 78,900,000 2024 79,600,000 2025 77,600,000 2026 78,400,000 2027 79,200,000 2028 80,000,000 2029 80,800,000 2030 79,600,000 2031 82,400,000 2032 83,200,000 2033 84,000,000 2034 84,900,000 2035 85,700,000 2036 86,500,000 2037 87, 400,000 2038 88,200,000 2039 89,100,000 2040 90,000,000 Table B 14. LBJ traffic & revenue forecast Year Lenders base transactions (average year ) Lenders base annual revenue (2008$) 2013 100,000 100,000 2014 1,700,000 2 ,200,000 2015 13,800,000 12,200,000 2016 71,200,000 70,800,000 2017 78,700,000 81,600,000 2018 79,800,000 85,200,000 2019 76,100,000 84,100,000

PAGE 153

153 Table B 14. Continued Year Lenders base transactions (average year ) Lenders base annual revenue (2008$) 2020 76,800,000 87,800,000 2021 77,500,000 91,700,000 2022 78,200,000 95,800,000 2023 78,900,000 100,000,000 2024 79,600,000 104,500,000 2025 77,600,000 104,700,000 2026 78,400,000 109,500,000 2027 79,200,000 114,600 ,000 2028 80,000,000 119,800,000 2029 80,800,000 125,400,000 2030 79,600,000 130,100,000 2031 82,400,000 137,200,000 2032 83,200,000 143,500,000 2033 84,000,000 150,100,000 2034 84,900,000 156,900,000 2035 85,700,000 164,100,000 2036 86,500,000 17 1,600,000 2037 87,400,000 179,400,000 2038 88,200,000 187,600,000 2039 89,100,000 196,200,000 2040 90,000,000 205,200,000 Figure B 4. LBJ revenue vs. traffic

PAGE 154

154 Table B 15. LBJ T&R regression analysis summary output Regression s tatistics Multiple R 0.98171539 R s quare 0.963765108 Adjusted r s quare 0.962922436 Standard e rror 0.09842671 Observations 45 ANOVA df SS MS F Significance F Regression 1 11.0799708 11.08 1143.70147 1.2929E 32 Residual 43 0.416576141 0.00969 Total 44 11.49654694 VARIABLES Coefficients Standard e rror t Stat P value Intercept 14.60667993 0.13111032 111.408 1.42347E 54 Traffic 4.93579E 08 1.45949E 09 33.8187 1.2929E 32 Intercept t ransformed 2,205,975 VARIABLES CONTINUED Lower 95% Upper 95% Lower 95.0% U pper 95.0% Intercept 14.34227077 14.87108908 14.34227077 14.87108908 Lenders base case 4.64145E 08 5.23012E 08 4.64145E 08 5.23012E 08 RESIDUAL OUTPUT Observation Predicted Y Residuals 1 18.12096007 0.045590512 2 18.49114407 0.273804254 3 18.5454 3773 0.284925736 4 18.36281362 0.115296494 5 18.39736413 0.106792068 6 18.43191463 0.097881696 7 18.46646514 0.088691898 8 18.50101565 0.080334904 9 18.53556615 0.070868525 10 18.43685042 0.029759256 11 18.47633671 0.035098393 12 18.5158230 1 0.041135355 13 18.5553093 0.046024942 14 18.5947956 0.052223591 15 18.53556615 0.148247789

PAGE 155

155 Table B 15. Continued Regression st atistics RESIDUAL OUTPUT Observation Predicted Y Residuals 16 18.67376818 0.063182091 17 18.71325448 0.068591117 18 18.75274077 0.074071527 19 18.79716285 0.073956367 20 18.83664914 0.079337412 21 18.87613544 0.084541307 22 18.92055752 0.084570989 23 18.96004381 0.089778782 24 19.00446589 0.090179213 25 19.04888797 0.090607698 26 19.09331005 0.089110663 27 19.1 3279635 0.090334312 28 19.17721843 0.085460444 29 19.22164051 0.079081835 30 19.26606259 0.070908887 31 19.30554888 0.066789737 32 19.34997096 0.056526576 33 19.39439304 0.045135021 34 19.43881512 0.032337534 35 19.47830142 0.023845456 36 19.52272 35 0.00948576 37 19.56714558 0.006071012 38 19.60663187 0.017191079 39 19.65105395 0.034029431 40 19.69547603 0.051608266 41 19.73989811 0.070175602 42 19.77938441 0.084179203 43 19.82380649 0.103751854 44 19.86822857 0.124193263 45 19.912 65065 0.144936257 Table B 16. LBJ O& M e xpenses Year Revenues (2008$) O&M e xpense (2008$) 2013 100,000 17,634,168 2014 2,200,000 18,360,145 2015 12,200,000 18,069,013 2016 70,800,000 18,243,903 2017 81,600,000 19,814,483 2018 85,200,000 18,448,198

PAGE 156

156 Table B 16. Continued Year Revenues (2008$) O&M e xpense (2008$) 2019 84,100,000 19,139,874 2020 87,800,000 19,054,027 2021 91,700,000 18,785,213 2022 95,800,000 19,885,794 2023 100,000,000 19,189,047 2024 104,500,000 19,575,815 2025 104,700,000 19,770,586 2026 109,500,000 20,238,435 2027 114,600,000 21,930,208 2028 119,800,000 20,549,737 2029 125,400,000 21,365,011 2030 130,100,000 21,734,112 2031 137,200,000 21,967,217 2032 143,500,000 23,629,942 2033 150,100,000 22,302,877 2034 156,900 ,000 22,325,186 2035 164,100,000 22,784,883 2036 171,600,000 23,927,247 2037 179,400,000 26,226,516 2038 187,600,000 17,634,168 2039 196,200,000 18,360,145 2040 205,200,000 18,069,013 Figure B 5. LBJ O&M vs. traffic

PAGE 157

157 Table B 17. LBJ O&M regressi on analysis summary output Regression s tatistics Multiple R 0.901034443 R s quare 0.811863068 Adjusted r s quare 0.803683202 Standard e rror 0.045040471 Observations 25 ANOVA df SS MS F Significance F Regression 1 0.201345729 0.201345729 99.2513825 8.20585E 10 Residual 23 0.046658814 0.002028644 Total 24 0.248004542 VARIABLES Coefficients Standard e rror t Stat P value Intercept 15.22726397 0.161696032 94.17215584 2.90798E 31 Traffic 1.97708E 08 1.98452E 09 9.962498808 8.20585E 10 Lower 95% Upper 95% Lower 95.0% Upper 95.0% Intercept 14.89277024 15.5617577 14.89277024 15.561758 Traffic 1.56655E 08 2.38761E 08 1.56655E 08 2.388E 08 Intercept t ransformed 4103143.604 RESIDUAL OUTPUT Observation Predicted in transformation of O&M e xpenses (2008$) Residuals 1 16.63494552 0.050403407 2 16.78322658 0.057533747 3 16.80497447 0.095265439 4 16.73182248 0.012480965 5 16.74566204 0.05626166 6 16.75950161 0.029024342 7 16.77334117 0.00605681 8 16.78718074 0.024391725 9 1 6.80102031 0.05243974 10 16.76147869 0.044037467 11 16.77729534 0.007445112 12 16.79311198 0.003306526 13 16.80892863 0.00922277 14 16.82474528 0.001651202

PAGE 158

158 Table B 17. Continued Regression s tatistics RESIDUAL OUTPUT Observation Predicted in transformation of O&M e xpenses (2008$) Residuals 15 16.80102031 0.102355304 16 16.85637857 0.018019852 17 16.87219522 0.005069932 18 16.88801186 0.006381712 19 16.90580559 0.000743805 20 16.92162224 0.056402949 21 16.93743888 0.01721266 22 1 6.95523261 0.034006587 23 16.97104926 0.029441406 24 16.98884298 0.001685445 25 17.00663671 0.075644814 Table B 18. LBJ accumulated major maintenance Year Traffic Accumulated traffic Major maintenance costs (2008$) Accum major maintenance cost (200 8$) 2016 71,200,000 71,200,000 651299 651299 2017 78,700,000 149,900,000 166688 817986 2018 79,800,000 229,700,000 1444781 2262767 2019 76,100,000 305,800,000 241194 2503962 2020 76,800,000 382,600,000 757089 3261050 2021 77,500,000 460,100,000 27802 31 6041281 2022 78,200,000 538,300,000 2386161 8427442 2023 78,900,000 617,200,000 5619810 14047253 2024 79,600,000 696,800,000 1169900 15217152 2025 77,600,000 774,400,000 2633281 17850434 2026 78,400,000 852,800,000 2664795 20515228 2027 79,200,000 932,000,000 3558475 24073703 2028 80,000,000 1,012,000,000 4811958 28885661 2029 80,800,000 1,092,800,000 4558454 33444115 2030 79,600,000 1,172,400,000 3858535 37302650 2031 82,400,000 1,254,800,000 7488034 44790684 2032 83,200,000 1,338,000,000 320 8954 47999637 2033 84,000,000 1,422,000,000 4827319 52826956 2034 84,900,000 1,506,900,000 1388376 54215332 2035 85,700,000 1,592,600,000 2723359 56938691 2036 86,500,000 1,679,100,000 2677537 59616228 2037 87,400,000 1,766,500,000 4418036 64034263 2 038 88,200,000 1,854,700,000 12072734 76106998 2039 89,100,000 1,943,800,000 11784075 87891073 2040 90,000,000 2,033,800,000 5299909 93190983

PAGE 159

159 Figure B 6. LBJ cost vs. accumulated traffic

PAGE 160

160 Table B 19. LBJ major maintenance regression analysis summar y output Regression s tatistics Multiple R 0.981380006 R s quare 0.963106716 Adjusted r s quare 0.96150266 Standard e rror 5558662.556 Observations 25 ANOVA df SS MS F Significance F Regression 1 1.85522E+16 1.85522E+16 600.4197039 5.54308E 18 Resid ual 23 7.10671E+14 3.08987E+13 Total 24 1.92629E+16 VARIABLES Coefficients Standard e rror t Stat P value Intercept (13,631,634) 2243457.669 6.07617173 3.38107E 06 Accum t raffic 0.046483426 0.001897015 24.5034631 5.54308E 18 VARIABLES CONT INUED Lower 95% Upper 95% Lower 95.0% Upper 95.0% Intercept 18272579.82 8990688.314 18272579.82 8990688.31 Accum t raffic 0.042559153 0.0504077 0.042559153 0.0504077 RESIDUAL OUTPUT Observation Predicted accum cost Residuals 1 10322014.13 1097331 2.64 2 6663768.488 7481754.628 3 2954391.08 5217158.355 4 582997.6504 1920964.059 5 4152924.779 891874.4681 6 7755390.307 1714109.098 7 11390394.23 2962952.041 8 15057936.56 1010684.033 9 18758017.28 3540864.84 10 22365131.15 4514697.353 11 26009431.76 5494203.447 12 29690919.11 5617215.623 13 33409593.2 4523932.014 14 37165454.04 3721339.21 15 40865534.76 3562885.219 16 44695769.08 94914.58621

PAGE 161

161 Table B 19. Continued Regression Statistics RESIDUAL OUTPUT Observation Predicted accum cost Residuals 17 48563190.13 563552.6753 18 52467797.93 359158.2193 19 56414240.81 2198908.545 20 60397870.43 3459179.532 21 64418686.8 4802459.036 22 68481338.24 4447074.746 23 72581176.43 3525821.339 24 76722849.7 11168223.53 25 8090 6358.06 12284624.52 Table B 20. LBJ revenue sharing Band 1 Band 2 Band 3 Year Floor Ceiling Floor Ceiling Floor Ceiling 1 0 3,068,880 3,068,880 3,773,849 3,773,849 4,596,304 2 0 20,342,124 20,342,124 25,015,027 25,015,027 30,466,680 3 0 142,831,743 142,831,743 175,642,420 175,642,420 213,921,076 4 0 303,443,025 303,443,025 373,148,616 373,148,616 454,470,812 5 0 476,134,309 476,134,309 585,509,778 585,509,778 713,112,937 6 0 661,828,137 661,828,137 813,860,372 813,860,372 991,229,150 7 0 861,517, 930 861,517,930 1,059,422,022 1,059,422,022 1,290,307,314 8 0 1,076,273,529 1,076,273,529 1,323,510,329 1,323,510,329 1,611,949,744 9 0 1,307,247,178 1,307,247,178 1,607,542,224 1,607,542,224 1,957,882,171 10 0 1,555,679,985 1,555,679,985 1,913,043,917 1,913,043,917 2,329,963,421 11 0 1,822,908,903 1,822,908,903 2,241,659,481 2,241,659,481 2,730,195,865 12 0 2,090,550,285 2,090,550,285 2,570,782,259 2,570,782,259 3,131,046,063 13 0 2,378,936,921 2,378,936,921 2,925,415,800 2,925,415,800 3,562,966,716 14 0 2,689,703,397 2,689,703,397 3,307,570,177 3,307,570,177 4,028,405,963 15 0 3,024,614,684 3,024,614,684 3,719,415,805 3,719,415,805 4,530,007,227 16 0 3,385,576,676 3,385,576,676 4,163,296,391 4,163,296,391 5,070,624,993 17 0 3,772,900,527 3,772,90 0,527 4,639,594,565 4,639,594,565 5,650,725,279 18 0 4,193,594,027 4,193,594,027 5,156,927,918 5,156,927,918 6,280,803,751 19 0 4,646,417,782 4,646,417,782 5,713,772,344 5,713,772,344 6,959,004,148 20 0 5,133,825,723 5,133,825,723 6,313,145,484 6,313,14 5,484 7,689,001,760 21 0 5,658,459,200 5,658,459,200 6,958,295,445 6,958,295,445 8,474,752,571 22 0 6,223,161,294 6,223,161,294 7,652,718,409 7,652,718,409 9,320,514,705 23 0 6,830,992,227 6,830,992,227 8,400,177,578 8,400,177,578 10,230,871,495 24 0 7 ,485,245,940 7,485,245,940 9,204,723,564 9,204,723,564 11,210,756,327 25 0 8,189,467,952 8,189,467,952 10,070,716,344 10,070,716,344 12,265,479,370

PAGE 162

162 Table B 20. Continued Band 1 Band 2 Band 3 Y ear Floor Ceiling Floor Ceiling Floor Ceiling 26 0 8,947,4 74,565 8,947,474,565 11,002,848,887 11,002,848,887 13,400,756,351 27 0 9,763,373,551 9,763,373,551 12,006,172,584 12,006,172,584 14,622,739,545 28 0 10,641,586,412 10,641,586,412 13,086,124,622 13,086,124,622 15,938,051,087 29 0 11,586,872,337 11,586,87 2,337 14,248,557,453 14,248,557,453 17,353,818,886 30 0 12,604,353,999 12,604,353,999 15,449,770,507 15,449,770,507 18,877,715,237 31 0 13,699,545,310 13,699,545,310 16,846,544,327 16,846,544,327 20,517,998,405 32 0 14,878,381,306 14,878,381,306 18,296, 177,319 18,296,177,319 22,283,557,373 33 0 16,147,250,307 16,147,250,307 19,856,525,300 19,856,525,300 24,183,960,018 34 0 17,513,028,539 17,513,028,539 21,536,044,073 21,536,044,073 26,229,504,959 35 0 18,983,117,396 18,983,117,396 23,343,835,247 23,34 3,835,247 28,431,277,363 36 0 20,565,483,549 20,565,483,549 25,289,695,560 25,289,695,560 30,801,209,026 37 0 22,268,702,119 22,268,702,119 27,384,169,975 27,384,169,975 33,352,143,026 38 0 24,107,003,151 24,107,003,151 29,638,608,819 29,638,608,819 36, 097,903,325 39 0 26,075,321,624 26,075,321,624 32,065,229,293 32,065,229,293 39,053,369,684 40 0 28,199,351,303 28,199,351,303 34,677,181,684 34,677,181,684 42,234,558,298 41 0 30,485,602,678 30,485,602,678 37,488,620,623 37,488,620,623 45,658,708,590 42 0 32,946,465,351 32,946,465,351 40,514,781,796 40,514,781,796 49,344,376,636 43 0 35,595,275,170 35,595,275,170 43,772,064,503 43,772,064,503 53,311,535,722 44 0 38,446,386,506 38,446,386,506 47,278,120,537 47,278,120,537 57,581,684,585 45 0 41,515,2 50,033 41,515,250,033 51,051,949,834 51,051,949,834 62,177,936,917 46 0 44,818,496,467 44,818,496,467 55,114,003,443 55,114,003,443 67,125,281,768 47 0 48,374,023,683 48,374,023,683 59,486,294,350 59,486,294,350 72,450,448,527 48 0 49,573,143,519 49,573 ,143,519 60,964,556,154 60,964,556,154 74,250,875,532 49 0 49,576,143,519 49,576,143,519 60,964,556,154 60,964,556,154 74,250,875,532 Table B 20. Continued Band 4 Band 5 Year Floor Ceiling Floor 1 4,596,304 5,205,965 5,205,965 2 30,466,680 34,507,83 2 34,507,832 3 213,921,076 242,295,932 242,295,932 4 454,470,812 514,752,598 514,752,598 5 713,112,937 807,701,454 807,701,454 6 991,229,150 1,122,707,476 1,122,707,476 7 1,290,307,314 1,461,455,878 1,461,455,878 8 1,611,949,744 1,825,761,509 1,825,7 61,509 9 1,957,882,171 2,217,579,003 2,217,579,003 10 2,329,963,421 2,639,013,745 2,639,013,745 11 2,730,195,865 3,092,333,703 3,092,333,703

PAGE 163

163 Table B 20. Continued Band 4 Band 5 Year Floor Ceiling Floor 12 3,131,046,063 3,546,353,356 3,546,353,356 13 3,562,966,716 4,035,564,701 4,035,564,701 14 4,028,405,963 4,562,740,604 4,562,740,604 15 4,530,007,227 5,130,875,115 5,130,875,115 16 5,070,624,993 5,743,201,345 5,743,201,345 17 5,650,725,279 6,400,247,122 6,400,247,122 18 6,280,803,751 7,113,900 ,277 7,113,900,277 19 6,959,004,148 7,882,058,333 7,882,058,333 20 7,689,001,760 8,708,884,073 8,708,884,073 21 8,474,752,571 9,598,858,213 9,598,858,213 22 9,320,514,705 10,556,803,679 10,556,803,679 23 10,230,871,495 11,587,911,748 11,587,911,748 2 4 11,210,756,327 12,697,770,176 12,697,770,176 25 12,265,479,370 13,892,393,483 13,892,393,483 26 13,400,756,351 15,178,255,543 15,178,255,543 27 14,622,739,545 16,562,324,670 16,562,324,670 28 15,938,051,087 18,052,101,380 18,052,101,380 29 17,353,81 8,886 19,655,659,035 19,655,659,035 30 18,877,715,237 21,381,687,599 21,381,687,599 31 20,517,998,405 23,239,540,725 23,239,540,725 32 22,283,557,373 25,239,286,447 25,239,286,447 33 24,183,960,018 27,391,761,742 27,391,761,742 34 26,229,504,959 29,70 8,631,254 29,708,631,254 35 28,431,277,363 32,202,450,507 32,202,450,507 36 30,801,209,026 34,886,733,949 34,886,733,949 37 33,352,143,026 37,776,028,188 37,776,028,188 38 36,097,903,325 40,885,990,819 40,885,990,819 39 39,053,369,684 44,233,475,279 4 4,233,475,279 40 42,234,558,298 47,836,622,179 47,836,622,179 41 45,658,708,590 51,714,957,608 51,714,957,608 42 49,344,376,636 55,889,498,953 55,889,498,953 43 53,311,535,722 60,382,868,789 60,382,868,789 44 57,581,684,585 65,219,417,484 65,219,417,4 84 45 62,177,936,917 70,425,355,150 70,425,355,150 46 67,125,281,768 76,028,893,682 76,028,893,682 47 72,450,448,527 82,060,399,648 82,060,399,648 48 74,250,875,532 84,099,638,362 84,099,638,362 49 74,250,875,532 84,099,638,362 84,099,638,362

PAGE 164

164 North Tarrant Expressway Table B 21. NTE s ummary Financial base year 2008 Term 52 Start d ate 2009 End date 2061 Construction end d ate 2015 Inflation r ate 2.5% Total c ost 2,105,671,000 Bond r ating BBB /Baa2 TIFIA r ate 4.45% PAB l ength 30 Risk free r ate 4.44% TIFIA l ength 40 Formulation of the Simulated Cash Flow Model The following information explains each step in the formulation of the cash flow model for each PPP project. Table B 22. NTE convert nominal to real dollars Year O&M e xpenses (y ear of $) O&M e xpenses (2008$) Major maintenance costs (year of $) Major maintenance cost s (2008$) 2015 7,463,443 6,278,735 182,625 153,636 2016 17,360,575 14,248,632 356,717 292,774 2017 19,495,894 15,610,915 584,175 467,765 2018 20,854,252 16,291,308 2,966,781 2,317,645 2019 22,305,707 17,000,178 2,827,915 2,155,281 2020 24,011,378 17,853,801 1,360,308 1,011,465 2021 23,540,161 17,076,512 1,226,520 889,743 2022 23,645,306 16,734,426 2,027,880 1,435,186 2023 26,046,114 17,983,945 1,404,791 969,960 2024 25,182,203 16,963,360 16,356,968 11,018,461 2025 26,742,834 17,575,258 10,207,916 6,708,592 2026 27,088,127 17,367,984 18,408,521 11,802,916 2027 28,750,186 17,984,038 18,001,670 11,260,544 2028 30,770,831 18,778,544 20,163,854 12,305,414 2029 31,189,767 18,569,960 54,359,508 32,364,906 2030 31,788,945 18,465,075 56,567,300 32,857,946 2031 32,545,006 18,443,165 11,579,019 6,561,798 2032 33,068,916 18,282,989 9,029,841 4,992,377 2033 35,114,793 18,940,589 8,162,902 4,402,993 2034 36,444,424 19,178,321 19,937,994 10,492,065 2035 37,871,025 19,442,974 6,057,623 3,109,982 2036 40,896,902 20,484,350 20,966,902 10,501,855 2037 40,848,450 19,961,055 14,045,023 6,863,259 2038 41,564,642 19,815,639 8,745,073 4,169,150 2039 44,420,594 20,660,676 12,144,305 5,648,496

PAGE 165

165 Table B 23. NTE traffic forecast Year Lenders base transactions (average year) 14 2014 2015 12,954,280 2016 22,237,619 2017 27,071,210 2018 30,394,367 2019 28,747,111 2020 29,314,619 2021 29,667,458 2022 30,021,308 20 23 30,375,832 2024 30,731,367 2025 31,346,729 2026 31,453,221 2027 31,605,208 2028 31,760,902 2029 31,920,303 2030 32,017,696 2031 34,122,598 2032 34,538,456 2033 34,950,607 2034 35,377,586 2035 35,804,565 2036 36,231,881 2037 36,658,860 203 8 37,086,176 2039 37,497,990 2040 37,925,306 2041 38,352,285 2042 38,772,524 2043 39,178,272 2044 39,596,826 2045 40,012,347 2046 40,438,652 2047 40,798,231 2048 41,225,547 2049 41,650,504 2050 42,077,483 2051 42,501,766 14 *Traffic annualization factor = 337

PAGE 166

166 Table B 23. Continued Year Lenders base transactions (average year) 2052 42,928,745 2053 43,300,456 2054 43,727,435 2055 44,095,776 2056 44,571,620 2057 44,891,433 2058 45,405,358 2059 45,669,903 2060 46,211,462 2061 46,530,601 2062 47,049,244 2063 47,234,59 4 2064 47,860,740 2065 47,960,829 Table B 24. NTE traffic & revenue forecast Year Lenders base transactions (average year) Lenders base annual revenue (2008$) 2014 2015 12,954,280 38,548,000 2016 22,237,619 59,289,00 0 2017 27,071,210 72,975,000 2018 30,394,367 84,006,000 2019 28,747,111 78,274,000 2020 29,314,619 81,197,000 2021 29,667,458 83,641,000 2022 30,021,308 86,106,000 2023 30,375,832 88,592,000 2024 30,731,367 91,099,000 2025 31,346,729 95,777,000 2 026 31,453,221 100,391,000 2027 31,605,208 105,062,000 2028 31,760,902 109,790,000 2029 31,920,303 114,576,000 2030 32,017,696 119,176,000 2031 34,122,598 114,995,000 2032 34,538,456 120,529,000 2033 34,950,607 125,965,000 2034 35,377,586 131,467,0 00 2035 35,804,565 137,054,000

PAGE 167

167 Table B 24. Continued Year Lenders base transactions (average year ) Lenders base annual revenue (2008$) 2036 36,231,881 142,724,000 2037 36,658,860 148,478,000 2038 37,086,176 154,316,000 2039 37,497,990 160,121,000 2040 37,925,306 165,992,000 2041 38,352,285 171,945,000 2042 38,772,524 177,924,000 2043 39,178,272 183,705,000 2044 39,596,826 189,365,000 2045 40,012,347 195,018,000 2046 40,438,652 200,009,000 2047 40,798,231 204,33 3,000 2048 41,225,547 208,608,000 2049 41,650,504 212,888,000 2050 42,077,483 217,212,000 2051 42,501,766 220,567,000 2052 42,928,745 223,944,000 2053 43,300,456 226,506,000 2054 43,727,435 228,986,000 2055 44,095,776 231,495,000 2056 44,571,620 2 33,872,000 2057 44,891,433 236,419,000 2058 45,405,358 238,660,000 2059 45,669,903 241,161,000 2060 46,211,462 243,237,000 2061 46,530,601 245,670,000 2062 47,049,244 247,848,000 2063 47,234,594 250,383,000 2064 47,860,740 252,335,000 2065 47,960, 829 254,895,000

PAGE 168

168 Figure B 7. NTE revenue vs. traffic Table B 25. NTE T&R regression analysis summary output Regression s tatistics Multiple R 0.984858379 R s quare 0.969946026 Adjusted r s quare 0.969332679 Standard e rror 0.080005077 Observations 51 ANOVA df SS MS F Significance F Regression 1 10.12224471 10.12224471 1581.400016 5.88282E 39 Residual 49 0.313639804 0.006400812 Total 50 10.43588452 VARIABLES Coefficients Standard e rror t Stat P value Intercept 16.51 0.059 279.07 4 .15E 80 Traffic 6.19E 08 1.56E 09 39.77 5.88E 39 Lower 95% Upper 95% Lower 95.0% Upper 95.0% Intercept 16.39 16.63 16.39 16.63 Traffic 5.88E 08 6.50E 08 5.88E 08 6.50E 08 Intercept t ransformed 14,827,478

PAGE 169

169 Table B 25. Continued Regression s tati stics RESIDUAL OUTPUT Observation Predicted Y Residuals 1 17.31373469 0.153680084 2 17.88828175 0.009652596 3 18.18743333 0.081805856 4 18.39310395 0.146705168 5 18.29115506 0.115429012 6 18.32627821 0.113889347 7 18.34811546 0.106071069 8 18 .37001528 0.098925627 9 18.39195682 0.092404699 10 18.41396092 0.086503539 11 18.45204576 0.074512628 12 18.45863656 0.034053443 13 18.46804306 0.002018152 14 18.47767898 0.036401028 15 18.48754433 0.069204588 16 18.49357199 0.102539956 17 18 .62384465 0.063445446 18 18.64958216 0.042181213 19 18.67509024 0.02357559 20 18.70151602 0.007249597 21 18.72794181 0.007943756 22 18.75438845 0.0220348 23 18.78081424 0.035133118 24 18.80726088 0.047252124 25 18.8327481 0.058692233 26 18.859 19475 0.068255405 27 18.88562053 0.077064683 28 18.91162918 0.085237871 29 18.93674098 0.092100792 30 18.96264534 0.096541589 31 18.98836199 0.100240433 32 19.01474606 0.099126864 33 19.03700045 0.098261124 34 19.06344709 0.092520357 35 19.0897477 4 0.086529026 36 19.11617352 0.08021087 37 19.14243245 0.069279608 38 19.16885824 0.058048339 39 19.19186348 0.046418511 40 19.21828927 0.030882156

