SCENARIO-BASED CONTRACTS: USING SCENARIO PLANNING AND VALUATION METHODS TO MANAGE RISK IN ALTERNATIVE DELIVERY METHODS. Adjo A. Amekudzi, Ph.D.

Transcription

1 SCENARIO-BASED CONTRACTS: USING SCENARIO PLANNING AND VALUATION METHODS TO MANAGE RISK IN ALTERNATIVE DELIVERY METHODS Adjo A. Amekudzi, Ph.D. Associate Professor, School of Civil and Environmental Engineering, Georgia Institute of Technology, Atlanta, GA 30332, Tel: , Fax: , Christy Mihyeon Jeon Graduate Research Assistant, School of Civil and Environmental Engineering, Georgia Institute of Technology, Atlanta, GA 30332, Tel: , Fax: , Total Word Count: 4,744 Submission Date: July 8, 2005

2 Amekudzi and Jeon 2 ABSTRACT In both developing and developed economies, several governments are considering alternative delivery methods such as design-build (DB), build-operate-transfer (BOT), and build-ownoperate-transfer (BOOT) schemes for multibillion-dollar mega infrastructure projects. These schemes come with various political, financial, and legal risks in both socioeconomic settings. This paper discusses sources of risk in the Channel Tunnel project, one of the most notable megaprojects in the world, and introduces applications of scenario planning and valuation methodologies to manage selected risks in BOT and BOOT schemes by creating scenario-based or dynamic contracts. Such contracts can insulate better the contractor and create enable host governments to provide more favorable conditions for infrastructure systems delivery. KEY WORDS Alternative Delivery Methods, Risk Management, Asset Management, Scenario Planning, Channel Tunnel

3 Amekudzi and Jeon 3 INTRODUCTION In both developing and developed economies, several governments are considering alternative delivery methods such as design-build (DB), build-operate-transfer (BOT), and build-ownoperate-transfer (BOOT) schemes for multibillion-dollar mega infrastructure projects. In the United States, the design-build (DB) method of contracting has been increasing steadily: the volume of domestic DB contracts has grown from $6 billion to $56 billion since 1982 and now represents 23 percent of the non-residential U.S. market (1). Also, highways, tunnels, and light rail systems are increasingly being built through build-operate-transfer (BOT) concession agreements in which private companies construct and collect tolls on new infrastructure for a mutually agreed-upon period of time (2). These schemes, DB, BOT, and BOOT, come with various political, financial, and legal risks in both socioeconomic settings. This paper discusses sources of risk in the Channel Tunnel project, one of the most notable megaprojects in the world, and introduces applications of scenario planning and valuation methodologies to manage selected risks in BOT and BOOT schemes by creating scenario-based or dynamic contracts. Such contracts can insulate better the contractor and enable host governments to provide more favorable conditions for infrastructure systems delivery. METHODOLOGY The intent of this paper is to demonstrate how scenario planning methodology coupled with valuation techniques can be used to manage risks associated with large-scale civil infrastructure projects that are procured through alternative delivery methods. In recent decades, several megaprojects have been completed all over the world. Megaprojects are projects worth several billions of dollars that typically have significant cost overruns and lower-than-predicted revenues, e.g. the Channel Tunnel that connects the United Kingdom to the European mainland, the Boston Central Artery Tunnel (CA/T) Project, Mexico City metro, and Malaysia s North-South Expressway (3). Table 1 shows examples of megaprojects and their associated construction cost overruns and lower-than-expected revenues. Some of these projects have been procured using alternative delivery methods. For example, as shown in Table 1, the Channel Tunnel was a BOOT project and the Malaysia s North-South Highway used a BOT scheme to finance them through private-sector involvement (3, 4). In a BOT structure, entrepreneurs form consortia to make substantial equity investments in a special-purpose project company, which then builds the facility, operates it long enough to pay back the debt finance, and eventually transfers rights and responsibilities of further operations of the facility to the government. On the other hand, a BOOT scheme refers to situations where the project company also owns the facility throughout the period (3). Also, the DB method, in which the agency or owner holds a single contract with a single entity for both the design and construction of a project, has been applied to mass transit rail projects, e.g., the Alameda Corridor rail project in Los Angeles and the Hudson Bergen light rail in New Jersey (5, 6). Megaprojects have various risks including political, financial, cultural, and legal risks. The United Nations Industrial Development Organization (UNIDO) identified two major categories of primary risks for large-scale BOT projects: (1) general risks related to the macroenvironmental factors of the host country and (2) project-specific risks controlled by the project participants (7). General risks can be categorized as (1) political risks, e.g., sovereign risks and instability risks; (2) financial risks, e.g., currency devaluation, foreign exchange fluctuation,