PAGE 170

170 Table B 25. Continued Regression s tatistics RESIDUAL OUTPUT Observation Predicted Y Residuals 41 19 .24108594 0.018982893 42 19.27053599 0.000251473 43 19.29032926 0.009213043 44 19.32213614 0.031585638 45 19.33850887 0.037533548 46 19.37202602 0.062479188 47 19.39177758 0.072277847 48 19.42387646 0.095550247 49 19.4353478 0.096845492 50 19.47410005 0.127831926 51 19.48029457 0.123932321 Table B 26. NTE operations & maintenance expenses Year Revenues (2008$) O&M e xpense (2008$) 2015 38,548,000 6,278,735 2016 59,289,000 14,248,632 2017 72,975,000 15,610,915 2018 84,006,000 16,291,308 2019 78,274,000 17,000,178 2020 81,197,000 17,853,801 2021 83,641,000 17,076,512 2022 86,106,000 16,734,426 2023 88,592,000 17,983,945 2024 91,099,000 16,963,360 2025 95,777,000 17,575,258 2026 100,391,000 17,367,984 2027 105,062,000 17,984,038 2028 109,790,000 18,778,544 2029 114,576,000 18,569,960 2030 119,176,000 18,465,075 2031 114,995,000 18,443,165 2032 120,529,000 18,282,989 2033 125,965,000 18,940,589 2034 131,467,000 19,178,321 2035 137 ,054,000 19,442,974 2036 142,724,000 20,484,350 2037 148,478,000 19,961,055 2038 154,316,000 19,815,639 2039 160,121,000 20,660,676

PAGE 171

171 Figure B 8. NTE O&M vs. traffic Table B 27. NTE O&M regression analysis summary output Regression s ta tistics Multiple R 0.962422132 R s quare 0.926256359 Adjusted r s quare 0.923050114 Standard e rror 775114.6992 Observations 25 ANOVA df SS MS F Significance F Regression 1 1.73567E+14 1.73567E+14 288.8913008 1.62349E 14 Residual 23 1.38185E+13 6.0 0803E+11 Total 24 1.87385E+14 VARIABLES Coefficients Standard e rror t Stat P value Intercept 1373409.27 967171.82 1.42 0.167 Lender's traffic 0.52 0.03 17.00 1.62E 14 Intercept transformed VARIABLES CONTINUED Lower 95% Upper 95% Lowe r 95.0% Upper 95.0% Intercept 627338.07 3374156.62 627338.07 3374156.62 Lender's tr affic 0.45 0.58 0.45 0.58

PAGE 172

172 Table B 27. Continued Regression s tatistics RESIDUAL OUTPUT Observation Predicted O&M e xpense (2008) Residuals 1 8077169.594 1798434.464 2 12881240.5 1367391.9 3 15382594.15 228321.1162 4 17102307.48 810999.1492 5 16249862.62 750315.5827 6 16543544.52 1310256.896 7 16726136.54 350375.8968 8 16909251.74 174825.6318 9 17092715.73 891228.8743 10 17276702.91 313343.0893 11 1759514 8.97 19890.64844 12 17650257.92 282274.3409 13 17728910.26 255127.9355 14 17809480.95 969063.0983 15 17891969.99 677989.5396 16 17942370.27 522704.7489 17 19031644.12 588479.1084 18 19246848.08 963859.4349 19 19460133.69 519544.7969 20 196810 92.7 502771.3904 21 19902051.71 459077.7687 22 20123185.12 361164.515 23 20344144.13 383089.3932 24 20565277.53 749638.4945 25 20778388.75 117712.3933 Table B 28. NTE accumulated major maintenance Year Traffic Accumulated traffic Major mainten ance costs (2008$) Accumulated major maintenance cost (2008$) 2015 12,954,280 12,954,280 153,636 153,636 2016 22,237,619 35,191,899 292,774 446,410 2017 27,071,210 62,263,109 467,765 914,176 2018 30,394,367 92,657,476 2,317,645 3,231,820 2019 28,747,111 121,404,587 2,155,281 5,387,101 2020 29,314,619 150,719,206 1,011,465 6,398,566 2021 29,667,458 180,386,664 889,743 7,288,309

PAGE 173

173 Table B 28. Continued Year Traffic Accumulated traffic Major maintenance costs (2008$) Accumul ated major maintenance cost (2008$) 2022 30,021,308 210,407,972 1,435,186 8,723,494 2023 30,375,832 240,783,804 969,960 9,693,454 2024 30,731,367 271,515,171 11,018,461 20,711,916 2025 31,346,729 302,861,900 6,708,592 27,420,508 2026 3 1,453,221 334,315,121 11,802,916 39,223,424 2027 31,605,208 365,920,329 11,260,544 50,483,967 2028 31,760,902 397,681,231 12,305,414 62,789,382 2029 31,920,303 429,601,534 32,364,906 95,154,287 2030 32,017,696 461,619,230 32,857,946 128,012,233 2031 34,122,598 495,741,828 6,561,798 134,574,031 2032 34,538,456 530,280,284 4,992,377 139,566,408 2033 34,950,607 565,230,891 4,402,993 143,969,400 2034 35,377,586 600,608,477 10,492,065 154,461,465 2035 35,804,565 636, 413,042 3,109,982 157,571,447 2036 36,231,881 672,644,923 10,501,855 168,073,302 2037 36,658,860 709,303,783 6,863,259 174,936,561 2038 37,086,176 746,389,959 4,169,150 179,105,710 2039 37,497,990 783,887,949 5,648,496 184,754,207 Figure B 9. NTE cost vs. accumulated traffic

PAGE 174

174 Table B 29. NTE major maintenance regression analysis summary output Regression s tatistics Multiple R 0.966229713 R s quare 0.933599858 Adjusted r s quare 0.930712895 Standard e rror 18779624.01 Observati ons 25 ANOVA df SS MS F Significance F Regression 1 1.1405E+17 1.1405E+17 323.3848002 4.84164E 15 Residual 23 8.11151E+15 3.52674E+14 Total 24 1.22161E+17 VARIABLE Coefficients Standard e rror t Stat P value Intercept 33296703.88 7150465.2 3 4.6565786 0.0001095 Accumulated traffic 0.290673192 0.016163864 17.982903 4.84164E 15 Intercept t ransformed VARIABLES CONTINUED Lower 95% Upper 95% Lower 95.0% Upper 95.0% Intercept 48088568.1 18504839.66 48088568.1 18504839.66 Accum tr affic 0.25723563 0.324110691 0.257235693 0.324110691 RESIDUAL OUTPUT Observation Predicted accum cost Residuals 1 29531241.96 29684878.03 2 23067362.27 23513772.59 3 15198487.25 16112663.06 4 6363659.581 9595479.965 5 1992354.931 3394746.115 6 10513328.8 4114762.739 7 19136863.52 11848554.85 8 27863252.94 19139758.45 9 36692692.98 26999238.69 10 45625477.51 24913561.74 11 54737131.29 27316623.58 12 63879739.43 24656315.61 13 73066526.12 22582558.78 14 82298568.88 19509187.34 15 91576945.24 3577341.859 16 100883631.1 27128601.96

PAGE 175

175 Table B 29. Continued Regression s tatistics RESIDUAL OUTPUT 17 110802155.6 23771875.57 18 120841558.9 18724848.86 19 131000763.3 12968636.89 20 141284079.2 13177385.76 21 151691506.4 5879940.574 22 162223142.9 5850159.479 23 172878890.7 2057670.161 24 183658847.9 4553137.407 25 194558508.3 9804301.678

PAGE 176

176 Table B 30. NTE revenue sharing schedule Band 1 Band 2 Band 3 Year Floor Ceiling Floor Ceiling Floor Ceiling 1 0 $29,392,000 $29,392,00 0 $36,301,000 $36,301,000 $44,720,000 2 0 $112,266,000 $112,266,000 $138,657,000 $138,657,000 $170,813,000 3 0 $222,696,000 $222,696,000 $275,046,000 $275,046,000 $338,831,000 4 0 $362,506,000 $362,506,000 $447,722,000 $447,722,000 $551,552,000 5 0 $51 5,847,000 $515,847,000 $637,108,000 $637,108,000 $784,859,000 6 0 $678,722,000 $678,722,000 $838,271,000 $838,271,000 $1,032,673,000 7 0 $852,772,000 $852,772,000 $1,053,235,000 $1,053,235,000 $1,297,489,000 8 0 $1,038,570,000 $1,038,570,000 $1,282,709, 000 $1,282,709,000 $1,580,180,000 9 0 $1,236,714,000 $1,236,714,000 $1,527,432,000 $1,527,432,000 $1,881,656,000 10 0 $1,447,830,000 $1,447,830,000 $1,788,175,000 $1,788,175,000 $2,202,868,000 11 0 $1,664,355,000 $1,664,355,000 $2,055,600,000 $2,055,600 ,000 $2,532,310,000 12 0 $1,900,953,000 $1,900,953,000 $2,347,816,000 $2,347,816,000 $2,892,294,000 13 0 $2,158,734,000 $2,158,734,000 $2,666,194,000 $2,666,194,000 $3,284,507,000 14 0 $2,438,858,000 $2,438,858,000 $3,012,168,000 $3,012,168,000 $3,710,7 14,000 15 0 $2,742,537,000 $2,742,537,000 $3,387,233,000 $3,387,233,000 $4,172,760,000 16 0 $3,070,366,000 $3,070,366,000 $3,792,126,000 $3,792,126,000 $4,671,551,000 17 0 $3,398,896,000 $3,398,896,000 $4,197,884,000 $4,197,884,000 $5,171,408,000 18 0 $3,751,848,000 $3,751,848,000 $4,633,805,000 $4,633,805,000 $5,708,423,000 19 0 $4,129,937,000 $4,129,937,000 $5,100,773,000 $5,100,773,000 $6,283,685,000 20 0 $4,534,408,000 $4,534,408,000 $5,600,324,000 $5,600,324,000 $6,899,085,000 21 0 $4,966,606,00 0 $4,966,606,000 $6,134,121,000 $6,134,121,000 $7,556,674,000 22 0 $5,427,939,000 $5,427,939,000 $6,703,900,000 $6,703,900,000 $8,258,590,000 23 0 $5,919,869,000 $5,919,869,000 $7,311,469,000 $7,311,469,000 $9,007,059,000 24 0 $6,443,922,000 $6,443,922, 000 $7,958,713,000 $7,958,713,000 $9,804,404,000 25 0 $7,001,284,000 $7,001,284,000 $8,647,096,000 $8,647,096,000 $10,652,429,000 26 0 $7,593,528,000 $7,593,528,000 $9,378,560,000 $9,378,560,000 $11,553,525,000 27 0 $8,222,349,000 $8,222,349,000 $10,155 ,200,000 $10,155,200,000 $12,510,274,000 28 0 $8,889,302,000 $8,889,302,000 $10,978,936,000 $10,978,936,000 $13,525,041,000

PAGE 177

177 Table B 30. Continued Band 1 Band 2 Band 3 Year Floor Ceiling Floor Ceiling Floor Ceiling 29 0 $9,595,141,000 $9,595,141,000 $ 11,850,699,000 $11,850,699,000 $14,598,974,000 30 0 $10,340,918,000 $10,340,918,000 $12,771,787,000 $12,771,787,000 $15,733,669,000 31 0 $11,128,156,000 $11,128,156,000 $13,744,083,000 $13,744,083,000 $16,931,449,000 32 0 $11,955,728,000 $11,955,728,000 $14,766,195,000 $14,766,195,000 $18,190,598,000 33 0 $12,822,927,000 $12,822,927,000 $15,836,509,000 $15,836,509,000 $19,509,126,000 34 0 $13,729,176,000 $13,729,176,000 $16,956,533,000 $16,956,533,000 $20,888,893,000 35 0 $14,677,767,000 $14,677,767,0 00 $18,128,112,000 $18,128,112,000 $22,332,171,000 36 0 $15,669,821,000 $15,669,821,000 $19,353,371,000 $19,353,371,000 $23,841,577,000 37 0 $16,702,382,000 $16,702,382,000 $20,628,659,000 $20,628,659,000 $25,412,615,000 38 0 $17,776,962,000 $17,776,962 ,000 $21,955,844,000 $21,955,844,000 $27,047,585,000 39 0 $18,891,006,000 $18,891,006,000 $23,331,769,000 $23,331,769,000 $28,742,599,000 40 0 $20,045,404,000 $20,045,404,000 $24,757,535,000 $24,757,535,000 $30,499,011,000 41 0 $21,241,624,000 $21,241,6 24,000 $26,234,954,000 $26,234,954,000 $32,319,056,000 42 0 $22,480,346,000 $22,480,346,000 $27,764,865,000 $27,764,865,000 $34,203,766,000 43 0 $23,763,859,000 $23,763,859,000 $29,350,098,000 $29,350,098,000 $36,156,627,000 44 0 $25,091,933,000 $25,091 ,933,000 $30,990,366,000 $30,990,366,000 $38,177,286,000 45 0 $26,467,470,000 $26,467,470,000 $32,689,255,000 $32,689,255,000 $40,270,162,000 46 0 $27,889,534,000 $27,889,534,000 $34,445,607,000 $34,445,607,000 $42,433,827,000 47 0 $28,257,583,000 $28,2 57,583,000 $34,900,175,000 $34,900,175,000 $42,993,812,000

PAGE 178

178 Table B 30. Continued Band 4 Band 5 Year Floor Ceiling Floor 1 $44,720,000 $51,194,000 $51,194,000 2 $170,813,000 $195,543,000 $195,543,000 3 $338,831,000 $387,887,000 $387,887,000 4 $551 ,552,000 $631,405,000 $631,405,000 5 $784,859,000 $898,490,000 $898,490,000 6 $1,032,673,000 $1,182,182,000 $1,182,182,000 7 $1,297,489,000 $1,485,338,000 $1,485,338,000 8 $1,580,180,000 $1,808,956,000 $1,808,956,000 9 $1,881,656,000 $2,154,079,000 $2 ,154,079,000 10 $2,202,868,000 $2,521,796,000 $2,521,796,000 11 $2,532,310,000 $2,898,935,000 $2,898,935,000 12 $2,892,294,000 $3,311,036,000 $3,311,036,000 13 $3,284,507,000 $3,760,033,000 $3,760,033,000 14 $3,710,714,000 $4,247,946,000 $4,247,946,00 0 15 $4,172,760,000 $4,776,886,000 $4,776,886,000 16 $4,671,551,000 $8,347,892,000 $8,347,892,000 17 $5,171,408,000 $5,920,118,000 $5,920,118,000 18 $5,708,423,000 $6,534,880,000 $6,534,880,000 19 $6,283,685,000 $7,193,428,000 $7,193,428,000 20 $6,89 9,085,000 $7,897,925,000 $7,897,925,000 21 $7,556,674,000 $8,650,719,000 $8,650,719,000 22 $8,258,590,000 $9,454,257,000 $9,454,257,000 23 $9,007,059,000 $10,311,088,000 $10,311,088,000 24 $9,804,404,000 $11,223,872,000 $11,223,872,000 25 $10,652,429, 000 $12,194,672,000 $12,194,672,000 26 $11,553,525,000 $13,226,228,000 $13,226,228,000 27 $12,510,274,000 $14,321,494,000 $14,321,494,000 28 $13,525,041,000 $15,483,178,000 $15,483,178,000

PAGE 179

179 Table B 30. Continued Band 4 Band 5 Year Floor Ceiling Floor 29 $14,598,974,000 $16,712,592,000 $16,712,592,000 30 $15,733,669,000 $18,011,568,000 $18,011,568,000 31 $16,931,449,000 $19,382,761,000 $19,382,761,000 32 $18,190,598,000 $20,824,207,000 $20,824,207,000 33 $19,509,126,000 $22,333,629,000 $22,333,629 ,000 34 $20,888,893,000 $23,913,157,000 $23,913,157,000 35 $22,332,171,000 $25,565,391,000 $25,565,391,000 36 $23,841,577,000 $27,293,327,000 $27,293,327,000 37 $25,412,615,000 $29,091,817,000 $29,091,817,000 38 $27,047,585,000 $30,963,496,000 $30,963 ,496,000 39 $28,742,599,000 $32,903,911,000 $32,903,911,000 40 $30,499,011,000 $34,914,614,000 $34,914,614,000 41 $32,319,056,000 $36,998,163,000 $36,998,163,000 42 $34,203,766,000 $39,155,739,000 $39,155,739,000 43 $36,156,627,000 $41,391,332,000 $41 ,391,332,000 44 $38,177,286,000 $43,704,540,000 $43,704,540,000 45 $40,270,162,000 $46,100,419,000 $46,100,419,000 46 $42,433,827,000 $48,577,336,000 $48,577,336,000 47 $42,993,812,000 $49,218,395,000 $49,218,395,000

PAGE 180

180 Elizabeth River Crossing Table B 31. ERC s ummary Financial base year 2010 Term 58 Start d ate 2012 End d ate 2070 Construction end date 2017 Inflation r ate 2.5% Total c ost 2,040,840,000 Bond r ating BBB /Baa3 TIFIA r ate 3.16% TIFIA l ength 40 Risk free rate 3.15% PAB l ength 3 0 Formulation of the Simulated Cash Flow Model The following information explains each step in the formulation of the cash flow model for each PPP project. Table B 32. ERC convert nominal to real dollars Year O&M e xpenses (year of $) O&M e xpenses (201 0$) Major maintenance costs (year of $) Major maintenance costs (2010$) 2012 35,528,000 29,888,471 11,897,000 10,008,533 2013 33,779,000 27,723,998 8,424,000 6,913,969 2014 34,104,000 27,308,040 9,182,000 7,352,288 2015 34,832,000 27,210,703 7,407,000 5,786,337 2016 36,532,000 27,842,673 9,663,000 7,364,605 2017 37,100,000 27,585,923 15,252,000 11,340,714 2018 37,706,000 27,352,701 10,041,000 7,283,946 2019 38,616,000 27,329,593 7,175,000 5,077,943 2020 39,672,000 27,392,150 10,254,000 7,080,034 2 021 40,930,000 27,571,469 12,086,000 8,141,431 2022 42,153,000 27,702,743 17,177,000 11,288,639 2023 43,249,000 27,729,784 13,016,000 8,345,415 2024 44,528,000 27,853,498 5,910,000 3,696,869 2025 45,882,000 28,000,451 18,053,000 11,017,221 2026 48,545 ,000 28,903,027 18,057,000 10,750,890 2027 49,350,000 28,665,671 25,778,000 14,973,529 2028 50,351,000 28,533,773 23,235,000 13,167,210 2029 51,821,000 28,650,554 3,091,000 1,708,938 2030 53,377,000 28,791,051 17,901,000 9,655,631 2031 55,328,000 29,1 15,515 14,296,000 7,523,052 2032 56,951,000 29,238,628 23,640,000 12,136,770 2033 58,343,000 29,222,713 11,070,000 5,544,717 2034 60,105,000 29,370,985 6,585,000 3,217,834 2035 61,904,000 29,512,279 19,092,000 9,101,971 2036 65,435,000 30,434,788 24,3 16,000 11,309,732 2037 35,528,000 29,888,471 11,897,000 10,008,533 2038 33,779,000 27,723,998 8,424,000 6,913,969 2039 34,104,000 27,308,040 9,182,000 7,352,288

PAGE 181

181 Table B 33. ERC traffic forecast Year Lenders base transactions (average year ) 20 12 6,931,872 2013 30,359,070 2014 37,456,210 2015 38,611,080 2016 42,921,455 2017 45,686,870 2018 45,774,751 2019 46,451,631 2020 47,140,871 2021 47,842,738 2022 48,557,503 2023 49,285,443 2024 50,026,848 2025 50,782,004 2026 51,515,073 2027 52,096,206 2028 52,688,374 2029 53,289,921 2030 53,902,941 2031 54,527,705 2032 55,164,497 2033 55,813,599 2034 56,337,154 2035 56,674,235 2036 57,014,631 2037 57,358,379 2038 57,705,522 2039 58,056,099

PAGE 182

182 Table B 34. ERC traffic & revenue f orecast Year Lenders base transactions (average year ) Lenders base annual revenue (2010 $) 2012 6,931,872 16,500,000 2013 30,359,070 70,500,000 2014 37,456,210 83,900,000 2015 38,611,080 84,700,000 2016 42,921,455 90,500 ,000 2017 45,686,870 94,100,000 2018 45,774,751 94,200,000 2019 46,451,631 95,700,000 2020 47,140,871 97,300,000 2021 47,842,738 98,700,000 2022 48,557,503 100,300,000 2023 49,285,443 102,000,000 2024 50,026,848 103,600,000 2025 50,782,004 105,200 ,000 2026 51,515,073 106,400,000 2027 52,096,206 109,100,000 2028 52,688,374 111,200,000 2029 53,289,921 113,600,000 2030 53,902,941 116,200,000 2031 54,527,705 118,800,000 2032 55,164,497 121,500,000 2033 55,813,599 124,400,000 2034 56,337,154 12 6,700,000 2035 56,674,235 128,800,000 2036 57,014,631 130,900,000 2037 57,358,379 133,000,000 2038 57,705,522 135,200,000 2039 58,056,099 137,400,000

PAGE 183

183 Figure B 10. ERC revenue vs. traffic Table B 35. ERC T&R regression analysis summary output Regre ssion s tatistics Multiple R 0.940187976 R s quare 0.88395343 Adjusted r s quare 0.881678007 Standard e rror 0.075592115 Observations 53 ANOVA df SS MS F Significance F Regression 1 2.219832726 2.219832726 388.4787356 1.65907E 25 Residual 51 0.29142 2564 0.005714168 Total 52 2.51125529 VARIABLES Coefficients Standard e rror t Stat P value Intercept 16.210372 0.129421623 125.252425 3.68729E 65 Traffic 4.498E 08 2.2819E 09 19.70986392 1.65907E 25 Intercept t ransformed 10,966,680 VARI ABLES CONTINUED Lower 95% Upper 95% Lower 95.0% Upper 95.0% Intercept 15.95054742 16.4701969 15.95054742 16.4701969 Traffic 4.03948E 08 4.9557E 08 4.03948E 08 4.9557E 08

PAGE 184

184 Table B 35. Continued Regression s tatistics RESIDUAL OUTPUT Observation Predict ed Residuals 1 18.265182 0.094686564 2 18.269135 0.091796169 3 18.299578 0.077150972 4 18.330577 0.062732447 5 18.362144 0.045451277 6 18.394291 0.029384798 7 18.427031 0.013452131 8 18.460377 0.004328738 9 18.49434 0.022966618 10 18.527311 0 .044594808 11 18.553448 0.045672494 12 18.580081 0.053240317 13 18.607136 0.058942334 14 18.634708 0.063884146 15 18.662807 0.069854931 16 18.691447 0.076022393 17 18.720641 0.081628447 18 18.744189 0.086855918 19 18.759349 0.085577726 2 0 18.774659 0.084714497 21 18.790119 0.084259431 22 18.805732 0.083466478 23 18.8215 0.083092791 24 18.837424 0.082415726 25 18.845465 0.077654388 26 18.853546 0.072397229 27 18.861667 0.138662017 28 18.869829 0.061847147 29 18.878032 0. 05723345 30 18.882154 0.050688502 31 18.886286 0.042955003 32 18.890428 0.036018916 33 18.894581 0.027933292 34 18.898744 0.021271896 35 18.900831 0.012650015 36 18.902921 0.004764363 37 18.905013 0.003634222 38 18.907108 0.012528589 39 18 .907856 0.022650222

PAGE 185

185 Table B 35. Continued Regression s tatistics RESIDUAL OUTPUT Observation Predicted Residuals 40 18.910255 0.030409328 41 18.911306 0.039414927 42 18.912357 0.048902353 43 18.913409 0.057702663 44 18.914462 0.066406221 45 18.9 14988 0.075541463 46 18.915515 0.084584066 47 18.916042 0.094090612 48 18.916569 0.103497351 49 18.917096 0.113349824 50 18.91736 0.122821109 51 18.917624 0.132198535 52 18.917888 0.14148385 53 18.918152 0.15172432 Table B 36. ERC O& M e xpenses Y ear Revenues (2010$) O&M e xpense (2010$) 2017 91,787,085 29,888,471 2018 92,433,300 27,723,998 2019 94,132,825 27,308,040 2020 95,888,198 27,210,703 2021 97,631,509 27,842,673 2022 97,907,721 27,585,923 2023 99,555,967 27,352,701 2024 101,255,238 2 7,329,593 2025 102,799,274 27,392,150 2026 104,368,079 27,571,469 2027 106,305,901 27,702,743 2028 108,737,891 27,729,784 2029 111,070,578 27,853,498 2030 113,746,570 28,000,451 2031 116,234,883 28,903,027 2032 118,804,832 28,665,671 2033 121,548, 057 28,533,773 2034 123,941,385 28,650,554 2035 126,042,096 28,791,051

PAGE 186

186 Table B 36. Continued Year Revenues (2010$) O&M e xpense (2010$) 2036 128,094,477 29,115,515 2037 130,107,813 29,238,628 2038 132,180,144 29,222,713 2039 134,465,894 29,370,985 2040 136,848,988 29,512,279 2041 138,527,936 30,434,788 Figure B 11. ERC O&M vs. t raffic Table B 37. ERC O&M regression analysis summary output Regression s tatistics Multiple R 0.878897529 R s quare 0.772460867 Adjusted R s quare 0.762118179 Sta ndard e rror 429944.9022 Observations 24 ANOVA df SS MS F Significance F Regression 1 1.3806E+13 1.3806E+13 74.68666522 1.60061E 08 Residual 22 4.06676E+12 1.84853E+11 Total 23 1.78728E+13

PAGE 187

187 Table B 37. Continued Regression s tatistics VARIABLES Coefficients Standard e rror t Stat P value Intercept 18295982.24 1160150.973 15.77034599 1.78577E 13 Traffic 0.188186788 0.021775473 8.642144712 1.60061E 08 VARIABLES CONTINUED Lower 95% Upper 95% Lower 95.0% Upper 95.0% Intercept 15889976.4 20701 988.09 15889976.4 20701988.09 Traffic 0.143027222 0.233346355 0.143027222 0.233346355 RESIDUAL OUTPUT Observation Predicted Y Residuals 1 26910185.63 813812.7869 2 27037565.5 270474.5477 3 27167271.36 43431.3648 4 27299353.46 543319.7215 5 27433862 .79 152060.5445 6 27570851.48 218150.7971 7 27710374.11 380780.7159 8 27852484.49 460334.9189 9 27990438.39 418969.8625 10 28099799.94 397056.7028 11 28211238.14 481453.7247 12 28324441.34 470943.1786 13 28439803.6 439352.2007 14 28557375. 93 345651.3085 15 28677211.77 11540.35833 16 28799364.19 265591.5613 17 28897890.33 247336.5986 18 28961324.52 170273.0257 19 29025382.55 90132.11763 20 29090071.38 148556.5403 21 29155399.11 67313.4224 22 29221373.07 149611.5031 23 29288000.9 8 224278.2076 24 29321646.33 1113141.58

PAGE 188

188 Table B 38. ERC accumulated major maintenance Year Traffic Accumulated traffic Major maintenance cos ts (2010$) Accum major maintenance cost (2010$) 2017 45,686,870 45,686,870 11,897,000 11,897,000 2018 45,774 ,751 91,461,621 8,424,000 20,321,000 2019 46,451,631 137,913,252 9,182,000 29,503,000 2020 47,140,871 185,054,123 7,407,000 36,910,000 2021 47,842,738 232,896,861 9,663,000 46,573,000 2022 48,557,503 281,454,364 15,252,000 61,825,000 2023 49,285,443 3 30,739,807 10,041,000 71,866,000 2024 50,026,848 380,766,655 7,175,000 79,041,000 2025 50,782,004 431,548,659 10,254,000 89,295,000 2026 51,515,073 483,063,732 12,086,000 101,381,000 2027 52,096,206 535,159,938 17,177,000 118,558,000 2028 52,688,374 5 87,848,312 13,016,000 131,574,000 2029 53,289,921 641,138,233 5,910,000 137,484,000 2030 53,902,941 695,041,174 18,053,000 155,537,000 2031 54,527,705 749,568,879 18,057,000 173,594,000 2032 55,164,497 804,733,376 25,778,000 199,372,000 2033 55,813,59 9 860,546,975 23,235,000 222,607,000 2034 56,337,154 916,884,129 3,091,000 225,698,000 2035 56,674,235 973,558,364 17,901,000 243,599,000 2036 57,014,631 1,030,572,995 14,296,000 257,895,000 2037 57,358,379 1,087,931,374 23,640,000 281,535,000 2038 57 ,705,522 1,145,636,896 11,070,000 292,605,000 2039 58,056,099 1,203,692,995 6,585,000 299,190,000 2040 58,410,151 1,262,103,146 19,092,000 318,282,000

PAGE 189

189 Figure B 12. ERC cost vs. accumulated traffic Table B 39. ERC accumulated major maintenance & accu mulated traffic regression analysis summary output Regression s tatistics Multiple R 0.996758114 R s quare 0.993526738 Adjusted r s quare 0.993245292 Standard e rror 8529601.048 Observations 25 ANOVA df SS MS F Significance F Regression 1 2.56828E+1 7 2.56828E+17 3530.077268 1.11063E 26 Residual 23 1.67334E+15 7.27541E+13 Total 24 2.58501E+17 VARIABLES Coefficients Standard e rror t Stat P value Intercept 15027005.26 3374337.796 4.453319783 0.00018186 Accumulated traffic 0.263431745 0. 004433799 59.41445336 1.11063E 26 VARIABLES CONTINUED Lower 95% Upper 95% Lower 95.0% Upper 95.0% Intercept 22007354.78 8046655.739 22007354.8 8046655.74 Accumulated traffic 0.254259733 0.272603757 0.254259733 0.272603757

PAGE 190

190 Table B 39. Continued Regression s tatistics RESIDUAL OUTPUT Observation Predicted accumulated cost Residuals 1 2991633.369 14888633.37 2 9066889.167 11254110.83 3 21303723.38 8199276.617 4 33722125.3 3187874.705 5 46325421.26 247578.7443 6 59117009.01 2707990.992 7 72 100359.26 234359.264 8 85279019.13 6238019.133 9 98656611.07 9361611.066 10 112227316.6 10846316.64 11 125951111.1 7393111.102 12 139830901.4 8256901.411 13 153869158.3 16385158.29 14 168068904.1 12531904.11 15 182433232.6 8839232.59 16 1 96965312.3 2406687.699 17 211668386.1 10938613.91 18 226509380.9 811380.8759 19 241439173.5 2159826.497 20 256458637.2 1436362.758 21 271568655.1 9966344.883 22 286770121.5 5834878.522 23 302063941 2873940.95 24 317451029 830971.0417 25 33288521 5.1 9712784.862

PAGE 191

191 Table B 40. ERC revenue s haring Band 1 Band 2 Band 3 Year Floor Ceiling Floor Ceiling Floor Ceiling 1 108,488,000 108,488,000 113,654,000 113,654,000 123,986,000 2 223,215,000 223,215,000 233,844,000 233,844,000 255,103,000 3 344,158,000 344,158,000 360,546,000 360,546,000 393,323,000 4 471,569,000 471,569,000 494,024,000 494,024,000 538,936,000 5 605,843,000 605,843,000 634,693,000 634,693,000 692,392,000 6 746,391,000 746,391,000 781,933,000 781,933,000 853,018,000 7 893,334,000 893,334,000 935,874,000 935,874,000 1,020,953,000 8 1,048,044,000 1,048,044,000 1,097,951,000 1,097,951,000 1,197,765,000 9 1,210,792,000 1,210,792,000 1,268,449,000 1,268,449,000 1,383,762,000 10 1,381,889,000 1,381,889,000 1,44 7,693,000 1,447,693,000 1,579,302,000 11 1,560,726,000 1,560,726,000 1,635,046,000 1,635,046,000 1,783,687,000 12 1,748,217,000 1,748,217,000 1,831,446,000 1,831,446,000 1,997,963,000 13 1,944,438,000 1,944,438,000 2,037,031,000 2,037,031,000 2,22 2,215,000 14 2,150,184,000 2,150,184,000 2,252,574,000 2,252,574,000 2,457,363,000 15 2,365,694,000 2,365,694,000 2,478,346,000 2,478,346,000 2,703,650,000 16 2,591,550,000 2,591,550,000 2,714,958,000 2,714,958,000 2,961,772,000 17 2,828,458,00 0 2,828,458,000 2,963,146,000 2,963,146,000 3,232,523,000 18 3,076,416,000 3,076,416,000 3,222,912,000 3,222,912,000 3,515,904,000 19 3,334,725,000 3,334,725,000 3,493,521,000 3,493,521,000 3,811,114,000 20 3,603,841,000 3,603,841,000 3,775,452,00 0 3,775,452,000 4,118,675,000 21 3,884,048,000 3,884,048,000 4,069,003,000 4,069,003,000 4,438,912,000 22 4,175,861,000 4,175,861,000 4,374,711,000 4,374,711,000 4,772,412,000 23 4,479,989,000 4,479,989,000 4,693,322,000 4,693,322,000 5,119,987,00 0 24 4,796,997,000 4,796,997,000 5,025,425,000 5,025,425,000 5,482,282,000 25 5,125,969,000 5,125,969,000 5,370,063,000 5,370,063,000 5,858,251,000 26 5,467,653,000 5,467,653,000 5,728,017,000 5,728,017,000 6,248,746,000

PAGE 192

192 Table B 40. Continued Band 1 Band 2 Band 3 Year Floor Ceiling Floor Ceiling Floor Ceiling 27 5,822,593,000 5,822,593,000 6,099,859,000 6,099,859,000 6,654,392,000 28 6,191,243,000 6,191,243,000 6,486,064,000 6,486,064,000 7,075,706,000 29 6,574,007,000 6,574,007,00 0 6,887,055,000 6,887,055,000 7,513,151,000 30 6,970,734,000 6,970,734,000 7,302,674,000 7,302,674,000 7,966,553,000 31 7,381,765,000 7,381,765,000 7,733,278,000 7,733,278,000 8,436,303,000 32 7,808,280,000 7,808,280,000 8,180,102,000 8,180,102,00 0 8,923,748,000 33 8,250,287,000 8,250,287,000 8,643,158,000 8,643,158,000 9,428,899,000 34 8,708,690,000 8,708,690,000 9,123,389,000 9,123,389,000 9,952,788,000 35 9,183,452,000 9,183,452,000 9,620,759,000 9,620,759,000 10,495,374,000 36 9,674 ,992,000 9,674,992,000 10,135,706,000 10,135,706,000 11,057,134,000 37 10,184,065,000 10,184,065,000 10,669,021,000 10,669,021,000 11,638,932,000 38 10,711,950,000 10,711,950,000 11,222,043,000 11,222,043,000 12,242,229,000 39 11,258,853,000 11,25 8,853,000 11,794,988,000 11,794,988,000 12,867,260,000 40 11,825,063,000 11,825,063,000 12,388,161,000 12,388,161,000 13,514,357,000 41 12,411,323,000 12,411,323,000 13,002,338,000 13,002,338,000 14,184,369,000 42 13,018,428,000 13,018,428,000 13, 638,353,000 13,638,353,000 14,878,203,000 43 13,646,933,000 13,646,933,000 14,296,787,000 14,296,787,000 15,596,495,000 44 14,298,079,000 14,298,079,000 14,978,940,000 14,978,940,000 16,340,662,000 45 14,972,386,000 14,972,386,000 15,685,356,000 1 5,685,356,000 17,111,298,000 46 15,670,767,000 15,670,767,000 16,416,994,000 16,416,994,000 17,909,448,000 47 16,394,002,000 16,394,002,000 17,174,669,000 17,174,669,000 18,736,003,000 48 17,143,191,000 17,143,191,000 17,959,534,000 17,959,534,000 19,592,218,000 49 17,919,004,000 17,919,004,000 18,772,290,000 18,772,290,000 20,478,861,000 50 18,722,111,000 18,722,111,000 19,613,640,000 19,613,640,000 21,396,698,000 51 19,552,840,000 19,552,840,000 20,483,927,000 20,483,927,000 22,346,102,0 00 52 20,411,102,000 20,411,102,000 21,383,059,000 21,383,059,000 23,326,974,000 53 21,298,201,000 21,298,201,000 22,312,401,000 22,312,401,000 24,340,801,000

PAGE 193

193 Table B 40. Continued Band 4 Band 5 Year Floor Ceiling Floor 1 123,986,000 134,318,00 0 134,318,000 2 255,103,000 276,361,000 276,361,000 3 393,323,000 426,100,000 426,100,000 4 538,936,000 583,847,000 583,847,000 5 692,392,000 750,092,000 750,092,000 6 853,018,000 924,103,000 924,103,000 7 1,020,953,000 1,106,033,000 1,106,033,000 8 1,197,765,000 1,297,578,000 1,297,578,000 9 1,383,762,000 1,499,076,000 1,499,076,000 10 1,579,302,000 1,710,910,000 1,710,910,000 11 1,783,687,000 1,932,328,000 1,932,328,000 12 1,997,963,000 2,164,460,000 2,164,460,000 13 2,222,215,000 2,407,400,00 0 2,407,400,000 14 2,457,363,000 2,662,133,000 2,662,133,000 15 2,703,650,000 2,928,955,000 2,928,955,000 16 2,961,772,000 3,208,586,000 3,208,586,000 17 3,232,523,000 3,501,900,000 3,501,900,000 18 3,515,904,000 3,808,896,000 3,808,896,000 19 3,811, 114,000 4,128,707,000 4,128,707,000 20 4,118,675,000 4,461,898,000 4,461,898,000 21 4,438,912,000 4,808,822,000 4,808,822,000 22 4,772,412,000 5,170,114,000 5,170,114,000 23 5,119,987,000 5,546,653,000 5,546,653,000 24 5,482,282,000 5,939,139,000 5,93 9,139,000 25 5,858,251,000 6,346,438,000 6,346,438,000 26 6,248,746,000 6,769,475,000 6,769,475,000 27 6,654,392,000 7,208,925,000 7,208,925,000

PAGE 194

194 Table B 40. Continued Band 4 Band 5 Year Floor Ceiling Floor 28 7,075,706,000 7,665,348,000 7,665,348, 000 29 7,513,151,000 8,139,247,000 8,139,247,000 30 7,966,553,000 8,630,432,000 8,630,432,000 31 8,436,303,000 9,139,329,000 9,139,329,000 32 8,923,748,000 9,667,394,000 9,667,394,000 33 9,428,899,000 10,214,641,000 10,214,641,000 34 9,952,788,000 10 ,782,187,000 10,782,187,000 35 10,495,374,000 11,369,988,000 11,369,988,000 36 11,057,134,000 11,978,562,000 11,978,562,000 37 11,638,932,000 12,608,843,000 12,608,843,000 38 12,242,229,000 13,262,415,000 13,262,415,000 39 12,867,260,000 13,939,532,00 0 13,939,532,000 40 13,514,357,000 14,640,554,000 14,640,554,000 41 14,184,369,000 15,366,399,000 15,366,399,000 42 14,878,203,000 16,118,054,000 16,118,054,000 43 15,596,495,000 16,896,203,000 16,896,203,000 44 16,340,662,000 17,702,384,000 17,702,38 4,000 45 17,111,298,000 18,537,239,000 18,537,239,000 46 17,909,448,000 19,401,902,000 19,401,902,000 47 18,736,003,000 20,297,336,000 20,297,336,000 48 19,592,218,000 21,224,903,000 21,224,903,000 49 20,478,861,000 22,185,433,000 22,185,433,000 50 2 1,396,698,000 23,179,756,000 23,179,756,000 51 22,346,102,000 24,208,278,000 24,208,278,000 52 23,326,974,000 25,270,888,000 25,270,888,000 53 24,340,801,000 26,369,201,000 26,369,201,000

PAGE 195

195 I 95 Table B 41. I 95 s ummary Base y ear 2010 Term 85 years Start d ate 2012 End d ate 2087 Construction end d ate 2015 Inflation r ate 2.5% Total c ost 924,649,000 Bond r ating BBB TIFIA r ate 2.56% PAB l ength 30 Risk free rate 2.55% TIFIA l ength 35 Formulation of the Simulated Cash Flow Model Table B 42. I 95 convert nominal to real dollars Year O&M e xpenses (year of $) O&M e xpenses (2010$) Major maintenance costs (year of $) Major maintenance costs (2010 $) 2015 34,027,000 30,074,910 2016 31,823,000 27,440,873 2017 31,586,000 26,572,204 2018 30,748,000 25,236,316 2019 31,725,000 25,403,107 4,906,000 3,928,373 2020 32,746,000 25,581,123 2021 33,792,000 25,754,396 4,490,000 3,422,030 2022 34,863,000 25,922,589 459,000 341,292 2023 37,052,000 26,878,276 21,364,000 15,497,881 2024 37,304,000 26,401,055 33,771,000 23,900,655 2025 38,393,000 26,509,044 1,676,000 1,157,220 2026 39,468,000 26,586,629 5,511,000 3,712,347

PAGE 196

1 96 Table B 42. Continued Year O&M e xpenses (year of $) O&M e xpenses (2010$) Major maintenance costs (year of $) Major maintenance costs (2010$) 2027 40,681,000 26,735,352 2,468,000 1,621,957 2028 41,923,000 26,879,598 6,042,000 3,873,924 2029 43,199,000 27,022,172 6,396,000 4,000,875 2030 44,510,000 27,163,160 9,892,000 6,036,800 2031 47,221,000 28,114,736 39, 720,000 23,648,743 2032 47,533,000 27,610,240 2033 48,856,000 27,686,560 14,060,000 7,967,763 2034 50,203,000 27,756,001 44,430,000 24,564,252 2035 51,776,000 27,927,487 13,402,000 7,228,913 2036 53,348,000 28,073,570 10,062,000 5,294,974 2037 54 ,970,000 28,221,583 9,424,000 4,838,279 2038 56,637,000 28,368,215 8,109,000 4,061,618 2039 60,043,000 29,340,688 19,187,000 9,375,943 2040 60,474,000 28,830,537 761,000 362,801

PAGE 197

197 Table B 43. I 95 traffic forecast Year Lenders base transactions (a verage year ) 2015 19,300,000 2016 23,200,000 2017 25,500,000 2018 26,000,000 2019 26,300,000 2020 26,800,000 2021 27,200,000 2022 27,500,000 2023 27,800,000 2024 28,100,000 2025 28,400,000 2026 28,600,000 2027 28,900,000 2028 29,000,000 2029 29,200,000 2030 29,400,000 2031 29,600,000 2032 29,800,000 2033 30,000,000 2034 30,100,000 2035 30,300,000 2036 30,500,000 2037 30,700,000 2038 30,900,000 2039 31,000,000 2040 31,200,000 2041 31,400,000 2042 31,500,000 2043 31,700,000 2044 31,800,000 2045 32,100,000 2046 32,200,000 2047 32,300,000 2048 32,500,000 2049 32,700,000 2050 32,800,000 2051 33,393,048 2052 33,597,357

PAGE 198

198 Table B 43. Continued Year Lenders base transactions (average year ) 2053 33,801,665 2054 34,005,974 2055 34,210,283 2056 34,414,591 2057 34,618,900 2058 34,823,209 2059 35,027,517 2060 35,231,826 2061 35,436,134 2062 35,640,443 2063 35,844,752 2064 36,049,060 2065 36,253,369 2066 36,457,678 2067 36,661,986 2068 36,866,295 2069 37,070,604 2070 37,274,912 2071 37,479,221 2072 37,683,529 2073 37,887,838 2074 38,092,147 2075 38,296,455 2076 38,500,764 2077 38,705,073 2078 38,909,381 2079 39,113,690 2080 39,317,998 2081 39,522,307 2082 39,726,616 2083 39,930,924 2084 40,135,233 2085 40,339,542 2086 40,543,850 2087 40,748,159

PAGE 199

199 Table B 44. I 95 traffic & revenue forecast Year Lenders base transactions (average year ) Lenders base annual revenue (2010 $) 2015 50,913,370 45,000,000 2016 69,349,666 59,800,000 2017 81,543,843 68,600,00 0 2018 86,141,085 70,700,000 2019 91,042,111 72,900,000 2020 96,134,349 75,100,000 2021 101,555,507 77,400,000 2022 107,187,639 79,700,000 2023 113,037,906 82,000,000 2024 119,113,693 84,300,000 2025 125,567,451 86,700,000 2026 132,269,451 89,100, 000 2027 139,228,071 91,500,000 2028 146,451,954 93,900,000 2029 154,109,878 96,400,000 2030 162,059,166 98,900,000 2031 170,309,600 101,400,000 2032 179,043,425 104,000,000 2033 188,107,499 106,600,000 2034 197,512,874 109,200,000 2035 207,456,34 5 111,900,000 2036 217,583,514 114,500,000 2037 228,282,162 117,200,000 2038 239,579,402 120,000,000 2039 251,094,187 122,700,000 2040 263,244,731 125,500,000 2041 275,845,868 128,300,000 2042 288,912,535 131,100,000 2043 302,460,130 133,900,000 2 044 316,504,536 136,700,000 2045 331,062,123 139,500,000 2046 345,906,522 142,200,000 2047 361,535,561 145,000,000 2048 377,729,861 147,800,000 2049 394,507,916 150,600,000 2050 411,888,793 153,400,000

PAGE 200

200 Figure B 13. I 95 revenue vs. traffic Table B 45. I 95 T& R regression analysis summary output Regression s tatistics Multiple R 0.983313844 R s quare 0.966906116 Adjusted r s quare 0.965932766 Standard e rror 0.053626482 Observations 36 ANOVA df SS MS F Significance F Regression 1 2.85676228 1 2.856762281 993.3801597 9.45399E 27 Residual 34 0.097777187 0.0028758 Total 35 2.954539468 VARIABLES Coefficients Standard e rror t Stat P value Intercept 15.5175537 0.093450097 166.0517661 4.6884E 51 Traffic 9.99211E 08 3.17029E 09 31.517 93394 9.45399E 27 Intercept t ransformed 5,485,143 VARIABLES CONTINUED Lower 95% Upper 95% Lower 95.0% Upper 95.0% Intercept 15.32764025 15.70746714 15.32764025 15.70746714 Traffic 9.34783E 08 1.06364E 07 9.34783E 08 1.06364E 07

PAGE 201

201 Table B 45. Conti nued Regression Statistics RESIDUAL OUTPUT Observation Predicted Y Residuals 1 17.44603113 0.180576518 2 17.83572346 0.075796943 3 18.06554201 0.014476722 4 18.11550257 0.03309568 5 18.1454789 0.031323315 6 18.19543945 0.050512212 7 18.2354079 0.060627693 8 18.26538423 0.060375024 9 18.29536056 0.062214944 10 18.3253369 0.063651885 11 18.35531323 0.065880773 12 18.37529745 0.057757468 13 18.40527379 0.061474086 14 18.4152659 0.044826369 15 18.43525012 0.038862067 16 18.45523434 0.032555594 17 18.47521856 0.027895888 18 18.49520279 0.022878808 19 18.51518701 0.018471577 20 18.52517912 0.006464634 21 18.54516334 0.000496617 22 18.56514756 0.002227561 23 18.58513178 0.005291734 24 18.60511601 0.007836625 25 18.615108 12 0.019877229 26 18.63509234 0.021450728 27 18.65507656 0.023342378 28 18.66506867 0.034001097 29 18.68505289 0.034991426 30 18.69504501 0.045542958 31 18.72502134 0.035696708 32 18.73501345 0.045437443 33 18.74500556 0.054796316 34 18.76498978 0 .053795715 35 18.78497401 0.052441439 36 18.79496612 0.060738538

PAGE 202

202 Table B 46. I 95 O& M Expenses Year Revenues (2010$) O&M e xpense (2010$) 2015 49,230,684 30,074,910 2016 64,261,812 27,440,873 2017 75,951,949 26,572,204 2 018 79,780,670 25,236,316 2019 83,667,306 25,403,107 2020 80,822,787 25,581,123 2021 83,468,572 25,754,396 2022 86,071,055 25,922,589 2023 89,381,221 26,878,276 2024 90,644,992 26,401,055 2025 92,834,475 26,509,044 2026 95,334,769 26,586,629 2027 98,006,842 26,735,352 2028 100,731,011 26,879,598 2029 103,482,301 27,022,172 2030 106,403,180 27,163,160 2031 109,948,199 28,1 14,736 2032 111,629,410 27,610,240 2033 114,540,846 27,686,560 2034 117,029,338 27,756,001 2035 119,787,323 27,927,487 2036 122,439,559 28,073,570 2037 125,207,412 28,221,583 2038 128,025,864 28,368,215 2039 131,729,880 29,340,688 2040 134,197,822 28,830,537

PAGE 203

203 Figure B 14. I 95 O&M vs. t raffic

PAGE 204

204 Table B 47. I 95 O&M regression analysis summary output Regression s tatistics Multiple R 0.964183226 R s quare 0.92964929 4 Adjusted r s quare 0.92629926 Standard e rror 304932.324 Observations 23 ANOVA df SS MS F Significance F Regression 1 2.58034E+13 2.58034E+13 277.5044674 1.40089E 13 Residual 21 1.95266E+12 92983722197 Total 22 2.77561E+13 VARIABLES Coef ficients Standard e rror t Stat P value Intercept 6709828.84 1227493.786 5.466283345 2.0106E 05 Traffic 0.703845904 0.042251545 16.65846534 1.40089E 13 VARIABLES CONTINUED Lower 95% Upper 95% Lower 95.0% Upper 95.0% Intercept 4157115.768 9262541.912 4 157115.768 9262541.912 Traffic 0.615979005 0.791712802 0.615979005 0.791712802 RESIDUAL OUTPUT Observation Predicted Y Residuals Standard Residuals 1 25009822.33 226493.2274 0.760244785 2 25220976.1 182131.1741 0.611339583 3 25572899.05 8223.808102 0 .027603948 4 25854437.42 100040.9371 0.335796357 5 26065591.19 143002.367 0.480000241 6 26276744.96 601530.8004 2.019091958 7 26487898.73 86843.41878 0.291497706 8 26699052.5 190008.3768 0.637780119 9 26839821.68 253192.8058 0.849864309 10 27050975.45 315623.3354 1.059417968 11 27121360.04 241761.6253 0.811494529 12 27262129.22 239957.4012 0.805438489 13 27402898.4 239738.7374 0.804704524 14 27543667.58 571068.2131 1.916841558 15 27684436.77 74196.45428 0.249047038 16 278252 05.95 138645.6845 0.465376646 17 27895590.54 139589.1303 0.468543407 18 28036359.72 108872.5579 0.365440483

PAGE 205

205 Table B 47. Continued Regression Statistics RESIDUAL OUTPUT Observation Predicted Y Residuals Standard Residuals 19 28177128.9 103558. 9799 0.347604983 20 28317898.08 96315.01899 0.32328998 21 28458667.26 90452.22855 0.303610999 22 28529051.85 811635.7228 2.724327931 23 28669821.03 160716.1132 0.539458015 24 29264995.72 434458.5729 0.614186105 25 29613678.73 272991.1556 0.1 03374935 26 29804734.72 974197.5763 0.3689043 Table B 48. I 95 accumulated major maintenance table Year Traffic Accumulated tr affic Major maintenance costs (2010$) A ccum major maintenance cost (2010$) 2015 19,300,000 19,300,000 2016 23,200,000 42,500,000 2017 25,500,000 68,000,000 2018 26,000,000 94,000,000 2019 26,300,000 3,928,373 120,300,000 3,928,373 2020 26,800,000 147,100,000 3,928,373 2021 27,200,000 3,422,030 174,300,000 7,350,403 2022 27,500,000 341,292 201,800,000 7, 691,696 2023 27,800,000 15,497,881 229,600,000 23,189,576 2024 28,100,000 23,900,655 257,700,000 47,090,232 2025 28,400,000 1,157,220 286,100,000 48,247,452 2026 28,600,000 3,712,347 314,700,000 51,959,799 2027 28,900,000 1,621,957 343,600,000 53,581, 756 2028 29,000,000 3,873,924 372,600,000 57,455,681 2029 29,200,000 4,000,875 401,800,000 61,456,556 2030 29,400,000 6,036,800 431,200,000 67,493,356 2031 29,600,000 23,648,743 460,800,000 91,142,099 2032 29,800,000 490,600,000 91,142,099 2033 30, 000,000 7,967,763 520,600,000 99,109,863 2034 30,100,000 24,564,252 550,700,000 123,674,115 2035 30,300,000 7,228,913 581,000,000 130,903,027 2036 30,500,000 5,294,974 611,500,000 136,198,001 2037 30,700,000 4,838,279 642,200,000 141,036,280 2038 30,9 00,000 4,061,618 673,100,000 145,097,898 2039 31,000,000 9,375,943 704,100,000 154,473,841 2040 31,200,000 362,801 735,300,000 154,836,643

PAGE 206

206 Figure B 15. I 95 cost vs. accumulated traffic Table B 49. I 95 major maintenance regression analysis summary o utput Regression s tatistics Multiple R 0.989502186 R s quare 0.979114576 Adjusted r s quare 0.978070304 Standard e rror 7722830.689 Observations 22 ANOVA df SS MS F Significance F Regression 1 5.59208E+16 5.59208E+16 937.6056327 2.80928E 18 Residua l 20 1.19284E+15 5.96421E+13 Total 21 5.71136E+16 VARIABLES Coefficients Standard e rror t Stat P value Intercept 36560467.86 4067211.308 8.989075092 1.83992E 08 Traffic 0.270824631 0.008844597 30.62034671 2.80928E 18 VARIABLES CONTINUED Lo wer 95% Upper 95% Lower 95.0% Upper 95.0% Intercept 45044521.96 28076413.77 45044521.96 28076413.77 Traffic 0.252375125 0.289274137 0.252375125 0.289274137

PAGE 207

207 Table B 49. Continued Regression Statistics RESIDUAL OUTPUT Observation Predicted Y Resid uals 1 3980264.768 7908638.111 2 3277835.34 650538.0027 3 10644265.3 3293861.886 4 18091942.65 10400247.08 5 25620867.39 2431290.917 6 33231039.52 13859192.08 7 40922459.03 7324992.838 8 48668043.48 3291755.403 9 56494875.31 2913119.029 10 6 4348789.61 6893108.9 11 72256868.83 10800312.85 12 80219112.98 12725756.83 13 88235522.05 2906577.364 14 96306096.05 5163996.636 15 104430835 5320972.396 16 112582656.4 11091458.2 17 120788642.7 10114384.56 18 129048793.9 7149207.089 19 13736 3110.1 3673169.959 20 145731591.2 633693.1886 21 154127154.7 346686.7018 22 162576883.2 7740240.599

PAGE 208

208 APPENDIX C SIMULATED CASH FLOW MODEL (BASE CASE) RESULTS Table C 1. I 495 traffic & revenue cash flow section Year Traffic Revenue (2004$) Reven ue (year of $) O&M Expense Major maintenance beginning Balance Major maintenance expenses Major maintenance ending Balance Net cash flow available for debt service Revenue account balance 2008 (1,981,700,000) 2009 2010 2011 2012 2013 20,786,400 45,957,995 59,964,759 26,296,584 19,000,000 0 19,000,000 33,668,175 0 2014 30,003,000 65,803,771 88,434,766 28,852,603 19,000,000 0 19,000,000 59,582,163 0 2015 35,963,700 76,777,842 106,278,489 30,895,326 1 9,000,000 0 19,000,000 75,383,163 0 2016 36,724,200 78,072,834 111,313,194 31,976,878 19,000,000 7,724,630 11,275,370 79,336,315 0 2017 37,333,500 79,093,193 116,151,021 33,063,840 11,275,370 12,777,518 0 81,585,033 0 2018 38,037,900 80,253, 775 121,391,036 34,207,762 0 13,409,159 0 73,774,115 0 2019 38,755,800 81,415,597 126,842,848 35,393,563 0 14,072,101 0 77,377,184 0 2020 40,662,900 84,399,000 135,435,619 36,891,978 0 15,207,499 0 83,336,142 0 2021 41,394,900 85,504,379 141 ,325,711 38,171,348 0 15,945,697 0 87,208,665 0 2022 42,140,100 86,607,048 147,442,702 39,497,484 0 16,719,738 0 91,225,480 0 2023 42,898,800 87,706,222 153,793,392 40,872,210 0 17,531,387 0 95,389,795 0 2024 43,671,300 88,801,058 160,384,58 9 42,297,428 0 18,382,496 0 99,704,665 0 2025 44,457,300 89,889,830 167,221,563 43,774,958 0 19,274,746 0 104,171,859 0 2026 45,257,700 90,972,430 174,312,583 45,307,007 0 20,210,417 0 108,795,159 0 2027 46,072,500 92,047,436 181,663,578 46, 895,637 0 21,191,505 0 113,576,436 0 2028 46,902,000 93,113,780 189,281,144 48,543,072 0 22,220,233 0 118,517,839 0 2029 47,746,500 94,170,328 197,171,755 50,251,627 0 23,298,932 0 123,621,196 0 2030 48,606,000 95,215,516 205,340,951 52,023, 622 0 24,429,894 0 128,887,435 0 2031 49,465,800 96,230,662 213,756,111 53,856,805 0 25,607,900 0 134,291,406 0 2032 50,325,300 97,215,056 222,421,018 55,753,060 0 26,834,439 0 139,833,519 0