4 Amekudzi and Jeon 4 fluctuation of interest rates, and inflation; and (3) legal risks, e.g., the host country s changes in the legal system such as taxation and regulations. Project-specific risks can be classified by specific risks pertaining to the different project stages: development, construction, and operation. Delays in project completion and cost overruns, caused by technical difficulties, poor management, or a combination of both, are general types of the construction stage risks. On the other hand, increased cost of operation and insufficient revenue from the completed project can be identified as common types of operation stage risks (7). Malaysia s North-South Highway project, for example, has faced six major sources of risks, including foreign exchange and interest risk, lack of equity, cost overruns, completion delays, force majeure, and revenue risk (4). Such risks contribute to the significant cost overruns that have been experienced by several megaprojects in the past several decades. For megaprojects that are procured using alternative delivery methods, opportunities exist for managing these risks using scenario-based methodologies combined with valuation techniques to explore the impacts of existing uncertainties or risk factors (for example, project delivery time, projected demand for the project, inflation etc.) on the project outcome. In essence, for each risk factor, plausible scenarios can be developed to capture the boundary conditions that might occur (e.g., on-time project delivery versus delayed delivery, high project demand versus low demand, and high inflation versus low inflation), and valuation methods can be applied to assess the impacts of these conditions on the project value. In particular, for BOT or BOOT projects, such scenarios can be used to evaluate feasible times for recouping project investments (i.e., the operation period) before the facility is transferred to the owner. The operate period would be longer in less favorable conditions i.e. conditions with project delays, low system demand and high inflation, for example. With a better handle on the impacts of these risk factors on the project value, contracts can be written to address different plausible scenarios that might reasonably occur -- and the scenarios that most closely match the actual outcome of the project can then be called into effect to protect both the contractor and government owner of the project. In the sections below, the proposed scenario/valuation methodology is applied to the Channel Tunnel project. Using the Channel Tunnel project as a case study, major system risks are identified, and plausible boundary conditions for these risk factors (e.g., time for project delivery, system demand, and inflation) are determined. These boundary conditions are then used to craft plausible scenarios for the system that capture plausibly favorable and nonfavorable conditions for the contractor and owner. The value of the system over its life cycle is then estimated based on these plausible scenarios to determine several likely values that the project might assume during its life cycle based on the prevailing risks. A discussion follows explaining how these values can be used to determine realistic operation times in the BOT or BOOT contract for the various plausible scenarios -- both to protect the contractor and create more favorable conditions for the government owner to attract contractors for such projects in the future. CASE STUDY As more multibillion-dollar mega infrastructure projects are being proposed and constructed around the world, it has become clearer that such projects tend to show noticeably poor performance records with respect to economy, environment, and public support (2). These projects are easily exposed to various risks such as project delays, inflation, cost overruns, and lower demand or revenues than predicted. For instance, the Channel Tunnel, opened in 1994 at a

5 Amekudzi and Jeon 5 construction cost of $12.6 billion (1997 prices converted to US$ at $1.59 to 1 British Pound -- March 20, 1997), has faced several near-bankruptcies resulting from construction cost overruns of 80 percent due to financing costs, 140 percent higher than what was predicted; and revenues less than half of those forecast (3, 8). This section discusses those risk factors and applies valuation methodology to estimate the difference in facility value based on the projected and actual conditions and characteristics of the Channel Tunnel project. Project Background The Channel Tunnel, also known as the Chunnel and Eurotunnel, opened in 1994, and is the longest underwater rail tunnel in Europe connecting France and the UK (3). This project is the flagship of private financing in the transport sector and was a build-own-operate-transfer (BOOT) project with a concession, originally lasting fifty-five years, later extended to sixty-five years as part of a settlement regarding responsibility for cost overruns, and later again extended to ninety-nine years in an attempt to secure project viability (3). As the concessionaire, Eurotunnel manages the infrastructure of the Channel Tunnel, operates accompanied truck shuttle and passenger shuttle services between UK and France, and earns toll revenue from other train operators, such as Eurostar, which use the Tunnel (9). The total investment costs for the project were predicted at $7.1 billion (1997 prices) in 1987, but actual costs turned out to be $12.6 billion (1997 prices) resulting in a cost overrun of 80 percent (3). Moreover, the actual financing cost turned out to be $ 18.6 billion (1997 prices), 140 percent higher than the forecast $7.8 billion (1997 prices) (3, 8). Traffic forecasting for the Channel Tunnel done at the time of the decision to construct the tunnel estimated 15.9 million passengers on the Eurostar trains in the opening year while actual traffic in 1995, the first full year of operations, was 2.9 million passengers, only 18 percent of the number of passengers predicted. In 2001, after more than six years of operations, the number of passengers had grown to 6.9 million, which was still 43 percent of the original estimate for the opening year. Rail freight traffic was projected at 7.2 million gross tonnes for the opening year while actual traffic was 1.3 million gross tonnes in 1995, 18 percent of the freight projected, and had grown to 2.4 million tonnes, 33 percent of the freight projected in 2001 (3). Table 2 shows the conspicuous differences between predicted and actual project delivery timeline, travel demand, investment/financing costs, and operating profit. Valuation Based on the project records presented in Table 2, this section analyzes the economic value of the project focusing on the differences between the envisaged and actual worth of the project. The net present value (NPV) of the project, the traditional method for quantifying the financial attractiveness of investment, was selected as the value criterion and calculated from the sum of the present values (today's dollars) of the annual cash flows minus the initial investment. These annual cash flows, discounted or adjusted by integrating uncertainty and time value of money, are the net benefits estimated from revenues minus costs (10). The predicted and actual NPVs for the Channel Tunnel project were generated by following three steps (10). First, the size and timing of the expected future cash flows were identified since we already know the projected and actual investment cost of the project ($7.1 billion versus $12.6 billion). The life span of this project is considered to be 120 years because