PAGE 209

209 Table C 1. Continued Year Traffic Revenue (2004$) Revenue (year of $) O&M Expense Major maintenance beginning Balance Major maintenance expenses Major maintenance ending Balance Net cash flow available for debt service Revenue account balance 2033 51,184,800 98,169,059 231,341,819 57,714,619 0 28,111,52 4 0 145,515,676 0 2034 52,044,600 99,092,988 240,524,691 59,743,798 0 29,441,252 0 151,339,641 0 2035 52,904,100 99,986,198 249,973,528 61,842,677 0 30,825,289 0 157,305,562 0 2036 52,904,100 99,986,198 257,472,734 63,697,957 0 31,750,048 0 162,024,729 0 2037 52,904,100 99,986,198 265,196,916 65,608,896 0 32,702,549 0 166,885,470 0 2038 52,904,100 99,986,198 273,152,823 67,577,163 0 33,683,626 0 171,892,035 0 2039 52,904,100 99,986,198 281,347,408 69,604,478 0 34,694,134 0 177,048,796 0 2040 52,904,1 00 99,986,198 289,787,830 71,692,612 0 35,734,958 0 182,360,259 0 2041 52,904,100 99,986,198 298,481,465 73,843,391 0 36,807,007 0 187,831,067 0 2042 52,904,100 99,986,198 307,435,909 76,058,692 0 37,911,217 0 193,465,999 0 2043 52,904,100 99,986,198 31 6,658,986 78,340,453 0 39,048,554 0 199,269,979 0 2044 52,904,100 99,986,198 326,158,756 80,690,667 0 40,220,011 0 205,248,079 0 2045 52,904,100 99,986,198 335,943,518 83,111,387 0 41,426,611 0 211,405,521 0 2046 52,904,100 99,986,198 346,021,824 85,604 ,728 0 42,669,409 0 217,747,687 0 2047 52,904,100 99,986,198 356,402,479 88,172,870 0 43,949,491 0 224,280,117 0 2048 52,904,100 99,986,198 367,094,553 90,818,056 0 45,267,976 0 231,008,521 0 2049 52,904,100 99,986,198 378,107,390 93,542,598 0 46,626,01 5 0 237,938,776 0 2050 52,904,100 99,986,198 389,450,611 96,348,876 0 48,024,796 0 245,076,940 0 2051 52,904,100 99,986,198 401,134,130 99,239,342 0 49,465,540 0 252,429,248 0 2052 52,904,100 99,986,198 413,168,154 102,216,522 0 50,949,506 0 260,002,125 0 2053 52,904,100 99,986,198 425,563,198 105,283,018 0 52,477,991 0 267,802,189 0 2054 52,904,100 99,986,198 438,330,094 108,441,509 0 54,052,331 0 275,836,255 0 2055 52,904,100 99,986,198 451,479,997 111,694,754 0 55,673,901 0 284,111,342 0 2056 52,9 04,100 99,986,198 465,024,397 115,045,596 0 57,344,118 0 292,634,683 0 2057 52,904,100 99,986,198 478,975,129 118,496,964 0 59,064,441 0 301,413,723 0 2058 52,904,100 99,986,198 493,344,383 122,051,873 0 60,836,375 0 310,456,135 0 2059 52,904,100 99,986 ,198 508,144,714 125,713,429 0 62,661,466 0 319,769,819 0 2060 52,904,100 99,986,198 523,389,055 129,484,832 0 64,541,310 0 329,362,913 0

PAGE 210

210 Table C 1. Continued Year Traffic Revenue (2004$) Revenue (year of $) O&M Expense Major maintenance beginning Balan ce Major maintenance expenses Major maintenance ending Balance Net cash flow available for debt service Revenue account balance 2061 52,904,100 99,986,198 539,090,727 133,369,377 0 66,477,549 0 339,243,801 0 2062 52,904,100 99,986,198 555,263,449 137,370 ,459 0 68,471,876 0 349,421,115 0 2063 52,904,100 99,986,198 571,921,352 141,491,572 0 70,526,032 0 359,903,748 0 2064 52,904,100 99,986,198 589,078,993 145,736,320 0 72,641,813 0 370,700,861 0 2065 52,904,100 99,986,198 606,751,363 150,108,409 0 74,821 ,067 0 381,821,886 0 2066 52,904,100 99,986,198 624,953,904 154,611,661 0 77,065,699 0 393,276,543 0 2067 52,904,100 99,986,198 643,702,521 159,250,011 0 79,377,670 0 405,074,839 0 2068 52,904,100 99,986,198 663,013,596 164,027,512 0 81,759,000 0 417,22 7,084 0 2069 52,904,100 99,986,198 682,904,004 168,948,337 0 84,211,770 0 429,743,897 0 2070 52,904,100 99,986,198 703,391,124 174,016,787 0 86,738,123 0 442,636,214 0 2071 52,904,100 99,986,198 724,492,858 179,237,291 0 89,340,267 0 455,915,300 0 2072 52,904,100 99,986,198 746,227,644 184,614,409 0 92,020,475 0 469,592,759 0 2073 52,904,100 99,986,198 768,614,473 190,152,842 0 94,781,089 0 483,680,542 0 2074 52,904,100 99,986,198 791,672,907 195,857,427 0 97,624,522 0 498,190,958 0 2075 52,904,100 9 9,986,198 815,423,095 201,733,150 0 100,553,258 0 513,136,687 0 2076 52,904,100 99,986,198 839,885,787 207,785,144 0 103,569,856 0 528,530,788 0 2077 52,904,100 99,986,198 865,082,361 214,018,698 0 106,676,951 0 544,386,711 0 2078 52,904,100 99,986,198 891,034,832 220,439,259 0 109,877,260 0 560,718,313 0 2079 52,904,100 99,986,198 917,765,877 227,052,437 0 113,173,578 0 577,539,862 0 2080 52,904,100 99,986,198 945,298,853 233,864,010 0 116,568,785 0 594,866,058 0 2081 52,904,100 99,986,198 973,657,81 9 240,879,931 0 120,065,848 0 612,712,040 0 2082 52,904,100 99,986,198 1,002,867,553 248,106,329 0 123,667,824 0 631,093,401 0 2083 52,904,100 99,986,198 1,032,953,580 255,549,518 0 127,377,859 0 650,026,203 0 2084 52,904,100 99,986,198 1,063,942,187 26 3,216,004 0 131,199,194 0 669,526,989 0 2085 52,904,100 99,986,198 1,095,860,453 271,112,484 0 135,135,170 0 689,612,799 0 2086 52,904,100 99,986,198 1,128,736,267 279,245,859 0 139,189,225 0 710,301,183 0 2087 52,904,100 99,986,198 1,162,598,355 287,62 3,234 0 143,364,902 0 731,610,218 0 2088 52,904,100 99,986,198 1,197,476,305 296,251,931 0 147,665,849 0 753,558,525 0

PAGE 211

211 Table C 1. Continued Year Traffic Revenue (2004$) Revenue (year of $) O&M Expense Major maintenance beginning Balance Major maintenan ce expenses Major maintenance ending Balance Net cash flow available for debt service Revenue account balance 2089 52,904,100 99,986,198 1,233,400,594 305,139,489 0 156,658,699 0 799,450,239 0 2090 52,904,100 99,986,198 1,270,402,612 314,293,674 0 16 1,358,460 0 823,433,746 0 2091 52,904,100 99,986,198 1,308,514,691 323,722,484 0 166,199,214 0 848,136,758 0 2092 52,904,100 99,986,198 1,347,770,131 333,434,159 0 171,185,191 0 873,580,861 0 2093 52,904,100 99,986,198 1,388,203,235 343,437, 184 0 156,658,699 0 799,450,239 0 Table C 2. I 495 PABs cash flow section Year Net cash flow after RURA a ccount Debt service reserve account beginning balance Total debt service on bonds RURA ending balanc e Debt service reserve account after paying senior debt Net cash flow available after senior debt service 2008 2009 2010 2011 2012 2013 33,668,175 58,900,000 22,984,968 80,000,000 58,900,000 10,683,207 2014 59,582,163 58,900,000 30,166,044 80,000,000 58,900,000 29,416,119 2015 75,383,163 58,900,000 30,166,044 80,000,000 58,900,000 45,217,119 2016 79,336,315 58,900,000 30,108,780 80,000,000 58,900,000 49,227,535 2017 81,585,033 58,900,000 30,166,044 80,000,000 58,900,000 51,418,989 2018 73,774,115 58,900,000 30,166,044 0 58,900,000 123,608,071 2019 77,377,184 58,900,000 30,166,044 0 58,900,000 47,211,140 2020 83,336,142 58,900,000 30,108,780 0 58,900,000 53,227,362 2021 87,208,665 58,900,000 30,1 66,044 0 58,900,000 57,042,621 2022 91,225,480 58,900,000 30,166,044 0 58,900,000 61,059,436 2023 95,389,795 58,900,000 30,166,044 0 58,900,000 65,223,751 2024 99,704,665 58,900,000 30,108,780 0 58,900,000 69,595,885 2025 104,171, 859 58,900,000 30,166,044 0 58,900,000 74,005,815

PAGE 212

212 Table C 2. Continued Year Net cash flow after RURA a ccount Debt service reserve account beginning balance Total debt service on bonds RURA ending balance Debt service reserve account after paying s enior debt Net cash flow available after senior debt service 2026 108,795,159 58,900,000 30,166,044 0 58,900,000 78,629,115 2027 113,576,436 58,900,000 30,166,044 0 58,900,000 83,410,392 2028 118,517,839 58,900,000 31,036,482 0 58,900,00 0 87,481,357 2029 123,621,196 58,900,000 31,764,001 0 58,900,000 91,857,195 2030 128,887,435 58,900,000 31,764,001 0 58,900,000 97,123,434 2031 134,291,406 58,900,000 31,764,001 0 58,900,000 102,527,405 2032 139,833,519 58,900,000 31,706,737 0 58,900,000 108,126,782 2033 145,515,676 58,900,000 31,764,001 0 58,900,000 113,751,675 2034 151,339,641 58,900,000 31,764,001 0 58,900,000 119,575,640 2035 157,305,562 58,900,000 31,764,001 0 58,900,000 125,541,561 2 036 162,024,729 58,900,000 31,706,737 0 58,900,000 130,317,992 2037 166,885,470 58,900,000 31,764,001 0 58,900,000 135,121,469 2038 171,892,035 58,900,000 49,195,218 0 58,900,000 122,696,817 2039 177,048,796 58,900,000 48,242,298 0 58,900,000 128,806,498 2040 182,360,259 58,900,000 53,046,779 0 58,900,000 129,313,480 2041 187,831,067 58,900,000 75,070,847 0 58,900,000 112,760,220 2042 193,465,999 58,900,000 89,960,945 0 58,900,000 103,505,054 2043 199,269,979 58,900,000 103,898,122 0 58,900,000 95,371,857 2044 205,248,079 58,900,000 99,421,401 0 58,900,000 105,826,678 2045 211,405,521 58,900,000 95,004,202 0 58,900,000 116,401,319 2046 217,747,687 58,900,000 93,422,648 0 58,900,000 12 4,325,039 2047 224,280,117 58,900,000 88,816,868 0 58,900,000 194,363,249 2048 231,008,521 0 231,008,521 2049 237,938,776 0 237,938,776 2050 245,076,940 0 245,076,940 2051 252,429,248 0 252,429,248 2052 260,002,125 0 260,002,125 2053 267,802,189 0 267,802,189

PAGE 213

213 Table C 2. Continued Year Net cash flow after RURA a ccount Debt service reserve account beginning balance Total debt service on bonds RURA ending balance Debt service reserve account after paying senior debt Net cash flow available after senior debt service 2054 275,836,255 0 275,836,255 2055 284,111,342 0 284,111,342 2056 292,634,683 0 292,634,683 2057 301,413,723 0 301,413,723 2058 310,456,135 0 310,456,135 205 9 319,769,819 0 319,769,819 2060 329,362,913 0 329,362,913 2061 339,243,801 0 339,243,801 2062 349,421,115 0 349,421,115 2063 359,903,748 0 359,903,748 2064 370,700,861 0 370,700,861 2065 381,821,886 0 381,821 ,886 2066 393,276,543 0 393,276,543 2067 405,074,839 0 405,074,839 2068 417,227,084 0 417,227,084 2069 429,743,897 0 429,743,897 2070 442,636,214 0 442,636,214 2071 455,915,300 0 455,915,300 2072 469,592,759 0 469,592,759 2073 483,680,542 0 483,680,542 2074 498,190,958 0 498,190,958 2075 513,136,687 0 513,136,687 2076 528,530,788 0 528,530,788 2077 544,386,711 0 544,386,711 2078 560,718,313 0 560,718,313 2079 577,53 9,862 0 577,539,862 2080 594,866,058 0 594,866,058 2081 612,712,040 0 612,712,040

PAGE 214

214 Table C 2. Continued Year Net cash flow after RURA a ccount Debt service reserve account beginning balance Total debt service on bonds RURA ending balan ce Debt service reserve account after paying senior debt Net cash flow available after senior debt service 2082 631,093,401 0 631,093,401 2083 650,026,203 0 650,026,203 2084 669,526,989 0 669,526,989 2085 689,612,799 0 689,612,79 9 2086 710,301,183 0 710,301,183 2087 731,610,218 0 731,610,218 2088 753,558,525 0 753,558,525 2089 776,165,280 0 776,165,280 2090 799,450,239 0 799,450,239 2091 823,433,746 0 823,433,746 2092 848,136,758 0 848,136,758 2093 873,580,861 0 873,580,861 Table C 3. I 495 TIFIA debt schedules section Year Unpaid balance on TIFIA mandatory Interest on unpaid balance TIFIA mandatory debt service Total compounded TIFIA mandatory debt service A mount actual ly paid to TIFIA mandatory Cash Flow Available after TIFIA mandatory debt service Unpaid balance on TIFIA s ched uled Interest on unpaid balance TIFIA required scheduled debt service Total r equired TIFIA scheduled debt service A mount actually paid to TIFIA scheduled 2008 2009 2010 2011 2012 2013 0 0 0 0 0 10,683,207 0 0 0 0 0 2014 0 0 0 0 0 29,416,119 0 0 0 0 0 2015 0 0 0 0 0 45,217,119 0 0 0 0 0 2016 0 0 0 0 0 49,227,535 0 0 0 0 0

PAGE 215

215 Table C 3. Continued Year Unpaid balance on TIFIA mandatory Interest on unpaid balance TIFIA mandatory debt service Total compounded TIFIA mandatory debt service A mount actually paid to TIFIA mandatory Cash Flow Available after TIFIA mandatory debt service Unpaid balance on TIFIA s ched uled Interest on unpaid balance TIFIA required scheduled debt service Total r equired TIFIA scheduled debt service A mount actually paid to TIFIA scheduled 2017 0 0 0 0 0 51,418,989 0 0 0 0 0 2018 0 0 28,990,914 28,990,914 28,990,914 94,617,157 0 0 0 0 0 2019 0 0 28,990,914 28,990,914 28,990,914 18,220,226 0 0 0 0 0 2020 0 0 28,990,914 28,990,914 28,990,914 24,236,448 0 0 0 0 0 2021 0 0 28,990,914 28,990,914 28,990,914 28,051,707 0 0 0 0 0 2022 0 0 28,990,914 28,990,914 28,990,914 32,068,522 0 0 0 0 0 2023 0 0 28,990,914 28,990,914 28,990,914 36,232,837 0 0 0 0 0 2024 0 0 28,990,914 28,990 ,914 28,990,914 40,604,971 0 0 0 0 0 2025 0 0 28,990,914 28,990,914 28,990,914 45,014,901 0 0 0 0 0 2026 0 0 28,990,914 28,990,914 28,990,914 49,638,201 0 0 0 0 0 2027 0 0 28,990,914 28,990,914 28,990,914 54,419,478 0 0 0 0 0 2028 0 0 28,990,914 28,990,914 28,990,914 58,490,443 0 0 0 0 0 2029 0 0 28,990,914 28,990,914 28,990,914 62,866,281 0 0 0 0 0 2030 0 0 28,990,914 28,990,914 28,990,914 68,132,520 0 0 0 0 0 2031 0 0 28,99 0,914 28,990,914 28,990,914 73,536,491 0 0 0 0 0 2032 0 0 28,990,914 28,990,914 28,990,914 79,135,868 0 0 0 0 0 2033 0 0 28,846,026 28,846,026 28,846,026 84,905,649 0 0 12,552,962 12,552,962 12,552,962 2034 0 0 28,432,30 7 28,432,307 28,432,307 91,143,333 0 0 0 0 0 2035 0 0 28,375,049 28,375,049 28,375,049 97,166,512 0 0 4,437,779 4,437,779 4,437,779 2036 0 0 27,832,297 27,832,297 27,832,297 102,485,695 0 0 35,025,941 35,025,941 35,025,941 2037 0 0 26,060,764 26,060,764 26,060,764 109,060,705 0 0 55,363,800 55,363,800 55,363,800 2038 0 0 23,719,059 23,719,059 23,719,059 98,977,758 0 0 44,285,541 44,285,541 44,285,541 2039 0 0 21,726,430 21,726,430 21,726,430 107 ,080,068 0 0 46,278,170 46,278,170 46,278,170 2040 0 0 19,646,234 19,646,234 19,646,234 109,667,246 0 0 48,358,366 48,358,366 48,358,366 2041 0 0 17,468,256 17,468,256 17,468,256 95,291,964 0 0 50,536,344 50,536,344 50,536,344 2042 0 0 15,194,372 15,194,372 15,194,372 88,310,682 0 0 52,810,228 52,810,228 52,810,228

PAGE 216

216 Table C 3. Continued Year Unpaid balance on TIFIA mandatory Interest on unpaid balance TIFIA mandatory debt service Total compounded TIFIA mandatory deb t service A mount actually paid to TIFIA Cash Flow Available after TIFIA mandatory debt service Unpaid balance on TIFIA s ched uled Interest on unpaid balance TIFIA required scheduled debt service Total r equired TIFIA scheduled debt service A mount actually p aid to TIFIA 2043 0 0 12,818,174 12,818,174 12,818,174 82,553,683 0 0 55,186,426 55,186,426 55,186,426 2044 0 0 10,337,148 10,337,148 10,337,148 95,489,530 0 0 57,667,452 57,667,452 57,667,452 2045 0 0 7,740,310 7,740,310 7,740 ,310 108,661,009 0 0 60,264,290 60,264,290 60,264,290 2046 0 0 5,028,717 5,028,717 5,028,717 119,296,322 0 0 62,975,883 62,975,883 62,975,883 2047 0 0 2,195,115 2,195,115 2,195,115 192,168,134 0 0 65,738,029 65,738,029 65,738, 029 2048 0 0 0 0 231,008,521 0 0 0 0 2049 0 0 0 0 237,938,776 0 0 0 0 2050 0 0 0 0 245,076,940 0 0 0 0 2051 0 0 0 0 252,429,248 0 0 0 0 2052 0 0 0 0 260,002,125 0 0 0 0 2053 0 0 0 0 267,80 2,189 0 0 0 0 2054 0 0 0 0 275,836,255 0 0 0 0 2055 0 0 0 0 284,111,342 0 0 0 0 2056 0 0 0 0 292,634,683 0 0 0 0 2057 0 0 0 0 301,413,723 0 0 0 0 2058 0 0 0 0 310,456,135 0 0 0 0 2059 0 0 0 0 319,769,819 0 0 0 0 2060 0 0 0 0 329,362,913 0 0 0 0 2061 0 0 0 0 339,243,801 0 0 0 0 2062 0 0 0 0 349,421,115 0 0 0 0 2063 0 0 0 0 359,903,748 0 0 0 0 2064 0 0 0 0 370,700,861 0 0 0 0 2065 0 0 0 0 381,821,886 0 0 0 0 2066 0 0 0 0 393,276,543 0 0 0 0 2067 0 0 0 0 405,074,839 0 0 0 0

PAGE 217

217 Table C 3. Continued Year Unpaid balance on TIFIA mandatory Interest on unpaid balance TIFIA mandatory debt service Total compounded TIFIA mandatory debt service A mount actually paid to TIFIA mandatory Cash Flow Available after TIFIA mandatory debt service Unpaid balance on TIFIA s ched uled Interest on unpaid balance TIFIA required scheduled debt service Total r equired TIFIA scheduled debt service A mount actually paid to TIFIA scheduled 2068 0 0 0 0 417,227,084 0 0 0 0 2069 0 0 0 0 429,743,897 0 0 0 0 2070 0 0 0 0 442,636,214 0 0 0 0 2071 0 0 0 0 455,915,300 0 0 0 0 2072 0 0 0 0 469,592,759 0 0 0 0 2073 0 0 0 0 483,680,542 0 0 0 0 2074 0 0 0 0 498,190,958 0 0 0 0 2075 0 0 0 0 513,136,687 0 0 0 0 2076 0 0 0 0 528,530,788 0 0 0 0 2077 0 0 0 0 544,386,711 0 0 0 0 2078 0 0 0 0 560,718,313 0 0 0 0 2079 0 0 0 0 577,539,862 0 0 0 0 2080 0 0 0 0 594,866,058 0 0 0 0 2081 0 0 0 0 612,712,040 0 0 0 0 2082 0 0 0 0 631,093,401 0 0 0 0 2083 0 0 0 0 650,0 26,203 0 0 0 0 2084 0 0 0 0 669,526,989 0 0 0 0 2085 0 0 0 0 689,612,799 0 0 0 0 2086 0 0 0 0 710,301,183 0 0 0 0 2087 0 0 0 0 731,610,218 0 0 0 0 2088 0 0 0 0 753,558,525 0 0 0 0 2089 0 0 0 0 776,165,280 0 0 0 0 2090 0 0 0 0 799,450,239 0 0 0 0 2091 0 0 0 0 823,433,746 0 0 0 0 2092 0 0 0 0 848,136,758 0 0 0 0 2093 0 0 0 0 873,580,861 0 0 0 0

PAGE 218

218 Table C 4. I 495 equity cash flow sect ion Year cash flows available after TIFIA scheduled debt service Permit fee Cash flows to equity Equity IRR Total DSCR r equired = 1.05 Senior debt service coverage ratio required = 1.45 2008 (348,700,000) 2009 0 2010 0 2011 0 2012 0 2013 10,683,207 0 0 1.46 4.95 2014 29,416,119 0 29,416,119 33.77% 1.98 4.63 2015 45,217,119 0 45,217,119 20.74% 2.50 2.50 2016 49,227,535 0 49,227,535 13.38% 2.63 2.63 2017 51,418,989 0 51,418,989 8.49% 2.70 2 .70 2018 94,617,157 0 94,617,157 2.96% 1.25 2.45 2019 18,220,226 0 18,220,226 2.17% 1.31 2.57 2020 24,236,448 0 24,236,448 1.22% 1.41 2.77 2021 28,051,707 0 28,051,707 0.26% 1.47 2.89 2022 32,068,522 0 32,068,522 0.69% 1.54 3.0 2 2023 36,232,837 0 36,232,837 1.59% 1.61 3.16 2024 40,604,971 0 40,604,971 2.43% 1.69 3.31 2025 45,014,901 0 45,014,901 3.20% 1.76 3.45 2026 49,638,201 0 49,638,201 3.91% 1.84 3.61 2027 54,419,478 0 54,419,478 4.55% 1.92 3.77 202 8 58,490,443 0 58,490,443 5.12% 1.97 3.82 2029 62,866,281 0 62,866,281 5.63% 2.03 3.89 2030 68,132,520 0 68,132,520 6.09% 2.12 4.06 2031 73,536,491 0 73,536,491 6.51% 2.21 4.23 2032 79,135,868 0 79,135,868 6.89% 2.30 4.41 2033 72,3 52,687 0 72,352,687 7.18% 1.99 4.58 2034 91,143,333 0 91,143,333 7.49% 2.51 4.76

PAGE 219

219 Table C 4. Continued Year cash flows available after TIFIA scheduled debt service Permit fee Cash flows to equity Equity IRR Total DSCR r equired = 1.05 Senior debt service coverage ratio required = 1.45 2035 92,728,733 0 92,728,733 7.76% 2.44 4.95 2036 67,459,754 0 67,459,754 7.93% 1.71 5.11 2037 53,696,905 0 53,696,905 8.04% 1.47 5.25 2038 54,692,217 0 54,692,217 8.15% 1.47 3.49 2039 60,801,898 0 60,801,898 8.26% 1.52 3.67 2040 61,308,880 0 61,308,880 8.35% 1.51 3.44 2041 44,755,620 0 44,755,620 8.41% 1.31 2.50 2042 35,500,454 0 35,500,454 8.46% 1.22 2.15 2043 27,367,257 0 27,367,257 8.49% 1.16 1.92 2044 37,822,078 0 37 ,822,078 8.53% 1.23 2.06 2045 48,396,719 0 48,396,719 8.57% 1.30 2.23 2046 56,320,439 0 56,320,439 8.62% 1.35 2.33 2047 126,430,105 0 126,430,105 8.71% 1.43 2.53 2048 231,008,521 0 231,008,521 8.86% 100.00 100.00 2049 237,938,776 0 237,938,776 8.99% 100.00 100.00 2050 245,076,940 0 245,076,940 9.10% 100.00 100.00 2051 252,429,248 0 252,429,248 9.20% 100.00 100.00 2052 260,002,125 0 260,002,125 9.29% 100.00 100.00 2053 267,802,189 0 267,802,189 9.38% 100.00 100. 00 2054 275,836,255 0 275,836,255 9.45% 100.00 100.00 2055 284,111,342 0 284,111,342 9.52% 100.00 100.00 2056 292,634,683 0 292,634,683 9.58% 100.00 100.00 2057 301,413,723 0 301,413,723 9.63% 100.00 100.00 2058 310,456,135 0 310,456 ,135 9.68% 100.00 100.00 2059 319,769,819 0 319,769,819 9.72% 100.00 100.00 2060 329,362,913 0 329,362,913 9.76% 100.00 100.00 2061 339,243,801 0 339,243,801 9.80% 100.00 100.00

PAGE 220

220 Table C 4. Continued Year cash flows available after TIFIA scheduled debt service Permit fee Cash flows to equity Equity IRR Total DSCR r equired = 1.05 Senior debt service coverage ratio required = 1.45 2062 349,421,115 0 349,421,115 9.84% 100.00 100.00 2063 359,903,748 0 359,903,748 9.87% 100.00 100.00 2064 370,700,861 0 370,700,861 9.90% 100.00 100.00 2065 381,821,886 0 381,821,886 9.92% 100.00 100.00 2066 393,276,543 0 393,276,543 9.95% 100.00 100.00 2067 405,074,839 0 405,074,839 9.97% 100.00 100.00 2068 417,227,084 0 417,227,08 4 9.99% 100.00 100.00 2069 429,743,897 0 429,743,897 10.01% 100.00 100.00 2070 442,636,214 0 442,636,214 10.03% 100.00 100.00 2071 455,915,300 0 455,915,300 10.04% 100.00 100.00 2072 469,592,759 0 469,592,759 10.06% 100.00 100.00 207 3 483,680,542 0 483,680,542 10.07% 100.00 100.00 2074 498,190,958 0 498,190,958 10.08% 100.00 100.00 2075 513,136,687 0 513,136,687 10.10% 100.00 100.00 2076 528,530,788 0 528,530,788 10.11% 100.00 100.00 2077 544,386,711 0 544,386,71 1 10.12% 100.00 100.00 2078 560,718,313 0 560,718,313 10.13% 100.00 100.00 2079 577,539,862 0 577,539,862 10.13% 100.00 100.00 2080 594,866,058 0 594,866,058 10.14% 100.00 100.00 2081 612,712,040 0 612,712,040 10.15% 100.00 100.00 20 82 631,093,401 0 631,093,401 10.16% 100.00 100.00 2083 650,026,203 0 650,026,203 10.16% 100.00 100.00 2084 669,526,989 0 669,526,989 10.17% 100.00 100.00 2085 689,612,799 0 689,612,799 10.17% 100.00 100.00 2086 710,301,183 0 710,301,1 83 10.18% 100.00 100.00 2087 731,610,218 0 731,610,218 10.18% 100.00 100.00 2088 753,558,525 0 753,558,525 10.19% 100.00 100.00

PAGE 221

221 Table C 4. Continued Year cash flows available after TIFIA scheduled debt service Permit fee Cash flows to equity Equity IRR Total DSCR r equired = 1.05 Senior debt service coverage ratio required = 1.45 2089 776,165,280 0 776,165,280 10.19% 100.00 100.00 2090 799,450,239 0 799,450,239 10.20% 100.00 100.00 2091 823,433,746 0 823,433,746 10.20% 100.00 100 .00 2092 848,136,758 0 848,136,758 10.20% 100.00 100.00 2093 873,580,861 0 873,580,861 10.21% 100.00 100.00

PAGE 222

222 Table C 5. NTE traffic & revenue cash flow section Year Traffic Revenue (2004$) Revenue (year of $) O&M e xpense Major maintenance beg inning Balance Major maintenance expenses Major maintenance ending Balance Net cash flow available for Debt Service Revenue account balance 2010 2011 2012 2013 2014 2015 12,954,280 33,056,696 39,294,024 9, 601,216 20,000,000 0 20,000,000 29,692,808 0 2016 22,237,619 58,719,463 71,543,964 15,694,541 20,000,000 7,875,610 12,124,390 55,849,423 1,818,808 2017 27,071,210 79,195,765 98,904,658 19,210,752 12,124,390 9,827,147 2,297,244 79,693,906 29,794,231 2018 30,394,367 97,280,003 124,526,629 21,892,399 2,297,244 11,309,326 0 93,622,146 81,614,137 2019 28,747,111 87,851,207 115,268,397 21,321,228 0 10,963,815 0 82,983,354 147,362,283 2020 29,314,619 90,991,646 122,373,648 22,249,228 0 11,459,763 0 88,664,658 202,471,637 2021 29,667,458 93,000,508 128,202,228 23,057,164 0 11,887,638 0 93,257,426 0 2022 30,021,308 95,059,668 134,316,823 23,892,330 0 12,330,160 0 98,094,333 0 2023 30,375,832 97,168,474 140,728,923 24,755,349 0 12,787,662 0 103,185,912 0 2024 30,731,367 99,330,277 147,456,354 25,647,363 0 13,260,769 0 108,548,223 0 2025 31,346,729 103,186,214 157,010,028 26,773,100 0 13,864,459 0 116,372,469 0 2026 31,453,221 103,868,540 161,999,475 27,528,379 0 14,259,348 0 120,211,747 0 2027 31,605,208 10 4,850,189 167,618,774 28,342,326 0 14,686,458 0 124,589,990 0 2028 31,760,902 105,865,401 173,472,786 29,182,908 0 15,127,777 0 129,162,101 0 2029 31,920,303 106,914,969 179,572,441 30,051,028 0 15,583,793 0 133,937,620 0 2030 32,017,696 107,561,362 185 ,174,565 30,889,071 0 16,022,124 0 138,263,369 0 2031 34,122,598 122,527,337 216,213,047 33,583,443 0 17,502,334 0 165,127,270 0 2032 34,538,456 125,721,817 227,396,314 34,812,274 0 18,158,529 0 174,425,511 0 2033 34,950,607 128,969,991 239,103,153 36,0 78,000 0 18,834,597 0 184,190,556 0 2034 35,377,586 132,423,555 251,643,515 37,399,837 0 19,541,310 0 194,702,368 0 2035 35,804,565 135,969,599 264,841,587 38,765,217 0 20,271,587 0 205,804,783 0 2036 36,231,881 139,613,510 278,737,678 40,175,839 0 21,0 26,360 0 217,535,480 0 2037 36,658,860 143,352,087 293,356,771 41,632,407 0 21,806,001 0 229,918,363 0

PAGE 223

223 Table C 5. Continued Year Traffic Revenue (2004$) Revenue (year of $) O&M e xpense Major maintenance beginning Balance Major maintenance expenses Major maintenance ending Balance Net cash flow available for Debt Service Revenue account balance 2038 37,086,176 147,193,846 308,749,038 43,137,059 0 22,611,689 0 243,000,290 0 2039 37,497,990 150,993,625 324,637,316 44,673,676 0 23,434,344 0 256,529,296 0 2040 37,925,306 155,040,173 341,670,856 46,277,843 0 24,293,929 0 271,099,084 0 2041 38,352,285 159,191,845 359,590,637 47,933,902 0 25,181,627 0 286,475,108 0 2042 38,772,524 163,386,522 378,292,432 49,635,765 0 26,093,989 0 302,562,677 0 2043 39,178,2 72 167,541,402 397,610,123 51,374,966 0 27,026,235 0 319,208,922 0 2044 39,596,826 171,938,156 418,245,638 53,186,226 0 27,997,839 0 337,061,573 0 2045 40,012,347 176,417,176 439,869,535 55,052,025 0 28,998,933 0 355,818,577 0 2046 40,438,652 181,133,72 7 462,920,280 56,992,135 0 30,040,595 0 375,887,550 0 2047 40,798,231 185,209,936 485,171,221 58,904,389 0 31,065,408 0 395,201,425 0 2048 41,225,547 190,173,462 510,627,887 60,970,756 0 32,175,553 0 417,481,578 0 2049 41,650,504 195,241,501 537,341,792 63,100,266 0 33,319,902 0 440,921,623 0 2050 42,077,483 200,469,687 565,524,023 65,301,097 0 34,503,018 0 465,719,908 0 2051 42,501,766 205,803,530 595,085,040 67,568,498 0 35,722,198 0 491,794,344 0 2052 42,928,745 211,314,546 626,295,759 69,912,590 0 36,983,095 0 519,400,074 0 2053 43,300,456 216,232,238 656,892,626 72,244,771 0 38,235,907 0 546,411,949 0 2054 43,727,435 222,022,515 691,344,999 74,738,923 0 39,578,269 0 577,027,807 0 2055 44,095,776 227,142,022 724,968,539 77,215,779 0 40,909,450 0 606,843,310 0 2056 44,571,620 233,930,840 765,302,301 79,951,766 0 42,384,683 0 642,965,852 0 2057 44,891,433 238,607,224 800,116,068 82,505,532 0 43,756,024 0 673,854,512 0 2058 45,405,358 246,318,563 846,623,682 85,482,279 0 45,363,374 0 715,778,028 0 2059 45,669,903 250,384,665 882,114,299 88,101,642 0 46,768,367 0 747,244,291 0 2060 46,211,462 258,919,073 934,985,863 91,316,209 0 48,506,025 0 795,163,628 0 2061 23,265,301 264,083,967 977,477,796 49,646,966 0 25,031,018 0 902,799,812 0

PAGE 224

224 Table C 6. NTE PABs cash flow section Year Net cash flow after RURA a ccount Debt service reserve account beginning balance Principle debt service on bonds Interest debts service on bonds Total debt service on bonds 2010 2011 0 28,958,000 28,958,000 2012 0 27,874,000 27,874,000 2013 0 27,874,000 27,874,000 2014 0 27,874,000 27,874,000 2015 29,692,808 40,000,000 0 27,874,000 27,874,000 2016 57,668,231 40,000,000 0 27,874,000 27,874,000 2017 109,488,137 40,000,000 0 27,874,000 27,874,000 2018 17 5,236,283 40,000,000 0 27,874,000 27,874,000 2019 230,345,637 40,000,000 0 27,874,000 27,874,000 2020 291,136,295 40,000,000 0 27,874,000 27,874,000 2021 93,257,426 33,139,584 0 27,874,000 27,874,000 2022 98,094,333 36,005,297 0 27,874,000 27,874,000 2023 103,185,912 37,426,377 0 27,874,000 27,874,000 2024 108,548,223 37,426,377 0 27,874,000 27,874,000 2025 116,372,469 37,426,377 0 27,874,000 27,874,000 2026 120,211,747 37,426,377 0 27,874,000 27,874,000 2027 124,589,990 37,426,377 0 27,874,000 27, 874,000 2028 129,162,101 37,426,377 0 27,874,000 27,874,000 2029 133,937,620 37,426,377 0 27,874,000 27,874,000 2030 138,263,369 66,246,377 0 27,874,000 27,874,000 2031 165,127,270 66,249,377 28,820,000 27,874,000 56,694,000 2032 174,425,511 66,250,37 7 30,985,000 25,712,000 56,697,000 2033 184,190,556 66,245,377 33,310,000 23,388,000 56,698,000 2034 194,702,368 66,248,377 35,595,000 21,098,000 56,693,000 2035 205,804,783 66,248,377 38,045,000 18,651,000 56,696,000 2036 217,535,480 66,165,102 40,660 ,000 16,036,000 56,696,000 2037 229,918,363 66,058,527 43,455,000 13,240,000 56,695,000

PAGE 225

225 Table C 6. Continued Year Net cash flow after RURA a ccount Debt service reserve account beginning balance Principle debt service on bonds Interest debts service on b onds Total debt service on bonds 2038 243,000,290 65,946,027 46,445,000 10,253,000 56,698,000 2039 256,529,296 65,838,527 49,635,000 7,060,000 56,695,000 2040 271,099,084 53,218,324 53,050,000 3,647,000 56,697,000 2041 286,475,108 97,451,266 2042 302,562,677 97,451,008 2043 319,208,922 97,450,740 2044 337,061,573 97,452,677 2045 355,818,577 97,450,166 2046 375,887,550 97,449,860 2047 395,201,425 97,449,540 2048 417,481,578 97,451,844 2049 440,921,623 97,448,858 2050 465,719,908 48,719,955 2051 491,794,344 0 2052 519,400,074 0 2053 546,411,949 0 2054 577,027,807 0 2055 606,843,310 0 2056 642,965,852 0 2057 673,854,512 0 2058 715,778,028 0 2059 747,244,291 0 2060 795,163,628 0 2061 902,799,812 0

PAGE 226

226 Table C 7. NTE TIFIA mandatory debt cash flow section Year Debt service reserve account after paying senior debt Net cash flow available after senior debt service TIFIA mandatory debt service Total c ompounded TIFIA mandatory debt service Amount actually paid to TIFIA Mandatory Cash flow available after TIFIA mandatory debt service 2010 2011 2012 2013 2014 2015 40,000,000 1,818,808 0 0 0 1,818,808 2016 40,000,000 29,794,231 0 0 0 29,794,231 2017 40,000,000 81,614,137 0 0 0 81,614,137 2018 40,000,000 147,362,283 0 0 0 147,362,283 2019 40,000,000 202,471,637 0 0 0 202,471,637 2020 40,000,000 263,262,295 1,920,915 1,920,915 1,920,915 261,341,380 2021 33,139,584 65,383,426 5,265,584 5,265,584 5,2 65,584 60,117,842 2022 36,005,297 70,220,333 8,131,297 8,131,297 8,131,297 62,089,036 2023 37,426,377 75,311,912 9,552,377 9,552,377 9,552,377 65,759,535 2024 37,426,377 80,674,223 9,552,377 9,552,377 9,552,377 71,121,846 2025 37,426,377 88,498,469 9,5 52,377 9,552,377 9,552,377 78,946,092 2026 37,426,377 92,337,747 9,552,377 9,552,377 9,552,377 82,785,370 2027 37,426,377 96,715,990 9,552,377 9,552,377 9,552,377 87,163,613 2028 37,426,377 101,288,101 9,552,377 9,552,377 9,552,377 91,735,724 2029 37,4 26,377 106,063,620 9,552,377 9,552,377 9,552,377 96,511,243 2030 66,246,377 81,569,369 9,552,377 9,552,377 9,552,377 72,016,992 2031 66,249,377 108,430,270 9,552,377 9,552,377 9,552,377 98,877,893 2032 66,250,377 117,727,511 9,552,377 9,552,377 9,552,37 7 108,175,134 2033 66,245,377 127,497,556 9,552,377 9,552,377 9,552,377 117,945,179 2034 66,248,377 138,006,368 9,552,377 9,552,377 9,552,377 128,453,991 2035 66,248,377 149,108,783 9,552,377 9,552,377 9,552,377 139,556,406 2036 66,165,102 160,840,480 9,470,102 9,470,102 9,470,102 151,370,378

PAGE 227

227 Table C 7. Continued Year Debt service reserve account after paying senior debt Net cash flow available after senior debt service TIFIA mandatory debt service Total c ompounded TIFIA mandatory debt service Amount actually paid to TIFIA Mandatory Cash flow available after TIFIA mandatory debt service 2037 66,058,527 173,220,363 9,360,527 9,360,527 9,360,527 163,859,836 2038 65,946,027 186,305,290 9,251,027 9,251,027 9,251,027 177,054,263 2039 65,838,527 199,832,2 96 9,141,527 9,141,527 9,141,527 190,690,769 2040 53,218,324 271,099,084 53,218,324 53,218,324 53,218,324 217,880,760 2041 97,451,266 286,475,108 97,451,266 97,451,266 97,451,266 189,023,842 2042 97,451,008 302,562,677 97,451,008 97,451,008 97,451,008 2 05,111,669 2043 97,450,740 319,208,922 97,450,740 97,450,740 97,450,740 221,758,182 2044 97,452,677 337,061,573 97,452,677 97,452,677 97,452,677 239,608,896 2045 97,450,166 355,818,577 97,450,166 97,450,166 97,450,166 258,368,411 2046 97,449,860 375,88 7,550 97,449,860 97,449,860 97,449,860 278,437,690 2047 97,449,540 395,201,425 97,449,540 97,449,540 97,449,540 297,751,885 2048 97,451,844 417,481,578 97,451,844 97,451,844 97,451,844 320,029,734 2049 97,448,858 440,921,623 97,448,858 97,448,858 97,448 ,858 343,472,765 2050 48,719,955 465,719,908 48,719,955 48,719,955 48,719,955 416,999,953 2051 0 491,794,344 0 0 0 491,794,344 2052 0 519,400,074 0 0 0 519,400,074 2053 0 546,411,949 0 0 0 546,411,949 2054 0 577,027,807 0 0 0 577,027,807 2055 0 606,8 43,310 0 0 0 606,843,310 2056 0 642,965,852 0 0 0 642,965,852 2057 0 673,854,512 0 0 0 673,854,512 2058 0 715,778,028 0 0 0 715,778,028 2059 0 747,244,291 0 0 0 747,244,291 2060 0 795,163,628 0 0 0 795,163,628 2061 0 902,799,812 0 0 0 902,799,812

PAGE 228

228 Table C 8. NTE TIFIA scheduled debt cash flow section Year Debt service reserve account required balance Debt service reserve ending balance Cash flow available after RURA Unpaid balance on TIFIA s cheduled Interest on unpaid balance TIFIA required schedul ed debt service Total r equired TIFIA scheduled debt service A mount actually paid to TIFIA 2010 2011 2012 2013 2014 2015 2016 40,000,000 40,000,000 1,818,808 0 0 0 0 0 2017 40,000,000 40,000,000 29,794 ,231 0 0 0 0 0 2018 40,000,000 40,000,000 81,614,137 0 0 0 0 0 2019 40,000,000 40,000,000 147,362,283 0 0 0 0 0 2020 40,000,000 40,000,000 202,471,637 0 0 0 0 0 2021 33,139,584 33,139,584 268,201,796 0 0 17,228,236 17,228,236 17,228,236 2022 36,005,29 7 36,005,297 57,252,129 0 0 32,943,923 32,943,923 32,943,923 2023 37,426,377 37,426,377 60,667,956 0 0 30,078,210 30,078,210 30,078,210 2024 37,426,377 37,426,377 65,759,535 0 0 28,657,130 28,657,130 28,657,130 2025 37,426,377 37,426,377 71,121,846 0 0 28,657,130 28,657,130 28,657,130 2026 37,426,377 37,426,377 78,946,092 0 0 28,657,130 28,657,130 28,657,130 2027 37,426,377 37,426,377 82,785,370 0 0 28,657,130 28,657,130 28,657,130 2028 37,426,377 37,426,377 87,163,613 0 0 28,657,130 28,657,130 28,657 ,130 2029 37,426,377 37,426,377 91,735,724 0 0 28,657,130 28,657,130 28,657,130 2030 66,246,377 66,246,377 67,691,243 0 0 28,657,130 28,657,130 28,657,130 2031 66,249,377 66,249,377 72,013,992 0 0 28,657,130 28,657,130 28,657,130 2032 66,250,377 66,250 ,377 98,876,893 0 0 28,657,130 28,657,130 28,657,130 2033 66,245,377 66,245,377 108,180,134 0 0 28,657,130 28,657,130 28,657,130 2034 66,248,377 66,248,377 117,942,179 0 0 28,657,130 28,657,130 28,657,130 2035 66,248,377 66,248,377 128,453,991 0 0 28,65 7,130 28,657,130 28,657,130

PAGE 229

229 Table C 8. Continued Year Debt service reserve account required balance Debt service reserve ending balance Cash flow available after RURA Unpaid balance on TIFIA s cheduled Interest on unpaid balance TIFIA required scheduled debt service Total r equired TIFIA scheduled debt service A mount actually paid to TIFIA 2036 66,165,102 66,165,102 139,639,681 0 0 28,657,130 28,657,130 28,657,130 2037 66,058,527 66,058,527 151,476,953 0 0 38,410,307 38,410,307 38,410,307 2038 65,946,02 7 65,946,027 163,972,336 0 0 38,081,580 38,081,580 38,081,580 2039 65,838,527 65,838,527 177,161,763 0 0 37,753,080 37,753,080 37,753,080 2040 53,218,324 53,218,324 203,310,972 0 0 57,424,580 57,424,580 57,424,580 2041 97,451,266 97,451,266 173,647,818 0 0 38,188,476 38,188,476 38,188,476 2042 97,451,008 97,451,008 189,024,100 0 0 0 0 0 2043 97,450,740 97,450,740 205,111,937 0 0 0 0 0 2044 97,452,677 97,452,677 221,756,245 0 0 0 0 0 2045 97,450,166 97,450,166 239,611,407 0 0 0 0 0 2046 97,449,860 97 ,449,860 258,368,717 0 0 0 0 0 2047 97,449,540 97,449,540 278,438,010 0 0 0 0 0 2048 97,451,844 97,451,844 297,749,581 0 0 0 0 0 2049 97,448,858 97,448,858 320,032,720 0 0 0 0 0 2050 48,719,955 48,719,955 392,201,668 0 0 0 0 0 2051 0 0 465,719,908 0 0 0 0 0 2052 0 0 491,794,344 0 0 0 0 0 2053 0 0 519,400,074 0 0 0 0 0 2054 0 0 546,411,949 0 0 0 0 0 2055 0 0 577,027,807 0 0 0 0 0 2056 0 0 606,843,310 0 0 0 0 0 2057 0 0 642,965,852 0 0 0 0 0 2058 0 0 673,854,512 0 0 0 0 0 2059 0 0 715,778,028 0 0 0 0 0 2060 0 0 747,244,291 0 0 0 0 0 2061 0 0 795,163,628 0 0 0 0 0

PAGE 230

230 Table C 9. NTE equity cash flow section Year Cash flows available after TIFIA scheduled debt service Permit f ee Cash flow after permit fee Cash f lows Equity IRR DSCR Required = 1.2 0.00 2010 (428,802,000) 0.00 2011 0 0.00 2012 0 0.00 2013 0 0.00 2014 0 1.89 2015 1,818,808 1,611,881 206,927 0 1.41 2016 29,794,231 0 29,794,231 0 1.48 2017 81,614,137 0 81,614,137 0 1.56 2018 147,362,283 0 147,362,283 0 1.64 2019 202,471,637 0 202,471,637 0 1.76 2020 250,973,560 0 250,973,560 250,973,560 5.22% 1.82 2021 24,308,206 0 24,308,206 24,308,206 4.30% 1.89 2022 30,589,746 0 30,589,746 30,589,746 3.23% 1.95 2023 37,102,405 0 37,102,405 37,102,405 2.0 9% 2.03 2024 42,464,716 0 42,464,716 42,464,716 0.97% 1.46 2025 50,288,962 0 50,288,962 50,288,962 0.14% 1.74 2026 54,128,240 0 54,128,240 54,128,240 1.13% 1.84 2027 58,506,483 0 58,506,483 58,506,483 2.00% 1.94 2028 63,078,594 0 63,078,594 63,078,59 4 2.79% 2.05 2029 39,034,113 0 39,034,113 39,034,113 3.20% 2.17 2030 43,356,862 0 43,356,862 43,356,862 3.60% 2.08 2031 70,219,763 0 70,219,763 70,219,763 4.17% 2.21 2032 79,523,004 0 79,523,004 79,523,004 4.70% 2.34 2033 89,285,049 0 89,285,049 89,28 5,049 5.19% 2.08 2034 99,796,861 0 99,796,861 99,796,861 5.65% 2.97 2035 110,982,551 0 110,982,551 110,982,551 6.07% 2.94 2036 113,066,646 0 113,066,646 113,066,646 6.43% 3.10

PAGE 231

231 Table C 9. Continued Year Cash flows available after TIFIA scheduled debt s ervice Permit f ee Cash flow after permit fee Cash f lows Equity IRR DSCR Required = 1.2 2037 125,890,756 0 125,890,756 125,890,756 6.77% 3.28 2038 139,408,683 0 139,408,683 139,408,683 7.09% 3.46 2039 145,886,392 0 145,886,392 145,886,392 7.37% 3.65 204 0 135,459,342 0 135,459,342 135,459,342 7.59% 3.86 2041 189,024,100 0 189,024,100 189,024,100 7.85% 4.06 2042 205,111,937 0 205,111,937 205,111,937 8.09% 4.28 2043 221,756,245 0 221,756,245 221,756,245 8.31% 4.52 2044 239,611,407 0 239,611,407 239,611, 407 8.51% 9.56 2045 258,368,717 0 258,368,717 258,368,717 8.69% 100.00 2046 278,438,010 0 278,438,010 278,438,010 8.86% 100.00 2047 297,749,581 0 297,749,581 297,749,581 9.02% 100.00 2048 320,032,720 0 320,032,720 320,032,720 9.16% 100.00 2049 392,201 ,668 0 392,201,668 392,201,668 9.31% 100.00 2050 465,719,908 0 465,719,908 465,719,908 9.46% 100.00 2051 491,794,344 0 491,794,344 491,794,344 9.60% 100.00 2052 519,400,074 0 519,400,074 519,400,074 9.72% 100.00 2053 546,411,949 0 546,411,949 546,411,9 49 9.83% 100.00 2054 577,027,807 0 577,027,807 577,027,807 9.93% 100.00 2055 606,843,310 0 606,843,310 606,843,310 10.03% 100.00 2056 642,965,852 0 642,965,852 642,965,852 10.11% 100.00 2057 673,854,512 0 673,854,512 673,854,512 10.19% 100.00 2058 715 ,778,028 0 715,778,028 715,778,028 10.26% 100.00 2059 747,244,291 0 747,244,291 747,244,291 10.32% 100.00 2060 795,163,628 0 795,163,628 795,163,628 10.38% 100.00 2061 902,799,812 0 902,799,812 902,799,812 10.44%

PAGE 232

232 Table C 10. LBJ traffic & revenue c ash flow section Year Traffic Revenue (2008$) Revenue (year of $) O&M e xpense Major maintenance beginning Balance Deposits into major maintenance Reserve Account Major maintenance ending Balance Net cash flow available for debt service Revenue account bal ance 2010 2011 2012 2013 2014 2015 2016 71,200,000 74,102,525 90,286,731 20,429,417 20,000,000 0 20,000,000 69,857,314 0 2017 78,700,000 107,300,533 134,003,662 24,287,202 20,000,000 4,568,648 1 5,431,352 109,716,461 26,099,314 2018 79,800,000 113,287,323 145,017,351 25,441,712 15,431,352 4,748,317 10,683,036 119,575,639 92,057,775 2019 76,100,000 94,377,544 123,831,516 24,238,219 10,683,036 4,641,361 6,041,675 99,593,297 167,875,414 2020 76,80 0,000 97,695,321 131,389,346 25,190,397 6,041,675 4,801,155 1,240,520 106,198,949 223,710,711 2021 77,500,000 101,129,733 139,408,454 26,179,981 1,240,520 4,966,039 0 109,502,955 273,107,056 2022 78,200,000 104,684,879 147,916,993 27,208,439 0 5,136,165 0 115,572,388 294,949,300 2023 78,900,000 108,365,004 156,944,836 28,277,300 0 5,311,695 0 123,355,841 310,365,337 2024 79,600,000 112,174,501 166,523,677 29,388,151 0 5,492,791 0 131,642,735 0 2025 77,600,000 101,630,121 154,642,249 28,954,990 0 5,488, 650 0 120,198,608 0 2026 78,400,000 105,723,401 164,892,424 30,152,017 0 5,683,865 0 129,056,542 0 2027 79,200,000 109,981,542 175,822,013 31,398,530 0 5,885,410 0 138,538,072 0 2028 80,000,000 114,411,185 187,476,049 32,696,575 0 6,093,481 0 148,685,99 4 0 2029 80,800,000 119,019,238 199,902,552 34,048,282 0 6,308,276 0 159,545,993 0 2030 79,600,000 112,174,501 193,116,412 34,081,245 0 6,369,953 0 152,665,214 0 2031 82,400,000 128,799,603 227,281,155 36,921,650 0 6,758,872 0 183,600,633 0 2032 83,200 ,000 133,987,167 242,346,066 38,448,026 0 6,995,105 0 196,902,934 0 2033 84,000,000 139,383,666 258,409,526 40,037,504 0 7,238,925 0 211,133,097 0 2034 84,900,000 145,714,963 276,901,080 41,775,204 0 7,499,397 0 227,626,479 0 2035 85,700,000 151,583,814 295,254,955 43,502,231 0 7,759,314 0 243,993,410 0 2036 86,500,000 157,689,040 314,825,383 45,300,655 0 8,027,540 0 261,497,188 0 2037 87,400,000 164,851,830 337,354,005 47,266,785 0 8,313,840 0 281,773,380 0

PAGE 233

233 Table C 10. Continued Year Traffic Revenu e (2008$) Revenue (year of $) O&M e xpense Major maintenance beginning Balance Deposits into major maintenance Reserve Account Major maintenance ending Balance Net cash flow available for debt service Revenue account balance 2038 88,200,000 171,491,443 359 ,714,890 49,220,839 0 8,599,688 0 301,894,364 0 2039 89,100,000 179,281,187 385,455,765 51,357,112 0 8,904,626 0 325,194,028 0 2040 90,000,000 187,424,769 413,038,634 53,586,103 0 9,219,436 0 350,233,096 0 2041 90,900,000 195,938,261 442,595,309 55,911, 837 0 9,544,421 0 377,139,052 0 2042 91,700,000 203,829,917 471,931,919 58,223,285 0 9,869,130 0 403,839,503 0 2043 92,600,000 213,088,589 505,702,944 60,750,281 0 10,215,142 0 434,737,521 0 2044 93,500,000 222,767,822 541,890,594 63,386,953 0 10,572,28 6 0 467,931,355 0 2045 94,400,000 232,886,719 580,667,799 66,138,061 0 10,940,902 0 503,588,836 0 2046 95,200,000 242,266,521 619,156,288 68,872,271 0 11,309,462 0 538,974,555 0 2047 96,100,000 253,271,118 663,462,557 71,861,452 0 11,701,789 0 579,899,3 16 0 2048 97,000,000 264,775,583 710,939,343 74,980,368 0 12,106,664 0 623,852,311 0 2049 97,900,000 276,802,621 761,813,525 78,234,652 0 12,524,468 0 671,054,405 0 2050 98,700,000 287,951,190 812,308,924 81,468,946 0 12,942,484 0 717,897,493 0 2051 99 ,600,000 301,030,946 870,437,022 85,004,846 0 13,387,013 0 772,045,164 0 2052 100,500,000 314,704,829 932,724,715 88,694,209 0 13,845,679 0 830,184,827 0 2053 101,300,000 327,379,957 994,548,645 92,360,912 0 14,304,791 0 887,882,942 0 2054 102,200,000 3 42,250,706 1,065,717,655 96,369,542 0 14,792,679 0 954,555,433 0 2055 103,100,000 357,796,936 1,141,979,455 100,552,154 0 15,296,021 0 1,026,131,280 0 2056 104,000,000 374,049,331 1,223,698,482 104,916,299 0 15,815,285 0 1,102,966,898 0 2057 104,800,000 389,114,632 1,304,809,069 109,253,639 0 16,335,364 0 1,179,220,066 0 2058 105,700,000 406,789,587 1,398,180,036 113,995,444 0 16,887,540 0 1,267,297,052 0 2059 106,600,000 425,267,400 1,498,232,546 118,943,052 0 17,457,115 0 1,361,832,380 0 2060 107,50 0,000 444,584,541 1,605,444,724 124,105,395 0 18,044,614 0 1,463,294,716 0 2061 108,407,598 464,953,480 1,720,974,233 129,511,248 0 18,651,885 0 1,572,811,101 0

PAGE 234

234 Table C 11. LBJ PABs cash flow section Year Net cash flow after OPEX+ CAPEX Debt s ervice rese rve account beginning balance Principle debt service on bonds Interest debts service on bo nds Total debt service on bo nds Reserve account after paying senior d ebt 2010 2011 2012 2013 2014 2015 2016 69,857,314 61, 500,000 0 43,758,000 43,758,000 61,500,000 2017 135,815,775 61,500,000 0 43,758,000 43,758,000 61,500,000 2018 211,633,414 61,500,000 0 43,758,000 43,758,000 61,500,000 2019 267,468,711 61,500,000 0 43,758,000 43,758,000 61,500,000 2020 329,909,660 61,500,000 0 43,758,000 43,758,000 61,500,000 2021 382,610,010 61,500,000 0 43,758,000 43,758,000 61,500,000 2022 410,521,689 52,939,127 0 43,758,000 43,758,000 52,939,127 2023 433,721,177 56,873,896 0 43,758,000 43,758,000 56,873,896 2024 131,642,735 56,873,896 0 43,758,000 43,758,000 56,873,896 2025 120,198,608 56,873,896 0 43,758,000 43,758,000 56,873,896 2026 129,056,542 56,873,896 0 43,758,000 43,758,000 56,873,896 2027 138,538,072 5 6,873,896 0 43,758,000 43,758,000 56,873,896 2028 148,685,994 56,873,896 0 43,758,000 43,758,000 56,873,896 2029 159,545,993 56,873,896 0 43,758,000 43,758,000 56,873,896 2030 152,665,214 78,753,896 21,880,000 43,758,000 65,638,000 78,753,896 2031 183,600,633 99,831,896 45,420,000 41,296,000 86,716,000 99,831,896 2032 196,902,934 99,837,896 48,895,000 37,827,000 86,722,000 99,837,896 2033 211,133,097 99,807,896 52,600,000 34,092,000 86,692,000 99,807,896 2034 227,626,479 99,836,896 56,440,000 30,281,000 86,721,000 99,836,896 2035 243,993,410 99,847,896 60,470,000 26,262,000 86,732,000 99,847,896 2036 261,497,188 99,814,497 64,770,000 21,956,000 86,726,000 99,814,497 2037 281,773,380 99,713 ,422 69,390,000 17,344,000 86,734,000 99,713,422 2038 301,894,364 99,598,422 74,325,000 12,403,000 86,728,000 99,598,422 2039 325,194,028 99,497,422 79,625,000 7,111,000 86,736,000 99,497,422 2040 350,233,096 55,168,901 41,185,000 1, 441,000 42,626,000 55,168,901 2041 377,139,052 138,537,151 0 0 0 138,537,151 2042 403,839,503 138,537,032 0 0 0 138,537,032 2043 434,737,521 138,536,915 0 0 0 138,536,915 2044 467,931,355 138,540,453 0 0 0 138,540,453 2045 5 03,588,836 138,536,667 0 0 0 138,536,667 2046 538,974,555 138,536,559 0 0 0 138,536,559 2047 579,899,316 138,536,455 0 0 0 138,536,455 2048 623,852,311 138,540,261 0 0 0 138,540,261 2049 671,054,405 138,536,230 0 0 0 138,53 6,230 2050 717,897,493 138,536,136 0 0 0 138,536,136 2051 772,045,164 0 0 0 0 0 2052 830,184,827 0 0 0 0 0 2053 887,882,942 0 0 0 0 0 2054 954,555,433 0 0 0 0 0 2055 1,026,131,280 0 0 0 0 0 2056 1,102,966,898 0 0 0 0 0 2057 1,179,220,066 0 0 0 0 0 2058 1,267,297,052 0 0 0 0 0 2059 1,361,832,380 0 0 0 0 0 2060 1,463,294,716 0 0 0 0 0 2061 1,572,811,101 0 0 0 0 0

PAGE 235

235 Table C 12. LBJ TIFIA mandatory debt cash flow section Year D ebt service reserve account after paying senior debt Unpaid b alance on TIFIA m andatory Interest on unpaid b alance TIFIA mandatory debt s ervice Total c ompounded TIFIA mandatory debt s ervice Amount actually paid to TIFIA Cash flow available after TIFIA manda tory debt service 2010 2011 2012 2013 2014 2015 2016 26,099,314 0 0 0 0 0 26,099,314 2017 92,057,775 0 0 0 0 0 92,057,775 2018 167,875,414 0 0 0 0 0 167,875,414 2019 223,710,711 0 0 0 0 0 223,710,711 2020 286,151,660 0 0 1,304,460 1,304,460 1,304,460 284,847,200 2021 338,852,010 0 0 5,246,358 5,246,358 5,246,358 333,605,652 2022 366,763,689 0 0 9,181,127 9,181,127 9,181,127 357,582,562 2023 389,963,177 0 0 13,115,896 13,115,896 13,115,896 376,847,281 2024 87,884,735 0 0 13,115,896 13,115,896 13,115,896 74,768,839 2025 76,440,608 0 0 13,115,896 13,115,896 13,115,896 63,324,712 2026 85,298,542 0 0 13,115,896 13,115,896 13,115,89 6 72,182,646 2027 94,780,072 0 0 13,115,896 13,115,896 13,115,896 81,664,176 2028 104,927,994 0 0 13,115,896 13,115,896 13,115,896 91,812,098 2029 115,787,993 0 0 13,115,896 13,115,896 13,115,896 102,672,097 2030 87,027,214 0 0 13,115,896 13,115,896 13,115,896 73,911,318 2031 96,884,633 0 0 13,115,896 13,115,896 13,115,896 83,768,737 2032 110,180,934 0 0 13,115,896 13,115,896 13,115,896 97,065,038 2033 124,441,097 0 0 13,115,896 13,115,896 13,115,896 111,325,201 2034 140,905,479 0 0 13,115,896 13,115,896 13,115,896 127,789,583 2035 157,261,410 0 0 13,115,896 13,115,896 13,115,896 144,145,514 2036 174,771,188 0 0 13,088,497 13,088,497 13,088,497 161,682,691 2037 195,039,380 0 0 12,979,422 12,979,422 12,979,422 182,059,958

PAGE 236

236 Table C 12.Continued Year Debt service reserve account after paying senior debt Unpaid b alance on TIFIA m andatory Interest on unpaid balance TIFIA mandatory debt service Total c ompounded TIFIA mandator y debt service Amount actually paid to TIFIA Cash flow available after TIFIA mandatory debt service 2038 215,166,364 0 0 12,870,422 12,870,422 12,870,422 202,295,942 2039 238,458,028 0 0 12,761,422 12,761,422 12,761,422 225,696,606 2040 30 7,607,096 0 0 12,542,901 12,542,901 12,542,901 295,064,195 2041 377,139,052 0 0 138,537,151 138,537,151 138,537,151 238,601,901 2042 403,839,503 0 0 138,537,032 138,537,032 138,537,032 265,302,471 2043 434,737,521 0 0 138,536,915 138,536,915 138,536,915 296,200,606 2044 467,931,355 0 0 138,540,453 138,540,453 138,540,453 329,390,902 2045 503,588,836 0 0 138,536,667 138,536,667 138,536,667 365,052,169 2046 538,974,555 0 0 138,536,559 138,536,559 138,536,55 9 400,437,996 2047 579,899,316 0 0 138,536,455 138,536,455 138,536,455 441,362,861 2048 623,852,311 0 0 138,540,261 138,540,261 138,540,261 485,312,050 2049 671,054,405 0 0 138,536,230 138,536,230 138,536,230 532,518,175 2050 717 ,897,493 0 0 138,536,136 138,536,136 138,536,136 579,361,357 2051 772,045,164 0 0 0 0 0 772,045,164 2052 830,184,827 0 0 0 0 0 830,184,827 2053 887,882,942 0 0 0 0 0 887,882,942 2054 954,555,433 0 0 0 0 0 954,555,433 2055 1,026,131,280 0 0 0 0 0 1,026,131,280 2056 1,102,966,898 0 0 0 0 0 1,102,966,898 2057 1,179,220,066 0 0 0 0 0 1,179,220,066 2058 1,267,297,052 0 0 0 0 0 1,267,297,052 2059 1,361,832,380 0 0 0 0 0 1,361,832,380 2060 1,463,294,716 0 0 0 0 0 1,463,294,716 2061 1,572,811,101 0 0 0 0 0 1,572,811,101

PAGE 237

237 Table C 13. LBJ reserve accounts cash flow section Year Debt service reserve account required balan ce Debt service reserve ending balan ce Cash flow ava ilable after RURA Unpaid b alance on TIFIA s cheduled Interest on unpaid b alance TIFIA required scheduled debt s ervice T otal required TIFIA scheduled debt service Amount actually paid to TIFIA 2010 2011 2012 2013 2014 2015 2016 61,500,000 61,500,000 26,099,314 0 0 0 0 0 2017 61,500,000 61,500,000 92,057,775 0 0 0 0 0 2018 61,500,000 61,500,000 167,875,414 0 0 0 0 0 2019 61,500,000 61,500,000 223,710,711 0 0 0 0 0 2020 61,500,000 61,500,00 0 284,847,200 0 0 11,740,144 11,740,144 11,740,144 2021 52,939,127 52,939,127 342,166,525 0 0 47,217,225 47,217,225 47,217,225 2022 56,873,896 56,873,896 353,647,793 0 0 43,282,456 43,282,456 43,282,456 2023 56,873,896 56,873,896 376,847,281 0 0 39,347,687 39,347,687 39,347,687 2024 56,873,896 56,873,896 74,768,839 0 0 39,347,687 39,347,687 39,347,687 2025 56,873,896 56,873,896 63,324,712 0 0 39,347,687 39,347,687 39,347,687 2026 56,873,896 56,873,896 72,182,646 0 0 39,347,687 39,3 47,687 39,347,687 2027 56,873,896 56,873,896 81,664,176 0 0 39,347,687 39,347,687 39,347,687 2028 56,873,896 56,873,896 91,812,098 0 0 39,347,687 39,347,687 39,347,687 2029 78,753,896 78,753,896 80,792,097 0 0 39,347,687 39,347,687 39,347,687 2030 99,831,896 99,831,896 52,833,318 0 0 39,347,687 39,347,687 39,347,687 2031 99,837,896 99,837,896 83,762,737 0 0 39,347,687 39,347,687 39,347,687 2032 99,807,896 99,807,896 97,095,038 0 0 39,347,687 39,347,687 39,347,687 2033 99,836,896 99 ,836,896 111,296,201 0 0 39,347,687 39,347,687 39,347,687 2034 99,847,896 99,847,896 127,778,583 0 0 39,347,687 39,347,687 39,347,687 2035 99,814,497 99,814,497 144,178,913 0 0 39,347,687 39,347,687 39,347,687

PAGE 238

238 Table C 13. Continued Year Debt s ervice reserve account required balance Debt service reserve ending balance Cash flow available after RURA Unpaid b alance on TIFIA s cheduled Interest on unpaid balance TIFIA required scheduled debt service T otal required TIFIA scheduled debt service Amount actually paid to TIFIA 2036 99,713,422 99,713,422 161,783,766 0 0 49,265,491 49,265,491 49,265,491 2037 99,598,422 99,598,422 182,174,958 0 0 48,938,265 48,938,265 48,938,265 2038 99,497,422 99,497,422 202,396,942 0 0 48,611,265 48,611,265 48, 611,265 2039 55,168,901 55,168,901 270,025,127 0 0 48,284,265 48,284,265 48,284,265 2040 138,537,151 138,537,151 211,695,945 0 0 87,628,704 87,628,704 87,628,704 2041 138,537,032 138,537,032 238,602,020 0 0 0 0 0 2042 138,536,915 138,536,915 265,302,588 0 0 0 0 0 2043 138,540,453 138,540,453 296,197,068 0 0 0 0 0 2044 138,536,667 138,536,667 329,394,688 0 0 0 0 0 2045 138,536,559 138,536,559 365,052,277 0 0 0 0 0 2046 138,536,455 138,536,455 400,438,100 0 0 0 0 0 2047 138,54 0,261 138,540,261 441,359,055 0 0 0 0 0 2048 138,536,230 138,536,230 485,316,081 0 0 0 0 0 2049 138,536,136 138,536,136 532,518,269 0 0 0 0 0 2050 0 0 717,897,493 0 0 0 0 0 2051 0 0 772,045,164 0 0 0 0 0 2052 0 0 830,184,827 0 0 0 0 0 2053 0 0 887,882,942 0 0 0 0 0 2054 0 0 954,555,433 0 0 0 0 0 2055 0 0 1,026,131,28 0 0 0 0 0 0 2056 0 0 1,102,966,89 8 0 0 0 0 0 2057 0 0 1,179,220,06 6 0 0 0 0 0 2058 0 0 1,267,297,05 2 0 0 0 0 0 2059 0 0 1,361,832,38 0 0 0 0 0 0 2060 0 0 1,463,294,71 6 0 0 0 0 0 2061 0 0 1,572,811,10 1 0 0 0 0 0

PAGE 239

240 Table C 14. LBJ equity cash flow section Year Cash flows available after TIFIA scheduled debt se rvice Permit f ee Cash flow after permit fe e Cash f lows Equity IRR Total DSCR r equi red = 1.20 2010 2011 2012 2013 (664,762,000 ) (664,762,000) 2014 0 0 2015 0 0 2016 26,099,314 26,099,314 0 0 0.00 2017 92,057,775 0 92,057,775 0 0 0.00 2018 167,875,414 0 167,875,414 0 0 0.00 2019 223,710,711 0 223,710,711 0 0 0.00 2020 273,107,056 0 273,107,056 0 0 0.00 2021 294,949,300 0 294,949,300 0 0 1.80 2022 310,365,337 0 310,365,337 0 0 1.15 2023 337,499,594 0 337,499,594 337,499,594 5.08% 1.23 2024 35,421,152 0 35,421,152 35,421,152 4.32% 1.37 2025 23,977,025 0 23,977,025 23,977,025 3.83% 1.25 2026 32,834,959 0 32,834,959 32,834,959 3.19% 1.34 2027 42,316,489 0 42,316,489 42,316,489 2.46% 1.44 2028 52,464,411 0 52,464,4 11 52,464,411 1.66% 1.55 2029 41,444,410 0 41,444,410 41,444,410 1.10% 1.66 2030 13,485,631 0 13,485,631 13,485,631 0.93% 1.29 2031 44,415,050 0 44,415,050 44,415,050 0.42% 1.32 2032 57,747,351 0 57,747,351 57,747,351 0.16% 1. 41 2033 71,948,514 0 71,948,514 71,948,514 0.77% 1.52 2034 88,430,896 0 88,430,896 88,430,896 1.39% 1.64 2035 104,831,226 0 104,831,226 104,831,226 2.00% 1.75 2036 112,518,275 0 112,518,275 112,518,275 2.54% 1.88 2037 133,236,693 0 133,236,693 133,236,693 3.07% 1.89 2038 153,785,677 0 153,785,677 153,785,677 3.56% 2.03 2039 221,740,862 0 221,740,862 221,740,862 4.14% 2.20 2040 124,067,241 0 124,067,241 124,067,241 4.41% 3.39 2041 238,602,020 0 238,602,020 238,602,020 4.85% 1.67 2042 265,302,588 0 265,302,588 265,302,588 5.25% 2.92 2043 296,197,068 0 296,197,068 296,197,068 5.61% 3.14 2044 329,394,688 0 329,394,688 329,394,688 5.95% 3.38 2045 365,052,277 0 365,052,277 365,052,277 6. 26% 3.64 2046 400,438,100 0 400,438,100 400,438,100 6.54% 3.89 2047 441,359,055 0 441,359,055 441,359,055 6.80% 4.19

PAGE 240

241 Table C 14. Continued Year Cash flows available after TIFIA scheduled debt service Permit f ee Cash flow after permit fee Ca sh f lows Equity IRR Total DSCR r equired = 1.20 2048 485,316,081 0 485,316,081 485,316,081 7.04% 4.50 2049 532,518,269 0 532,518,269 532,518,269 7.27% 4.84 2050 717,897,493 0 717,897,493 717,897,493 7.52% 5.18 2051 772,045,164 0 772,045 ,164 772,045,164 7.74% 100.00 2052 830,184,827 0 830,184,827 830,184,827 7.95% 100.00 2053 887,882,942 0 887,882,942 887,882,942 8.13% 100.00 2054 954,555,433 0 954,555,433 954,555,433 8.30% 100.00 2055 1,026,131,280 0 1,026,131,280 1,026,131,280 8.45% 100.00 2056 1,102,966,898 0 1,102,966,898 1,102,966,898 8.59% 100.00 2057 1,179,220,066 0 1,179,220,066 1,179,220,066 8.72% 100.00 2058 1,267,297,052 0 1,267,297,052 1,267,297,052 8.84% 100.00 2059 1,361,832,380 0 1,361,832,380 1,361,832,380 8.95% 100.00 2060 1,463,294,716 0 1,463,294,716 1,463,294,716 9.05% 100.00 2061 1,572,811,101 0 1,572,811,101 1,572,811,101 9.15% 100.00

PAGE 241

242 Table C 15. ERC traffic & revenue cash flow section Year Traffic Reve nue (2010$) Revenue (year of $) O&M e xpense Major maintenance beginning balance Deposits into major maintenance reserve account Major maintenance ending balance Net cash flow available for debt s ervice Revenue account balance 2012 2013 2014 2015 2016 2017 45,686,870 85,598,823 101,750,101 31,968,096 46,573,000 0 46,573,000 69,782,006 0 2018 45,774,751 85,937,824 104,706,894 32,787,448 46,573,000 14,692,139 31,880,861 71,919,446 0 2019 46,451,631 88,594,28 7 110,642,125 33,766,214 31,880,861 15,282,129 16,598,732 76,875,910 0 2020 47,140,871 91,383,652 116,978,800 34,776,404 16,598,732 15,896,604 702,128 82,202,396 0 2021 47,842,738 94,314,385 123,748,646 35,819,117 702,128 16,536,616 0 72,095,040 0 2022 48,557,503 97,395,592 130,986,243 36,895,495 0 17,203,263 0 76,887,484 0 2023 49,285,443 100,637,076 138,729,321 38,006,723 0 17,897,692 0 82,824,906 0 2024 50,026,848 104,049,435 147,019,128 39,154,033 0 18,621,101 0 89,243,993 0 2025 50,782,004 107,64 4,053 155,900,684 40,338,702 0 19,374,742 0 96,187,240 0 2026 51,515,073 111,252,283 165,154,639 41,551,963 0 20,145,789 0 103,456,888 0 2027 52,096,206 114,198,418 173,766,398 42,757,169 0 20,882,376 0 110,126,853 0 2028 52,688,374 117,280,765 182,917, 967 43,999,903 0 21,647,736 0 117,270,327 0 2029 53,289,921 120,497,126 192,632,752 45,280,873 0 22,442,262 0 124,909,617 0 2030 53,902,941 123,865,593 202,968,197 46,601,930 0 23,267,937 0 133,098,330 0 2031 54,527,705 127,395,497 213,971,165 47,964,45 0 0 24,126,065 0 141,880,649 0 2032 55,164,497 131,096,896 225,692,667 49,369,868 0 25,018,013 0 151,304,787 0 2033 55,813,599 134,980,549 238,188,120 50,819,666 0 25,945,201 0 161,423,253 0 2034 56,337,154 138,196,705 249,959,966 52,268,364 0 26,843,29 2 0 170,848,310 0 2035 56,674,235 140,307,803 260,122,824 53,692,677 0 27,679,001 0 178,751,146 0 2036 57,014,631 142,472,391 270,739,245 55,156,723 0 28,541,377 0 187,041,145 0 2037 57,358,379 144,692,185 281,831,440 56,661,642 0 29,431,293 0 195,738,5 06 0 2038 57,705,522 146,969,004 293,422,884 58,208,609 0 30,349,652 0 204,864,623 0

PAGE 242

243 Table C 15. Continued Year Traffic Revenue (2010$) Revenue (year of $) O&M e xpense Major maintenance beginning balance Deposits into major maintenance reserve account Ma jor maintenance ending balance Net cash flow available for debt s ervice Revenue account balance 2039 58,056,099 149,304,708 305,538,258 59,798,834 0 31,297,385 0 214,442,039 0 2040 58,410,151 151,701,240 318,203,602 61,433,561 0 32,275,457 0 224,494,583 0 2041 58,588,938 152,926,004 328,791,944 63,041,738 0 33,183,605 0 232,566,601 0 2042 58,768,613 154,166,814 339,746,185 64,692,296 0 34,117,503 0 240,936,386 0 2043 58,949,185 155,423,961 351,079,549 66,386,362 0 35,077,891 0 249,615,296 0 2044 59,13 0,658 156,697,710 362,805,677 68,125,091 0 36,065,524 0 258,615,062 0 2045 59,313,037 157,988,335 374,938,737 69,909,670 0 37,081,181 0 267,947,886 0 2046 59,404,682 158,640,879 385,899,540 71,699,364 0 38,066,937 0 276,133,239 0 2047 59,496,557 159,297 ,765 397,184,876 73,534,957 0 39,078,957 0 284,570,961 0 2048 59,588,663 159,959,034 408,804,489 75,417,629 0 40,117,941 0 293,268,919 0 2049 59,680,998 160,624,701 420,768,368 77,348,588 0 41,184,608 0 302,235,171 0 2050 59,773,565 161,294,822 433,086, 894 79,329,077 0 42,279,699 0 311,478,119 0 2051 59,819,965 161,631,777 444,841,431 81,336,335 0 43,370,332 0 320,134,764 0 2052 59,866,423 161,969,859 456,916,193 83,394,407 0 44,489,115 0 329,032,671 0 2053 59,912,939 162,309,071 469,319,936 85,504,57 9 0 45,636,775 0 338,178,582 0 2054 59,959,514 162,649,424 482,061,678 87,668,170 0 46,814,058 0 347,579,450 0 2055 59,976,146 162,771,138 494,482,975 89,869,383 0 47,997,720 0 356,615,872 0 2056 60,029,493 163,162,149 508,062,598 92,147,378 0 49,241,42 3 0 366,673,798 0 2057 60,052,854 163,333,671 521,311,609 94,465,094 0 50,492,100 0 376,354,415 0 2058 60,076,229 163,505,476 534,906,457 96,841,112 0 51,774,547 0 386,290,798 0 2059 60,099,619 163,677,572 548,856,205 99,276,900 0 53,089,573 0 396,489,7 32 0 2060 60,123,023 163,849,953 563,170,101 101,773,961 0 54,438,003 0 406,958,137 0 2061 60,134,733 163,936,270 577,553,452 104,326,073 0 55,809,821 0 417,417,558 0 2062 60,146,447 164,022,662 592,304,261 106,942,186 0 57,216,210 0 428,145,865 0 2063 60,158,164 164,109,122 607,431,889 109,623,902 0 58,658,040 0 439,149,947 0 2064 60,169,884 164,195,650 622,945,966 112,372,867 0 60,136,205 0 450,436,894 0 2065 60,181,608 164,282,253 638,856,394 115,190,768 0 61,651,620 0 462,014,005 0

PAGE 243

244 Table C 15. Co ntinued Year Traffic Revenue (2010$) Revenue (year of $) O&M e xpense Major maintenance beginning balance Deposits into major maintenance reserve account Major maintenance ending balance Net cash flow available for debt s ervice Revenue account balance 2066 60,187,473 164,325,593 655,000,560 118,074,937 0 63,199,069 0 473,726,554 0 2067 60,193,338 164,368,946 671,552,695 121,031,320 0 64,785,358 0 485,736,017 0 2068 60,199,204 164,412,317 688,523,141 124,061,726 0 66,411,463 0 498,049,951 0 2069 60,205,07 0 164,455,699 705,922,437 127,168,007 0 68,078,383 0 510,676,047 0

PAGE 244

245 Table C 16. ERC PABs cash flow section Year Net cash flow after RURA a ccount Debt service reserve account beginning balance Total debt service on bonds 2012 2013 2014 2015 2016 2017 69,782,006 18,547,000 37,095,000 2018 71,919,446 18,547,000 37,095,000 2019 76,875,910 18,547,000 37,095,000 2020 82,202,396 18,547,000 37,095,000 2021 72,095,040 18,547,000 37,095,000 2022 76,887,484 18,547,000 38,43 5,000 2023 82,824,906 18,547,000 40,526,000 2024 89,243,993 18,547,000 42,766,000 2025 96,187,240 18,547,000 46,610,000 2026 103,456,888 18,547,000 48,742,000 2027 110,126,853 18,547,000 52,295,000 2028 117,270,327 18,547,000 54,620 ,000 2029 124,909,617 18,547,000 56,142,000 2030 133,098,330 18,547,000 49,930,000 2031 141,880,649 18,547,000 53,024,000 2032 151,304,787 18,547,000 58,948,000 2033 161,423,253 18,547,000 63,423,000 2034 170,848,310 18,547,000 66,4 26,000 2035 178,751,146 18,547,000 71,881,000 2036 187,041,145 18,547,000 75,026,000 2037 195,738,506 18,547,000 69,605,000 2038 204,864,623 18,547,000 72,119,000 2039 214,442,039 18,547,000 74,079,000 2040 224,494,583 18,547,000 80 ,855,000 2041 232,566,601 18,547,000 75,254,000 2042 240,936,386 18,547,000 2043 249,615,296 18,547,000 2044 258,615,062 18,547,000 2045 267,947,886 18,547,000 2046 276,133,239 18,547,000 2047 284,570,961 2048 293,268,9 19 2049 302,235,171 2050 311,478,119 2051 320,134,764 2052 329,032,671

PAGE 245

246 Table C 16. Continued Year Net cash flow after RURA a ccount Debt service reserve account beginning balance Total debt service on bonds 2053 338,178,582 2054 347,579,450 2055 356,615,872 2056 366,673,798 2057 376,354,415 2058 386,290,798 2059 396,489,732 2060 406,958,137 2061 417,417,558 2060 428,145,865 2061 439,149,947 2062 450,436,894 2063 462,014,0 05 2064 473,726,554 2065 485,736,017 2066 498,049,951 2067 510,676,047 2068 498,049,951 2069 510,676,047

PAGE 246

247 Table C 17. ERC TIFIA mandatory debt cash flow section Year Debt service reserve account after paying senior debt N et cash flow available after senior debt service Unpaid balance on TIFIA m andatory Interest on unpaid b alance TIFIA mandatory debt service Total c ompounded TIFIA mandatory debt service Amount actually paid to TIFIA Cash flow available after TIFIA mandatory debt service 2012 2013 2014 2015 2016 2017 18,547,000 32,687,006 0 0 0 32,687,006 2018 18,547,000 34,824,446 0 0 0 0 0 34,824,446 2019 18,547,000 39,780,910 0 0 0 0 0 39,780,910 2020 18,547,000 45,107,396 0 0 0 0 0 45,107,396 2021 18,547,000 35,000,040 0 0 0 0 0 35,000,040 2022 18,547,000 38,452,484 0 0 2,777,000 2,777,000 2,777,000 35,675,484 2023 18,547,000 42,298,906 0 0 3,703,000 3,703,000 3,703,000 38,595,906 2024 18,547,000 46,477,993 0 0 4,629,0 00 4,629,000 4,629,000 41,848,993 2025 18,547,000 49,577,240 0 0 4,629,000 4,629,000 4,629,000 44,948,240 2026 18,547,000 54,714,888 0 0 4,629,000 4,629,000 4,629,000 50,085,888 2027 18,547,000 57,831,853 0 0 4,629,000 4,629,000 4,629,000 53,202,853 20 28 18,547,000 62,650,327 0 0 4,629,000 4,629,000 4,629,000 58,021,327 2029 18,547,000 68,767,617 0 0 4,629,000 4,629,000 4,629,000 64,138,617 2030 18,547,000 83,168,330 0 0 18,058,000 18,058,000 18,058,000 65,110,330 2031 18,547,000 88,856,649 0 0 18,05 8,000 18,058,000 18,058,000 70,798,649 2032 18,547,000 92,356,787 0 0 18,058,000 18,058,000 18,058,000 74,298,787 2033 18,547,000 98,000,253 0 0 18,058,000 18,058,000 18,058,000 79,942,253 2034 18,547,000 104,422,310 0 0 18,058,000 18,058,000 18,058,000 86,364,310 2035 18,547,000 106,870,146 0 0 18,058,000 18,058,000 18,058,000 88,812,146 2036 18,547,000 112,015,145 0 0 18,058,000 18,058,000 18,058,000 93,957,145 2037 18,547,000 126,133,506 0 0 27,972,000 27,972,000 27,972,000 98,161,506

PAGE 247

248 Table C 17. Continued Year Debt service reserve account after paying senior debt Net cash flow available after senior debt service Unpaid balance on TIFIA m andatory Interest on unpaid balance TIFIA mandatory debt service Total c ompounded TIFIA mandatory debt service Amount actually paid to TIFIA Cash flow available after TIFIA mandatory debt service 2038 18,547,000 132,745,623 0 0 27,632,000 27,632,000 27,632,000 105,113,623 2039 18,547,000 140,363,039 0 0 27,927,000 27,927,000 27,927,000 112,436,039 2040 18,547,00 0 143,639,583 0 0 26,952,000 26,952,000 26,952,000 116,687,583 2041 18,547,000 157,312,601 0 0 26,612,000 26,612,000 26,612,000 130,700,601 2042 18,547,000 240,936,386 0 0 105,439,000 105,439,000 105,439,000 135,497,386 2043 18,547,000 249,615,296 0 0 1 05,439,000 105,439,000 105,439,000 144,176,296 2044 18,547,000 258,615,062 0 0 105,441,000 105,441,000 105,441,000 153,174,062 2045 18,547,000 267,947,886 0 0 105,438,000 105,438,000 105,438,000 162,509,886 2046 18,547,000 276,133,239 0 0 105,438,000 10 5,438,000 105,438,000 170,695,239 2047 284,570,961 0 0 0 0 0 284,570,961 2048 293,268,919 0 0 0 0 0 293,268,919 2049 302,235,171 0 0 0 0 0 302,235,171 2050 311,478,119 0 0 0 0 0 311,478,119 2051 320,134,764 0 0 0 0 0 320,134,764 2052 329,032,67 1 0 0 0 0 0 329,032,671 2053 338,178,582 0 0 0 0 0 338,178,582 2054 347,579,450 0 0 0 0 0 347,579,450 2055 356,615,872 0 0 0 0 0 356,615,872 2056 366,673,798 0 0 0 0 0 366,673,798 2057 376,354,415 0 0 0 0 0 376,354,415 2058 386,290,798 0 0 0 0 0 386,290,798 2059 396,489,732 0 0 0 0 0 396,489,732 2060 406,958,137 0 0 0 0 0 406,958,137 2061 417,417,558 0 0 0 0 0 417,417,558 2062 428,145,865 0 0 0 0 0 428,145,865 2063 439,149,947 0 0 0 0 0 439,149,947

PAGE 248

249 Table C 17. Continued Year Debt ser vice reserve account after paying senior debt Net cash flow available after senior debt service Unpaid balance on TIFIA m andatory Interest on unpaid balance TIFIA mandatory debt service Total c ompounded TIFIA mandatory debt service Amount actually paid to TIFIA Cash flow available after TIFIA mandatory debt service 2064 450,436,894 0 0 0 0 0 450,436,894 2065 462,014,005 0 0 0 0 0 462,014,005 2066 473,726,554 0 0 0 0 0 473,726,554 2067 485,736,017 0 0 0 0 0 485,736,017 2068 498,049,951 0 0 0 0 0 49 8,049,951 2069 510,676,047 0 0 0 0 0 510,676,047

PAGE 249

250 Table C 18. ERC reserve accounts cash flow sec tion Year Debt service reserve account required balance Debt service reserve ending balance Cash flow available after RURA 2012 2013 2014 2015 2016 2017 18,547,000 18,547,000 32,687,006 2018 18,547,000 18,547,000 34,824,446 2019 18,547,000 18,547,000 39,780,910 2020 18,547,000 18,547,000 45,107,396 2021 18,547,000 18,547,000 35,000,040 2022 18,547,000 18,547,0 00 35,675,484 2023 18,547,000 18,547,000 38,595,906 2024 18,547,000 18,547,000 41,848,993 2025 18,547,000 18,547,000 44,948,240 2026 18,547,000 18,547,000 50,085,888 2027 18,547,000 18,547,000 53,202,853 2028 18,547,000 18,547,000 58,021,327 2029 18,547,000 18,547,000 64,138,617 2030 18,547,000 18,547,000 65,110,330 2031 18,547,000 18,547,000 70,798,649 2032 18,547,000 18,547,000 74,298,787 2033 18,547,000 18,547,000 79,942,253 2034 18,547,000 18,547,000 86,3 64,310 2035 18,547,000 18,547,000 88,812,146 2036 18,547,000 18,547,000 93,957,145 2037 18,547,000 18,547,000 98,161,506 2038 18,547,000 18,547,000 105,113,623 2039 18,547,000 18,547,000 112,436,039 2040 18,547,000 18,547,000 116,68 7,583 2041 18,547,000 18,547,000 130,700,601 2042 18,547,000 18,547,000 135,497,386 2043 18,547,000 18,547,000 144,176,296 2044 18,547,000 18,547,000 153,174,062 2045 0 18,547,000 162,509,886 2046 189,242,239 2047 284,570,96 1 2048 293,268,919 2049 302,235,171 2050 311,478,119 2051 320,134,764 2052 329,032,671

PAGE 250

251 Table C 18. Continued Year Debt service reserve account required balance Debt service reserve ending balance Cash flow available after RURA 2053 338,178,582 2054 347,579,450 2055 356,615,872 2056 366,673,798 2057 376,354,415 2058 386,290,798 2059 396,489,732 2060 406,958,137 2061 417,417,558 2062 428,145,865 2063 439,149,947 2064 450,436,894 2065 4 62,014,005 2066 473,726,554 2067 485,736,017 2068 498,049,951 2069 510,676,047

PAGE 251

252 Table C 19. ERC TIFIA scheduled cash flow section Year U npaid b alance on TIFIA s cheduled Interest on unpaid balance TIFIA required scheduled debt service To tal r equired TIFIA scheduled debt service Amount actually paid to TIFIA 2010 2011 2012 2013 2014 2015 2016 2017 0 0 0 0 0 2018 0 0 0 0 0 2019 0 0 0 0 0 2020 0 0 0 0 0 2021 0 0 9,334,000 9,334,000 9,334,000 2022 0 0 15,280,000 15,280,000 15,280,000 2023 0 0 14,355,000 14,355,000 14,355,000 2024 0 0 13,429,000 13,429,000 13,429,000 2025 0 0 13,429,000 13,429,000 13,429,000 2026 0 0 13,429,000 13,429,000 13,429 ,000 2027 0 0 13,429,000 13,429,000 13,429,000 2028 0 0 13,429,000 13,429,000 13,429,000 2029 0 0 13,429,000 13,429,000 13,429,000 2030 0 0 0 0 0 2031 0 0 0 0 0 2032 0 0 0 0 0 2033 0 0 0 0 0 2034 0 0 0 0 0 2035 0 0 0 0 0 2036 0 0 0 0 0 2037 0 0 0 0 0

PAGE 252

253 Table C 19. Continued Year U npaid b alance on TIFIA s cheduled Interest on unpaid balance TIFIA required scheduled debt service Total r equired TIFIA scheduled debt service Amount actually paid t o TIFIA 2038 0 0 0 0 0 2039 0 0 0 0 0 2040 0 0 0 0 0 2041 0 0 0 0 0 2042 0 0 0 0 0 2043 0 0 0 0 0 2044 0 0 0 0 0 2045 0 0 0 0 0 2046 0 0 0 0 0 2047 0 0 0 0 0 2048 0 0 0 0 0 2049 0 0 0 0 0 2050 0 0 0 0 0 2051 0 0 0 0 0 2052 0 0 0 0 0 2053 0 0 0 0 0 2054 0 0 0 0 0 2055 0 0 0 0 0 2056 0 0 0 0 0 2057 0 0 0 0 0 2058 0 0 0 0 0 2059 0 0 0 0 0 2060 0 0 0 0 0 2061 0 0 0 0 0 2062 0 0 0 0 0 2063 0 0 0 0 0 2064 0 0 0 0 0 2065 0 0 0 0 0

PAGE 253

254 Table C 19. Continued Year U npaid b alance on TIFIA s cheduled Interest on unpaid balance TIFIA required scheduled debt service Total r equired TIFIA scheduled debt service Amount actually paid to TIFIA 2066 0 0 0 0 0 2067 0 0 0 0 0 2068 0 0 0 0 0 2069 0 0 0 0 0 Table C 20. ERC equity cash flow section Year Cash flows available after TIFIA scheduled debt service Cash flows to equity P ermit f ee Cash flow after permit fee Equity IRR Total DSCR r equired = 1.30 Senior debt service cov erage ratio r equired = 1.45 2012 (272,467,000) 2013 0 2014 0 2015 0 2016 0 2017 32,687,006 32,687,006 0 32,687,006 3 4.56% 2018 34,824,446 34,824,446 0 34,824,446 22.24% 1.88 1.88 2019 39,780,910 39,780,910 0 39,780,910 14.13% 1.94 1.94 2020 45,107,396 45,107,396 0 45,107,396 8.32% 2.07 2.07 2021 25,666,040 25,666,040 0 25,666,040 5.87% 2.22 2. 22 2022 20,395,484 20,395,484 0 20,395,484 4.22% 1.55 1.94 2023 24,240,906 24,240,906 0 24,240,906 2.57% 1.36 2.00 2024 28,419,993 28,419,993 0 28,419,993 0.99% 1.41 2.04 2025 31,519,240 31,519,240 0 31,519,240 0.42% 1.47 2.09 2026 36,656,888 36,656,888 0 36,656,888 1.73% 1.49 2.06 2027 39,773,853 39,773,853 0 39,773,853 2.86% 1.55 2.12 2028 44,592,327 44,592,327 0 44,592,327 3.86% 1.57 2.11 2029 50,709,617 50,709,617 0 50,709,617 4.78% 1.61 2.15

PAGE 254

255 Table C 20. Co ntinued Year Cash flows available after TIFIA scheduled debt service Cash flows to equity Permit f ee Cash flow after permit fee Equity IRR Total DSCR r equired = 1.30 Senior debt service cov erage ratio r equired = 1.45 2030 65,110,330 65,110,330 0 65 ,110,330 5.70% 1.68 2.22 2031 70,798,649 70,798,649 0 70,798,649 6.49% 1.96 2.67 2032 74,298,787 74,298,787 0 74,298,787 7.14% 2.00 2.68 2033 79,942,253 79,942,253 0 79,942,253 7.71% 1.96 2.57 2034 86,364,310 86,364,310 0 86,364,310 8. 21% 1.98 2.55 2035 88,812,146 88,812,146 0 88,812,146 8.63% 2.02 2.57 2036 93,957,145 93,957,145 0 93,957,145 9.00% 1.99 2.49 2037 98,161,506 98,161,506 0 98,161,506 9.32% 2.01 2.49 2038 105,113,623 105,113,623 0 105,113,623 9.60% 2.01 2.81 2039 112,436,039 112,436,039 0 112,436,039 9.86% 2.05 2.84 2040 116,687,583 116,687,583 0 116,687,583 10.08% 2.10 2.89 2041 130,700,601 130,700,601 0 130,700,601 10.29% 2.08 2.78 2042 135,497,386 135,497,386 0 135,497,386 10.48% 2. 28 3.09 2043 144,176,296 144,176,296 0 144,176,296 10.64% 2.29 100.00 2044 153,174,062 153,174,062 0 153,174,062 10.79% 2.37 100.00 2045 162,509,886 162,509,886 0 162,509,886 10.93% 2.45 100.00 2046 189,242,239 189,242,239 0 189,242,239 11.07% 2.54 100.00 2047 284,570,961 284,570,961 0 284,570,961 11.24% 2.62 100.00 2048 293,268,919 293,268,919 0 293,268,919 11.39% 100.00 100.00 2049 302,235,171 302,235,171 0 302,235,171 11.52% 100.00 100.00 2050 311,478,119 311,478,119 0 311,478,119 11.63% 100.00 100.00 2051 320,134,764 320,134,764 0 320,134,764 11.73% 100.00 100.00 2052 329,032,671 329,032,671 0 329,032,671 11.81% 100.00 100.00 2053 338,178,582 338,178,582 0 338,178,582 11.89% 100.00 100.00 2054 347,5 79,450 347,579,450 0 347,579,450 11.96% 100.00 100.00

PAGE 255

256 Table C 20. Continued Year Cash flows available after TIFIA scheduled debt service Cash flows to equity Permit f ee Cash flow after permit fee Equity IRR Total DSCR r equired = 1.30 Senior debt service cov erage ratio r equired = 1.45 2055 356,615,872 356,615,872 0 356,615,872 12.02% 100.00 100.00 2056 366,673,798 366,673,798 0 366,673,798 12.07% 100.00 100.00 2057 376,354,415 376,354,415 0 376,354,415 12.12% 100.00 100.00 2058 386,290,798 386,290,798 0 386,290,798 12.16% 100.00 100.00 2059 396,489,732 396,489,732 0 396,489,732 12.20% 100.00 100.00 2060 406,958,137 406,958,137 0 406,958,137 12.24% 100.00 100.00 2061 417,417,558 417,417,558 0 417,417,558 12.2 7% 100.00 100.00 2062 428,145,865 428,145,865 0 428,145,865 12.29% 100.00 100.00 2063 439,149,947 439,149,947 0 439,149,947 12.32% 100.00 100.00 2064 450,436,894 450,436,894 0 450,436,894 12.34% 100.00 100.00 2065 462,014,005 462,014,005 0 462,014,005 12.36% 100.00 100.00 2066 473,726,554 473,726,554 0 473,726,554 12.38% 100.00 100.00 2067 485,736,017 485,736,017 0 485,736,017 12.39% 100.00 100.00 2068 498,049,951 498,049,951 0 498,049,951 12.41% 100.00 100.00 2069 510,6 76,047 510,676,047 0 510,676,047 12.42% 100.00 100.00

PAGE 256

257 Table C 21. I 95 traffic & revenue cash flow section Year Traffic Revenue (2010$) Revenue (year of $) O&M Expense Major maintenance beginning balance Major maintenance expenses Major maintenance ending balance Net cash flow available for debt service Revenue account balance 2012 2013 2014 2015 19,300,000 37,732,457 42,690,812 22,960,860 1,000,000 0 1,000,000 20,729,952 2016 23,200,000 55,712,972 64,609,967 26,718, 239 1,000,000 0 1,000,000 38,891,728 0 2017 25,500,000 70,107,626 83,335,937 29,310,494 1,000,000 0 1,000,000 55,025,443 0 2018 26,000,000 73,699,214 89,795,336 30,472,040 1,000,000 0 1,000,000 60,323,296 0 2019 26,300,000 75,941,892 94,841,017 31,497,5 43 1,000,000 0 1,000,000 64,343,474 0 2020 26,800,000 79,832,368 102,192,180 32,735,473 1,000,000 4,195,906 0 66,260,801 0 2021 27,200,000 83,087,767 109,018,350 33,923,262 0 9,665,394 0 65,429,693 0 2022 27,500,000 85,616,140 115,144,189 35,055,322 0 1 0,016,298 0 70,072,569 0 2023 27,800,000 88,221,451 121,614,245 36,222,783 0 10,378,706 0 75,012,756 0 2024 28,100,000 90,906,043 128,447,859 37,426,707 0 10,752,974 0 80,268,178 0 2025 28,400,000 93,672,328 135,665,460 38,668,189 0 11,139,469 0 85,857, 803 0 2026 28,600,000 95,563,126 141,863,998 39,843,866 0 11,498,364 0 90,521,769 0 2027 28,900,000 98,471,127 149,835,464 41,161,258 0 11,909,450 0 96,764,756 0 2028 29,000,000 99,459,993 155,123,645 42,300,066 0 12,249,426 0 100,574,154 0 2029 29,200 ,000 101,467,617 162,211,226 43,582,608 0 12,642,252 0 105,986,365 0 2030 29,400,000 103,515,766 169,622,636 44,902,840 0 13,047,064 0 111,672,732 0 2031 29,600,000 105,605,257 177,372,673 46,261,844 0 13,464,215 0 117,646,614 0 2032 29,800,000 107,736, 925 185,476,809 47,660,734 0 13,894,069 0 123,922,005 0 2033 30,000,000 109,911,621 193,951,221 49,100,656 0 14,337,001 0 130,513,564 0 2034 30,100,000 111,015,376 200,796,391 50,455,478 0 14,744,411 0 135,596,502 0 2035 30,300,000 113,256,248 209,970,7 53 51,977,844 0 15,213,440 0 142,779,469 0 2036 30,500,000 115,542,353 219,564,291 53,544,792 0 15,696,705 0 150,322,793 0 2037 30,700,000 117,874,604 229,596,156 55,157,602 0 16,194,625 0 158,243,928 0 2038 30,900,000 120,253,932 240,086,376 56,817,587 0 16,707,631 0 166,561,157 0 2039 31,000,000 121,461,546 248,559,805 58,382,063 0 17,180,743 0 172,996,999 0

PAGE 257

258 Table C 21. Continued Year Traffic Revenue (2010$) Revenue (year of $) O&M Expense Major maintenance beginning balance Major maintenance expense s Major maintenance ending balance Net cash flow available for debt service Revenue account balance 2040 31,200,000 123,913,277 259,916,472 60,136,887 0 17,723,877 0 182,055,708 0 2041 31,400,000 126,414,496 271,792,023 61,942,964 0 18,283,428 0 191,565, 631 0 2042 31,500,000 127,683,976 281,384,448 63,646,649 0 18,800,197 0 198,937,602 0 2043 31,700,000 130,261,308 294,240,868 65,555,791 0 19,392,553 0 209,292,524 0 2044 31,800,000 131,569,418 304,625,586 67,357,649 0 19,940,071 0 217,327,866 0 2045 3 2,100,000 135,573,095 321,742,773 69,542,702 0 20,631,390 0 231,568,681 0 2046 32,200,000 136,934,547 333,098,122 71,452,482 0 21,213,053 0 240,432,586 0 2047 32,300,000 138,309,671 344,854,239 73,414,288 0 21,810,906 0 249,629,045 0 2048 32,500,000 141 ,101,486 360,610,586 75,609,406 0 22,494,607 0 262,506,573 0 2049 32,700,000 143,949,654 377,086,839 77,868,397 0 23,198,861 0 276,019,581 0 2050 32,800,000 145,395,225 390,395,461 80,004,094 0 23,851,551 0 286,539,816 0 2051 33,393,048 154,271,456 424, 584,426 83,153,000 0 24,889,874 0 316,541,552 0 2052 33,597,357 157,453,236 444,174,822 85,637,489 0 25,668,212 0 332,869,121 0 2053 33,801,665 160,700,639 464,669,123 88,194,232 0 26,469,910 0 350,004,980 0 2054 34,005,974 164,015,017 486,109,033 90,82 5,289 0 27,295,651 0 367,988,093 0 2055 34,210,283 167,397,754 508,538,185 93,532,777 0 28,146,135 0 386,859,273 0 2056 34,414,591 170,850,258 532,002,222 96,318,874 0 29,022,083 0 406,661,264 0 2057 34,618,900 174,373,968 556,548,893 99,185,818 0 29,92 4,238 0 427,438,837 0 2058 34,823,209 177,970,353 582,228,151 102,135,909 0 30,853,361 0 449,238,881 0 2059 35,027,517 181,640,912 609,092,255 105,171,514 0 31,810,238 0 472,110,503 0 2060 35,231,826 185,387,174 637,195,873 108,295,064 0 32,795,675 0 49 6,105,133 0 2061 35,436,134 189,210,702 666,596,197 111,509,060 0 33,810,503 0 521,276,634 0 2062 35,640,443 193,113,088 697,353,056 114,816,071 0 34,855,575 0 547,681,410 0 2063 35,844,752 197,095,959 729,529,042 118,218,739 0 35,931,769 0 575,378,534 0 2064 36,049,060 201,160,975 763,189,634 121,719,781 0 37,039,988 0 604,429,865 0 2065 36,253,369 205,309,830 798,403,331 125,321,988 0 38,181,161 0 634,900,182 0 2066 36,457,678 209,544,253 835,241,793 129,028,230 0 39,356,242 0 666,857,321 0 2067 36 ,661,986 213,866,010 873,779,989 132,841,459 0 40,566,214 0 700,372,316 0

PAGE 258

259 Table C 21. Continue Year Traffic Revenue (2010$) Revenue (year of $) O&M Expense Major maintenance beginning balance Major maintenance expenses Major maintenance ending balance Net cash flow available for debt service Revenue account balance 2068 36,866,295 218,276,901 914,096,343 136,764,706 0 41,812,087 0 735,519,550 0 2069 37,070,604 222,778,764 956,272,900 140,801,090 0 43,094,900 0 772,376,911 0 2070 37,274,912 227,373,476 1 ,000,395,491 144,953,815 0 44,415,721 0 811,025,956 0 2071 37,479,221 232,062,953 1,046,553,905 149,226,175 0 45,775,648 0 851,552,082 0 2072 37,683,529 236,849,147 1,094,842,077 153,621,557 0 47,175,812 0 894,044,707 0 2073 37,887,838 241,734,055 1,145 ,358,272 158,143,443 0 48,617,375 0 938,597,455 0 2074 38,092,147 246,719,712 1,198,205,294 162,795,408 0 50,101,531 0 985,308,355 0 2075 38,296,455 251,808,195 1,253,490,686 167,581,133 0 51,629,508 0 1,034,280,044 0 2076 38,500,764 257,001,627 1,311,3 26,955 172,504,397 0 53,202,571 0 1,085,619,987 0 2077 38,705,073 262,302,170 1,371,831,799 177,569,085 0 54,822,019 0 1,139,440,694 0 2078 38,909,381 267,712,035 1,435,128,346 182,779,193 0 56,489,188 0 1,195,859,965 0 2079 39,113,690 273,233,476 1,501 ,345,407 188,138,826 0 58,205,451 0 1,255,001,130 0 2080 39,317,998 278,868,795 1,570,617,734 193,652,203 0 59,972,222 0 1,316,993,309 0 2081 39,522,307 284,620,339 1,643,086,298 199,323,662 0 61,790,953 0 1,381,971,683 0 2082 39,726,616 290,490,506 1,7 18,898,573 205,157,661 0 63,663,137 0 1,450,077,774 0 2083 39,930,924 296,481,743 1,798,208,839 211,158,784 0 65,590,312 0 1,521,459,743 0 2084 40,135,233 302,596,546 1,881,178,494 217,331,738 0 67,574,056 0 1,596,272,700 0 2085 40,339,542 308,837,464 1 ,967,976,383 223,681,367 0 69,615,993 0 1,674,679,023 0 2086 40,543,850 315,207,099 2,058,779,141 230,212,644 0 71,717,793 0 1,756,848,704 0 2087 40,748,159 321,708,104 2,153,771,554 236,930,685 0 73,881,174 0 1,842,959,695 0

PAGE 259

260 Table C 22. I 95 PABs cas h flow section Year Opening balance in RURA RURA after c uring O&M d eficiencies Net cash flow after RURA a ccount Debt service reserve account beginning balance Principle debt service on bonds Interest debts service on bonds Total debt service on bonds 2012 2013 2014 2015 35,000,000 35,000,000 20,729,952 12,097,500 0 12,098,000 12,098,000 2016 35,000,000 35,000,000 38,891,728 12,097,500 0 12,098,000 12,098,000 2017 35,000,000 35,000,000 55,025,443 12,097,500 0 12,098,000 12,098,000 2018 35,000,000 35,000,000 60,323,296 12,097,500 0 12,098,000 12,098,000 2019 478,500 478,500 64,343,474 12,097,500 0 12,098,000 12,098,000 2020 949,500 949,500 66,260,801 12,097,500 0 12,098,000 12,098,000 2021 1,661,250 1,661,250 65,429,693 0 12,098,000 12,098,000 2022 2,373,750 2,373,750 70,072,569 0 12,098,000 12,098,000 2023 2,373,750 2,373,750 75,012,756 0 12,098,000 12,098,000 2024 2,373,750 2,373,750 80,268,178 0 12,098,0 00 12,098,000 2025 2,373,750 2,373,750 85,857,803 0 12,098,000 12,098,000 2026 2,373,750 2,373,750 90,521,769 0 12,098,000 12,098,000 2027 2,373,750 2,373,750 96,764,756 0 12,098,000 12,098,000 2028 2,373,750 2,373,750 100,574, 154 0 12,098,000 12,098,000 2029 2,373,750 2,373,750 105,986,365 0 12,098,000 12,098,000 2030 2,373,750 2,373,750 111,672,732 8,535,000 11,993,000 20,528,000 2031 2,373,750 2,373,750 117,646,614 14,080,000 11,501,000 25,581,000 2032 2,373,750 2,373,750 123,922,005 16,820,000 10,762,000 27,582,000 2033 2,373,750 2,373,750 130,513,564 20,150,000 9,879,000 30,029,000 2034 2,373,750 2,373,750 135,596,502 25,695,000 8,816,000 34,511,000 2035 2,373,750 2,373, 750 142,779,469 28,320,000 7,483,000 35,803,000 2036 2,373,750 2,373,750 150,322,793 29,750,000 6,049,000 35,799,000 2037 2,373,750 2,373,750 158,243,928 31,255,000 4,543,000 35,798,000 2038 2,373,750 2,373,750 166,561,157 32,840 ,000 2,961,000 35,801,000

PAGE 260

261 Table C 22. Continued Year Opening balance in RURA RURA after c uring O&M d eficiencies Net cash flow after RURA a ccount Debt service reserve account beginning balance Principle debt service on bonds Interest debts service on bo nds Total debt service on bonds 2039 2,373,750 2,373,750 172,996,999 34,505,000 1,298,000 35,803,000 2040 32,306,250 32,306,250 182,055,708 0 2041 32,306,250 32,306,250 191,565,631 0 2042 32,306,250 32,306,250 198,937,602 0 2043 32,306,250 32,306,250 209,292,524 0 2044 32,306,250 32,306,250 217,327,866 0 2045 32,306,250 32,306,250 231,568,681 0 2046 32,306,250 32,306,250 240,432,586 0 2047 32,306,250 32,306,250 249,629,045 0 2048 32,30 6,250 32,306,250 262,506,573 0 2049 32,281,500 32,281,500 276,019,581 0 2050 0 0 286,539,816 0 2051 0 0 316,541,552 0 2052 0 0 332,869,121 0 2053 0 0 350,004,980 0 2054 0 0 367,988,093 0 2055 0 0 386,859,273 0 2056 0 0 406,661,264 0 2057 0 0 427,438,837 0 2058 0 0 449,238,881 0 2059 0 0 472,110,503 0 2060 0 0 496,105,133 0 2061 0 0 521,276,634 0 2062 0 0 547,681,410 0 2063 0 0 575,378 ,534 0 2064 0 0 604,429,865 0 2065 0 0 634,900,182 0

PAGE 261

262 Table C 22. Continued Year Opening balance in RURA RURA after c uring O&M d eficiencies Net cash flow after RURA a ccount Debt service reserve account beginning balance Principle debt service on bonds Interest debts service on bonds Total debt service on bonds 2066 0 0 666,857,321 0 2067 0 0 700,372,316 0 2068 0 0 735,519,550 0 2069 0 0 772,376,911 0 2070 0 0 811,025,956 0 2071 0 0 851,552,082 0 2072 0 0 894,044,707 0 2073 0 0 938,597,455 0 2074 0 0 985,308,355 0 2075 0 0 1,034,280,0 44 0 2076 0 0 1,085,619,9 87 0 2077 0 0 1,139,440,6 94 0 2078 0 0 1,195,859,9 65 0 2079 0 0 1,255,001,1 30 0 2080 0 0 1,316,993,3 09 0 2081 0 0 1,381,971,6 83 0 2082 0 0 1,450,077,7 74 0 2083 0 0 1,521,459,7 43 0 2084 0 0 1,596,272,7 00 0 2085 0 0 1,674,679,0 23 0 2086 0 0 1,756,848,7 04 0 2087 0 0 1,8 42,959,6 95 0

PAGE 262

263 Table C 23. I 95 TIFIA mandatory debt cash flow section Year RURA reserve account after paying senior d ebt Debt service reserve account after paying senior debt Net cash flow available after senior debt service Unpaid b alance on TIF IA m andatory Interest on unpaid balance TIFIA mandatory debt service Total c ompounded TIFIA mandatory debt service Amount actually paid to TIFIA RURA enhanced debt service ratio Cash flow available after TIFIA mandatory debt service 2012 2013 2014 2015 35,000,000 12,097,500 8,631,952 0 0 0 0 0 8,631,952 2016 35,000,000 12,097,500 26,793,728 0 0 0 0 0 26,793,728 2017 35,000,000 12,097,500 42,927,443 0 0 0 0 0 42,927,443 2018 35,000,000 12,097,500 48,225,296 0 0 0 0 0 48,225,296 2019 478,500 12,097,500 52,245,474 0 0 638,000 638,000 638,000 81.89 51,607,474 2020 949,500 12,097,500 54,162,801 0 0 1,266,000 1,266,000 1,266,000 42.78 52,896,801 2021 1,661,250 53,331,693 0 0 2,215,000 2,215,000 2,215,000 24.08 51,116,69 3 2022 2,373,750 57,974,569 0 0 3,165,000 3,165,000 3,165,000 18.32 54,809,569 2023 2,373,750 62,914,756 0 0 3,165,000 3,165,000 3,165,000 19.88 59,749,756 2024 2,373,750 68,170,178 0 0 3,165,000 3,165,000 3,165,000 21.54 65,005,178 2025 2,373,750 73,759,803 0 0 3,165,000 3,165,000 3,165,000 23.30 70,594,803 2026 2,373,750 78,423,769 0 0 3,165,000 3,165,000 3,165,000 24.78 75,258,769 2027 2,373,750 84,666,756 0 0 3,165,000 3,165,000 3,165,000 26.75 81,501,756 2028 2,373,750 88,476,154 0 0 3,16 5,000 3,165,000 3,165,000 27.95 85,311,154 2029 2,373,750 93,888,365 0 0 3,165,000 3,165,000 3,165,000 29.66 90,723,365 2030 2,373,750 91,144,732 0 0 3,165,000 3,165,000 3,165,000 28.80 87,979,732 2031 2,373,750 92,065,614 0 0 3,165,000 3,165,000 3,1 65,000 29.09 88,900,614 2032 2,373,750 96,340,005 0 0 3,165,000 3,165,000 3,165,000 30.44 93,175,005 2033 2,373,750 100,484,564 0 0 3,165,000 3,165,000 3,165,000 31.75 97,319,564 2034 2,373,750 101,085,502 0 0 3,165,000 3,165,000 3,165,000 31.94 97,9 20,502 2035 2,373,750 106,976,469 0 0 3,165,000 3,165,000 3,165,000 33.80 103,811,469

PAGE 263

264 Table C 23. Continued Year RURA reserve account after paying senior debt Debt service reserve account after paying senior debt Net cash flow available after senior de bt service Unpaid b alance on TIFIA m andatory Interest on unpaid balance TIFIA mandatory debt service Total c ompounded TIFIA mandatory debt service Amount actually paid to TIFIA RURA enhanced debt service ratio Cash flow available after TIFIA mandatory debt service 2036 2,373,750 114,523,793 0 0 3,165,000 3,165,000 3,165,000 36.18 111,358,793 2037 2,373,750 122,445,928 0 0 3,165,000 3,165,000 3,165,000 38.69 119,280,928 2038 2,373,750 130,760,157 0 0 3,165,000 3,165,000 3,165,000 41.31 127,595,157 203 9 2,373,750 137,193,999 0 0 3,165,000 3,165,000 3,165,000 43.35 134,028,999 2040 32,306,250 182,055,708 0 0 43,075,000 43,075,000 43,075,000 4.23 138,980,708 2041 32,306,250 191,565,631 0 0 43,075,000 43,075,000 43,075,000 4.45 148,490,631 2042 32,30 6,250 198,937,602 0 0 43,075,000 43,075,000 43,075,000 4.62 155,862,602 2043 32,306,250 209,292,524 0 0 43,075,000 43,075,000 43,075,000 4.86 166,217,524 2044 32,306,250 217,327,866 0 0 43,075,000 43,075,000 43,075,000 5.05 174,252,866 2045 32,306,25 0 231,568,681 0 0 43,075,000 43,075,000 43,075,000 5.38 188,493,681 2046 32,306,250 240,432,586 0 0 43,075,000 43,075,000 43,075,000 5.58 197,357,586 2047 32,306,250 249,629,045 0 0 43,075,000 43,075,000 43,075,000 5.80 206,554,045 2048 32,306,250 2 62,506,573 0 0 43,075,000 43,075,000 43,075,000 6.09 219,431,573 2049 32,281,500 276,019,581 0 0 43,042,000 43,042,000 43,042,000 6.41 232,977,581 2050 0 286,539,816 0 0 0 0 0 286,539,816 2051 0 316,541,552 0 0 0 0 0 316,541,552 2052 0 332,869,12 1 0 0 0 0 0 332,869,121 2053 0 350,004,980 0 0 0 0 0 350,004,980 2054 0 367,988,093 0 0 0 0 0 367,988,093 2055 0 386,859,273 0 0 0 0 0 386,859,273 2056 0 406,661,264 0 0 0 0 0 406,661,264 2057 0 427,438,837 0 0 0 0 0 427,438,837 2058 0 44 9,238,881 0 0 0 0 0 449,238,881 2059 0 472,110,503 0 0 0 0 0 472,110,503

PAGE 264

265 Table C 23. Continued Year RURA reserve account after paying senior debt Debt service reserve account after paying senior debt Net cash flow available after senior debt service Unpaid b alance on TIFIA m andatory Interest on unpaid balance TIFIA mandatory debt service Total c ompounded TIFIA mandatory debt service Amount actually paid to TIFIA RURA enhanced debt service ratio Cash flow available after TIFIA mandatory debt service 2060 0 496,105,133 0 0 0 0 0 496,105,133 2061 0 521,276,634 0 0 0 0 0 521,276,634 2062 0 547,681,410 0 0 0 0 0 547,681,410 2063 0 575,378,534 0 0 0 0 0 575,378,534 2064 0 604,429,865 0 0 0 0 0 604,429,865 2065 0 634,900,182 0 0 0 0 0 634,9 00,182 2066 0 666,857,321 0 0 0 0 0 666,857,321 2067 0 700,372,316 0 0 0 0 0 700,372,316 2068 0 735,519,550 0 0 0 0 0 735,519,550 2069 0 772,376,911 0 0 0 0 0 772,376,911 2070 0 811,025,956 0 0 0 0 0 811,025,956 2071 0 851,552,082 0 0 0 0 0 851,552,082 2072 0 894,044,707 0 0 0 0 0 894,044,707 2073 0 938,597,455 0 0 0 0 0 938,597,455 2074 0 985,308,355 0 0 0 0 0 985,308,355 2075 0 1,034,280,044 0 0 0 0 0 1,034,280,044 2076 0 1,085,619,987 0 0 0 0 0 1,085,619,987 2077 0 1,13 9,440,694 0 0 0 0 0 1,139,440,694 2078 0 1,195,859,965 0 0 0 0 0 1,195,859,965 2079 0 1,255,001,130 0 0 0 0 0 1,255,001,130 2080 0 1,316,993,309 0 0 0 0 0 1,316,993,309 2081 0 1,381,971,683 0 0 0 0 0 1,381,971,683 2082 0 1,450,077,774 0 0 0 0 0 1,450,077,774 2083 0 1,521,459,743 0 0 0 0 0 1,521,459,743

PAGE 265

266 Table C 23. Continued Year RURA reserve account after paying senior debt Debt service reserve account after paying senior debt Net cash flow available after senior debt service Unpaid b a lance on TIFIA m andatory Interest on unpaid balance TIFIA mandatory debt service Total c ompounded TIFIA mandatory debt service Amount actually paid to TIFIA RURA enhanced debt service ratio Cash flow available after TIFIA mandatory debt service 2084 0 1, 596,272,700 0 0 0 0 0 1,596,272,700 2085 0 1,674,679,023 0 0 0 0 0 1,674,679,023 2086 0 1,756,848,704 0 0 0 0 0 1,756,848,704 2087 0 1,842,959,695 0 0 0 0 0 1,842,959,695 Table C 24. I 95 reserve accounts cash flow section Year RURA after TIF I A mandatory debt service RURA necessary for next year DSCR RURA ending balance Cash flow available after RURA Debt service reserve account required balance Debt service reserve ending balance Cash flow available after RURA 2012 2013 20 14 2015 35,000,000 35,000,000 35,000,000 8,631,952 12,097,500 12,097,500 8,631,952 2016 35,000,000 35,000,000 35,000,000 26,793,728 12,097,500 12,097,500 26,793,728 2017 35,000,000 35,000,000 35,000,000 42,927,443 12,097,500 12,097,500 42 ,927,443 2018 35,000,000 478,500 478,500 82,746,796 12,097,500 12,097,500 82,746,796 2019 478,500 949,500 949,500 51,136,474 12,097,500 12,097,500 51,136,474 2020 949,500 1,661,250 1,661,250 52,185,051 0 12,097,500 64,282,551 2021 1,661,250 2 ,373,750 2,373,750 50,404,193 50,404,193 2022 2,373,750 2,373,750 2,373,750 54,809,569 54,809,569 2023 2,373,750 2,373,750 2,373,750 59,749,756 59,749,756 2024 2,373,750 2,373,750 2,373,750 65,005,178 65,005,178 2025 2,373,750 2,373 ,750 2,373,750 70,594,803 70,594,803 2026 2,373,750 2,373,750 2,373,750 75,258,769 75,258,769 2027 2,373,750 2,373,750 2,373,750 81,501,756 81,501,756

PAGE 266

267 Table C 24. Continued Year RURA after TIF IA mandatory debt service RURA necessary fo r next year DSCR RURA ending balance Cash flow available after RURA Debt service reserve account required balance Debt service reserve ending balance Cash flow available after RURA 2028 2,373,750 2,373,750 2,373,750 85,311,154 85,311,154 2029 2,373, 750 2,373,750 2,373,750 90,723,365 90,723,365 2030 2,373,750 2,373,750 2,373,750 87,979,732 87,979,732 2031 2,373,750 2,373,750 2,373,750 88,900,614 88,900,614 2032 2,373,750 2,373,750 2,373,750 93,175,005 93,175,005 2033 2,373,750 2,373,750 2,373,750 97,319,564 97,319,564 2034 2,373,750 2,373,750 2,373,750 97,920,502 97,920,502 2035 2,373,750 2,373,750 2,373,750 103,811,469 103,811,469 2036 2,373,750 2,373,750 2,373,750 111,358,793 111,358,793 2037 2,373,750 2,373,750 2,373,750 119,280,928 119,280,928 2038 2,373,750 2,373,750 2,373,750 127,595,157 127,595,157 2039 2,373,750 32,306,250 32,306,250 104,096,499 104,096,499 2040 32,306,250 32,306,250 32,306,250 138,980,708 138,980,708 2041 3 2,306,250 32,306,250 32,306,250 148,490,631 148,490,631 2042 32,306,250 32,306,250 32,306,250 155,862,602 155,862,602 2043 32,306,250 32,306,250 32,306,250 166,217,524 166,217,524 2044 32,306,250 32,306,250 32,306,250 174,252,866 174, 252,866 2045 32,306,250 32,306,250 32,306,250 188,493,681 188,493,681 2046 32,306,250 32,306,250 32,306,250 197,357,586 197,357,586 2047 32,306,250 32,306,250 32,306,250 206,554,045 206,554,045 2048 32,306,250 32,281,500 32,281,500 219, 456,323 219,456,323 2049 32,281,500 0 0 265,259,081 265,259,081 2050 0 0 0 286,539,816 286,539,816 2051 0 0 0 316,541,552 316,541,552 2052 0 0 0 332,869,121 332,869,121 2053 0 0 0 350,004,980 350,004,980 2054 0 0 0 367,9 88,093 367,988,093 2055 0 0 0 386,859,273 386,859,273

PAGE 267

268 Table C 24. Continued Year RURA after TIF IA mandatory debt service RURA necessary for next year DSCR RURA ending balance Cash flow available after RURA Debt service reserve account require d balance Debt service reserve ending balance Cash flow available after RURA 2056 0 0 0 406,661,264 406,661,264 2057 0 0 0 427,438,837 427,438,837 2058 0 0 0 449,238,881 449,238,881 2059 0 0 0 472,110,503 472,110,503 2060 0 0 0 496 ,105,133 496,105,133 2061 0 0 0 521,276,634 521,276,634 2062 0 0 0 547,681,410 547,681,410 2063 0 0 0 575,378,534 575,378,534 2064 0 0 0 604,429,865 604,429,865 2065 0 0 0 634,900,182 634,900,182 2066 0 0 0 666,857,321 666,857,321 2067 0 0 0 700,372,316 700,372,316 2068 0 0 0 735,519,550 735,519,550 2069 0 0 0 772,376,911 772,376,911 2070 0 0 0 811,025,956 811,025,956 2071 0 0 0 851,552,082 851,552,082 2072 0 0 0 894,044,707 894,044 ,707 2073 0 0 0 938,597,455 938,597,455 2074 0 0 0 985,308,355 985,308,355 2075 0 0 0 1,034,280,044 1,034,280,044 2076 0 0 0 1,085,619,987 1,085,619,987 2077 0 0 0 1,139,440,694 1,139,440,694 2078 0 0 0 1,195,859,965 1,1 95,859,965 2079 0 0 0 1,255,001,130 1,255,001,130 2080 0 0 0 1,316,993,309 1,316,993,309 2081 0 0 0 1,381,971,683 1,381,971,683 2082 0 0 0 1,450,077,774 1,450,077,774 2083 0 0 0 1,521,459,743 1,521,459,743

PAGE 268

269 Table C 24. Conti nued Year RURA after TIF IA mandatory debt service RURA necessary for next year DSCR RURA ending balance Cash flow available after RURA Debt service reserve account required balance Debt service reserve ending balance Cash flow available after RURA 2084 0 0 0 1,596,272,700 1,596,272,700 2085 0 0 0 1,674,679,023 1,674,679,023 2086 0 0 0 1,756,848,704 1,756,848,704 2087 0 0 0 1,842,959,695 1,842,959,695 Table C 25. I 95 TIFIA scheduled cash flow section Year unpaid balance on TIFIA s cheduled Interest on unpaid balance TIFIA required scheduled debt service T otal required TIFIA scheduled debt service Amount actually paid to TIFIA 2012 2013 2014 2015 0 0 0 0 0 2016 0 0 0 0 0 2017 0 0 0 0 0 2018 0 0 0 0 0 2019 0 0 5 ,743,000 5,743,000 5,743,000 2020 0 0 11,393,000 11,393,000 11,393,000 2021 0 0 10,443,000 10,443,000 10,443,000 2022 0 0 9,494,000 9,494,000 9,494,000 2023 0 0 9,494,000 9,494,000 9,494,000 2024 0 0 9,494,000 9,494,000 9,494,000 2025 0 0 9,494,000 9 ,494,000 9,494,000 2026 0 0 9,494,000 9,494,000 9,494,000 2027 0 0 9,494,000 9,494,000 9,494,000 2028 0 0 9,494,000 9,494,000 9,494,000 2029 0 0 9,494,000 9,494,000 9,494,000 2030 0 0 9,494,000 9,494,000 9,494,000 2031 0 0 9,494,000 9,494,000 9,494,0 00

PAGE 269

270 Table C 25. Continued Year unpaid balance on TIFIA s cheduled Interest on unpaid balance TIFIA required scheduled debt service T otal required TIFIA scheduled debt service Amount actually paid to TIFIA 2032 0 0 9,494,000 9,494,000 9,494,000 2033 0 0 9 ,494,000 9,494,000 9,494,000 2034 0 0 9,494,000 9,494,000 9,494,000 2035 0 0 9,494,000 9,494,000 9,494,000 2036 0 0 9,494,000 9,494,000 9,494,000 2037 0 0 9,494,000 9,494,000 9,494,000 2038 0 0 9,494,000 9,494,000 9,494,000 2039 0 0 9,494,000 9,494,0 00 9,494,000 2040 0 0 0 0 0 2041 0 0 0 0 0 2042 0 0 0 0 0 2043 0 0 0 0 0 2044 0 0 0 0 0 2045 0 0 0 0 0 2046 0 0 0 0 0 2047 0 0 0 0 0 2048 0 0 0 0 0 2049 0 0 0 0 0 2050 0 0 0 0 0 2051 0 0 0 0 0 2052 0 0 0 0 0 2053 0 0 0 0 0 2054 0 0 0 0 0 20 55 0 0 0 0 0 2056 0 0 0 0 0 2057 0 0 0 0 0 2058 0 0 0 0 0 2059 0 0 0 0 0

PAGE 270

271 Table C 25. Continued Year unpaid balance on TIFIA s cheduled Interest on unpaid balance TIFIA required scheduled debt service T otal required TIFIA scheduled debt service Amount actually paid to TIFIA 2060 0 0 0 0 0 2061 0 0 0 0 0 2062 0 0 0 0 0 2063 0 0 0 0 0 2064 0 0 0 0 0 2065 0 0 0 0 0 2066 0 0 0 0 0 2067 0 0 0 0 0 2068 0 0 0 0 0 2069 0 0 0 0 0 2070 0 0 0 0 0 2071 0 0 0 0 0 2072 0 0 0 0 0 2073 0 0 0 0 0 2074 0 0 0 0 0 2075 0 0 0 0 0 2076 0 0 0 0 0 2077 0 0 0 0 0 2078 0 0 0 0 0 2079 0 0 0 0 0 2080 0 0 0 0 0 2081 0 0 0 0 0 2082 0 0 0 0 0 2083 0 0 0 0 0 2084 0 0 0 0 0 2085 0 0 0 0 0 2086 0 0 0 0 0 2087 0 0 0 0 0

PAGE 271

272 APPENDIX D DEFAULT RATE CALCULATIONS O N RATED DEBT The U.S. Municipal Bond Rating Scale: Mapping to the Global Rating Scale And Assigning Global Scale Ratings to Municipal Obligations, from 1970 2006 the ten year cumulative default rate on Baa rated munici pal bonds was 0.1349%. for Baa3 (Baa is inclusive of B aa1 Baa 2 and Baa 3), is more accurate to account for certain economic factors (Moody's, 2006) Note: Figure2 9 pro vides proposed idealized cumulative probability of default (PD) rates for municipal debt obligations, by rating category. For instance, municipal obligations rated in the Ba1 category have an idealized PD of 3.6761% within ten years of the initial rating date Figure D Year Cumulative Probability of Default Rates To calculate the implied probability of default on a Baa3 rated bond, the suggested ten year cumulative default rate was divided by ten to give an annualized defa ult rate. This results in an annualized probability of default of 0.24507%. The PAB term, which is normally 30 years, is multiplying by the annualized default rate, 0.24507%, and results in a 7.3521% probability of default over 30 years. This was done f or the PAB and TIFIA loan for each case study.

PAGE 272

273 The idealized probability of default on each project as used in Chapter 5, question 1 and 2 are shown in Table D 1. Table D 1. Probability of d efault on PABs Project Debt t erm (years) Recommended BBB defaul t per year Recommended BBB default risk I 495 40 0.012% 0.49% LBJ 28 0.245% 6.86% NTE 30 0.245% 7.35% ERC 30 0.245% 7.35% I 95 30 0.245% 7.35%

PAGE 273

274 APPENDIX E EFFECTS OF OPTIMISM BIAS ON SELECTED CASES Figure E 1. Effects of optimism bias on I 495 e quity IRR (%) Figure E 2. Effects of optimism bias on LBJ e quity IRR (%)

PAGE 274

275 Figure E 3. Effects of optimism bias on NTE e quity IRR (%) Figure E 4. Effects of optimism bias on ERC e quity IRR (%)

PAGE 275

276 Figure E 5. Effects of optimism bias on I 95 e quity IRR (%)

PAGE 276

277 LIST OF REFERENCES Akintoye, A., Edwards, P., Hardcastle, C., & Li, B. (2005). Perceptions of Positive and Negative Factors Influencing the Attractiveness of PPP/PFI Procurement for Construction Projects in the UK. Engineering, Constructi on and Architectural Management, 12 (2), 125 148. ASCE. (2013). Report Card for America's Infrastructure. Washington, D.C: American Socitey of Civil Engineers. Bain, R. (2009). Error and Optimism Bias in Toll Road Traffic Forecasts. London: Springer Ba in, R. (2009). Review of Lessons from Completed PPP Projects Financed by the EIB. London: European Investment Bank. Canadian Council for Public Private Partnerships. (2001). About PPP Retrieved Jan 28, 2013, from Canadian Council for Public Private Partn erships: www.pppcouncil.ca Cervero, R. (2004). TCRP Report 102: Transit Oriented Development in the United States. Washington, D.C.: Transportation Research Board. Chan, A., Lam P., Chan D., & Cheung, E. (2008). Risk Sharing Mechanism for PPP Project s the Case Study of the Sydney Cross City Tunnel. Surveying and Built Environment 67 80. Cohen I., Frieling, T., & Robinson, E. (2012). The Economic Impact and Financing of Infrastructure Spending. Williamsburg, VA: Associated Equipment Distributors (AED). Congressional Budget Office. (2010). Public Spending on Transportation and Water Infrastructure. Washington, D.C.: Congressional Budget Office. Endel, E., Fischer, R., & Galetovic, A. (2010). The Economics of Infrastructure Finance: Public Private Partnerships versus Public Provision. European Investment Bank 41 69. Etsy, B. (2004). Modern Project Finance: A Casebook by Benjamin Etsy. Hoboken : John Wiley and Sons. European PPP Expertise Centre. (2013). PPP Project Finance Retrieved July 10, 20 13, from European Investment Bank: www.eib.org Eurpoean PPP Expertise Centre. (2010). Capital markets in PPP financing: Where we were and where are we going? London: European International Bank. Farrell, L. (2003). Principal Agency Risk in Project Financ e. International Journal of Project Management 547 561.

PAGE 277

278 Federal Highway Administration. (2013, October 17). Gas Tax Retrieved November 11, 2013, from Federal Highway Administration : www.fhwa.dot.gov FHWA. (2005). Manual for using Public private Partne rships on Highway Projects. Washington, D.C.: U.S. Department of Transportation. FHWA. (2005). Safe, Accountable, Flexible, Efficient Transportation Equity Act:A Legacy for Users, A Summary of Highway Provisions. Washington, D.C.: Federal Highway Administ ration. FHWA. (2007). US PPP Case Studies Final Report Washington, D.C. : Federal Highway Administration FHWA. (2013, Jan 01). Project Profiles Retrieved Feb 21, 2013, from Federal Highway Administration: www.fhwa.dot.gov Finnerty, J. D. (2007). Pr oject Financing: Asset Based Financial Engineering. Hoboken: John Wiley and Sons Ltd. Flyvbjerg, B., Holm, M., & Buhl, S. L. (Spring 2005 ). How (In)accurate Are Demand Forecasts in Public Works Project Journal of the American Planning Association 131 146. GAO. (2012). HIGHWAY TRUST FUND: Pilot Program Could Help Determine the Viability of Mileage Fees for Certain Vehicles. Washington, D.C.: United States Government Accountability Office. Gilroy L., & Kenny, H. (2013). Annual Privatization Report 20 13. Los Angeles: Reason Foundation. Grimsey, D., & Lewis, M. (2004). Public Private Partnerships: The World wide Revolution in Infrastructure Provision and Project Finance. Cheltenham, United Kingdom: Edward Elgar Publishing. Hoffman, S. L. (2001). The L aw and Business of International Project Finance: A Resource for Governments, Sponsors, Lawyers, and Project Participants (2nd edition). Boston: Cambridge Jeffrey A. Parker & Associates, Inc. (2009 ). I 595 Corridor Roadway Improvements Value for Money Analysis. Tallahassee: Florida Department of Transportation Retrieved from www.transportation finance.org Jeffrey A. Parker & Associates, Inc. (2010). The Port of Miami Tunnel and Access Improvement Project: A Value for Money Analysis. Tallahassee: Flor ida Department of Transportation. Kessler, F., & Denton, P. (2012, July 6). MAP 21: Surface Transportation Reauthorization Ushers in Significant Changes to TIFIA. Nossaman Infra Insight pp. 1 6.

PAGE 278

279 Kwak, Y. H., Chih, Y. Y., & Ibbs, C. W. (2009). Towards a Comprehensive Understanding of Public Private Partnerships for Infrastructure Develoment. California management Review, 51 (2), 51 77. Loosemore, M., Raftery, J., Reilly, C., & Higgon, D. (2006). Risk Management in Projects. London: Taylor & Francis. Mood y's (2007). The U.S. Municipal Bond Rating Scale: Mapping to the Global Rating Scale And Assigning Global Scale Ratings to Municipal Obligations. New York: Moody's Investors Service, Inc. Moody's. (2006). Stable Outlook for U.S. State and Local Govern ment: Toll Facilities Strong Trend in Toll Revenues Fueled by Economic Growth and Growth in Capital Needs. New York: Moody's Moody's. (2012). U.S. Municipal Bond Defaults and Recoveries. New York: Moody's Investory Service, Inc. Moody's. (2013, May 7). Announcement: Moody's: Municipal bond defaults have increased since financial crisis, but numbers remain low Retrieved November 14, 2013 from Moody's Investory Service : www.moodys.com Moody's Investor's Service. (2006). Rating Methodology: Operationa l Toll Roads. London: Moody's Investor's Service, Inc. National Chamber Foundation. (2008). The Transportation Challenge Moving the U.S. Economy. Washington, D.C.: National Chamber Foundation. NCHRP. (2006). National Cooperative Highway Research Program: Estimating Toll Road Demand and Revenue. Washington, D.C.: Transportation Research Board. Ng, A., & Loosemore, M. (2007). Risk Allocation in the Private Provisions of Public Infrastructure. International Journal of Project Management 66 67. PriceWaterh ouseCoopers. (2005). Delivering the PPP Promise: A review of PPP issues and activities. New York City : PriceWaterhouseCoopers. Public works Finance. (2013). U.S. Transportation DBFOM Concessions, 1993 2012. New York: Public Works Finance. Public Private Infrastructure Advisory Facility. (2009). Toolkitfor for Public Private Partnerships in Roads and Highways. London: Public Private Infrastructure Advisory Facility.

PAGE 279

280 Rall, J., Reed, J., & Farber N. (2010). Public Private Partnerships for Transportation: A Toolkit for Legislators. Washington, D.C.: National Conference of State Legislatures. Reinhardt, W. (2012, Octobe 1). Public Works Financing Company Profiles: Experienced P3 Develpers Flock to the U.S. pp. 1 7. Rockart, J. F. (1982). The Changing Role of the Innovation Systems Executive: A Critical Success Factors Perspective. MIT Sloan Management Review, 24 (1), 3 13. Senate Finance Committee. (2013, April 25). INFRASTRUCTURE, ENERGY, AND NATURAL RESOURCES: Senate Finance Committee Staff Tax Refor m Options for Discussion. Washington, D.C.: The United States Senate Committee on Finance. Singh, L. K. (2006). Traffic Revenue Risk Management through Annuity Model of PPP Road Projects in India. International Journal of Project Management 605 603. Sta ndard & Poor's. (2002). Traffic Risk in Start Up Toll Facilities. London: McGraw Hill. The World Bank. (2008). Success and Failures of PPP Projects Warsaw: The World Bank TIFIA. (1998). Factsheet: Transportation Infrastructure Finance and Innovation Act. Washington D.C.: TIFIA. Retrieved from CA.GOV: www.ca.gov Tiong, R. (1996). CSFs in competative tendering and negotiation model for BOT projects. Journal of Construction Engineering and Managment 122. TRIP. (2012). Key Facts About America's Surface Transportation System and Federal Funding. Washington, D.C.: TRIP. Retrieved from www.tripnet.org US DOT. (2013, July 9). Transportation Infrastructure Finance and Innovation Act (TIFIA) Program Retrieved July 9, 2013, from U.S Department of Transportat ion: www.fta.dot.gov World Bank. (2011). Transportation Sector Data Snapshots 1984 2011. Washington, D.C.: Private Participation in Infrastructure Projects Database. Retrieved from ppi.worldbank.org World Bank Institute. (2012). Public Private Partnershi ps Reference Guide. Washington DC: The World Bank. Retrieved from www.ppiaf.org Yescombe, E. R. (2002). Principles of Project Finance. New York: Academic Press. Zhang, X. (2004). Improving concessionaire selection protocols in public/private partnered in frastructure projects. Journal of Construction Engineering Management 670 679.

PAGE 280

28 1 private partnered Journal of Construction Engineering Management 351 358.

PAGE 281

282 BIOGRAPHICAL SKETCH Kyle Fisher, a native of Hamilton, Ohio, earned his Bachelor of Science degree in civil e ngineering from Purdue University in 2003. After becoming an engineer Dr. Fisher relocated to Bradenton, Florida and worked as a civil engineer and project ma nager in land development. In addition to being a civil engineer, Dr. Fisher is a licensed Florida real estate broker and a Florida certified building contractor. He has started companies in both real estate and construction. In 2008, Dr. Fisher earned h is Master of Science degree in civil e ngineering from the University of Florida through the Electronic Delivery of Graduate Engineering (EDGE) program while working full time as an engineer. In 2009, he returned to full time academic studies at the Unive rsity of Florida as an Alumni Fellow, the highest graduate level award given at the University of Florida. In addition to his doctoral studies in civil engineering, Dr. Fisher earned an MBA from the University of Florida, Hough Graduate School of Business. Dr. Fisher has passed the Principles and Practice of Engineering exam and will be licensed as a Professional Engineer upon completion of his doctoral degree. For the past three years Dr. Fisher has taught CGN 2002, Introduction to Civil Engineering, at t he University of Florida. He enjoys mentoring civil engineering students and is a professional mentor to the UF chapter of Engineers with Borders.


xml version 1.0 encoding UTF-8
REPORT xmlns http:www.fcla.edudlsmddaitss xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.fcla.edudlsmddaitssdaitssReport.xsd
INGEST IEID E3896OBL5_BAD21W INGEST_TIME 2014-05-13T20:25:19Z PACKAGE UFE0045422_00001
AGREEMENT_INFO ACCOUNT UF PROJECT UFDC
FILES