6 Amekudzi and Jeon 6 the Tunnel lining was designed for a 120-year life, according to the Channel Tunnel construction facts, which means that no significant deterioration in performance is expected to occur over this period (11). After opening a year late in May 1994, its first and second year of operation produced a loss of $2.1 billion (12). Operating profit of the project, calculated by subtracting operating expenditure from total turnover, has been annually reported as approximately $300 million per year since Cash flow without actual records, including operating profit for the years and , is assumed to have a regular cycle and repeat the same ups and downs of the operating profit for the 7-year period of operations from 1998 through Based on the available actual cash flow, the operating profit that might have been predicted before the project was started was also estimated based on the ratio of the actual demand to predicted demand for Eurostar, the passenger rail system that uses the Tunnel. Second, the discount rate or the estimated rate of return for the project was determined to convert a future amount of money to its equivalent present value. In 2004, Eurotunnel reported an implicit discount rate of 7.2% determined in accordance with the existing standard on the basis of the Eurotunnel constituting a single income generating unit and using the adjusted present value (APV) methodology (13). The real discount rate of 7.2% per year was adjusted with an average inflation rate of 2.75%, for the period from (14), and thus 10.15% was considered an appropriate inflation-adjusted interest rate for calculating NPVs of this project (15). Third and lastly, the NPVs based on predicted and actual project records were calculated using the generic NPV equation: n n Bt Ct Operating Pr ofit NPV = ( ) = C + t 0 t t= 0 (1 + r) t= 1 (1 + r) where B = benefit, C = cost, r = discount rate, t = 0,1,, n, n = life span of the project (n=120), C 0 = investment cost. Predicted and Actual NPVs of the Channel Tunnel Table 3 reveals the differences between predicted and actual NPVs of the Channel Tunnel project based on predicted and actual cash flows. Cash flows are assumed to repeat the same upand-down cycles of the operating profit for the 7-year period from 1998 through The table shows cash flows for the first seventeen years of facility operation through While the predicted NPV of the project is negative $200.2 million, the actual NPV turns out to reach the value of negative $11.8 billion, about sixty times lower than the projected value, due to significant construction cost overruns (80%) and lower-than-expected revenues. Therefore, we can conclude that the investment for the project can never be recouped during the 120-year project life cycle and the project will not possibly attain the status of economic sustainability with the expected cash flows. It should be noted that we could only quantify benefits that accrue from the operation of the Tunnel excluding any other social or environmental benefits such as higher quality of life, better air quality from decreased usage of other modes of transportation,

7 Amekudzi and Jeon 7 etc. Figure 1 graphically captures all the project costs and revenues, e.g., initial construction cost and annual operating costs, annual revenues, and operating profit of the project. Sensitivity Analysis When conducting cost-benefit analysis, sensitivity analysis is generally performed to identify those parameters that are both uncertain and for which the project decision is sensitive (16). This section applies sensitivity analysis to capture the impacts of various risk factors of the Channel Tunnel project on the value of the project. Using plausible bounds for the projected risk parameters, several plausible scenarios are developed. The project value (life cycle) for each plausible scenario is then estimated. As presented earlier, the Channel Tunnel experienced a one-year delay for project delivery, severe construction cost overruns, and lower-than-predicted revenues. In addition, the inflation rate is a classic type of the financial risk, in particular, for longer and large-scale projects such as the Channel Tunnel. Sensitivity and risk analysis are conducted following three steps using the base case of the actual NPV of the project. First, plausible boundary conditions based on those risk factors, (i.e., time for project delivery, travel demand, and inflation), are identified and the resulting scenarios are used to depict favorable and non-favorable conditions that the project may face. Second, the value (NPV) of the system over its life cycle is calculated based on these likely scenarios, to determine several plausible values that the project might assume based on the prevailing risks. Finally, a discussion is presented to explain how these values might be used to determine realistic operation times in the BOT or BOOT contract for the various plausible scenarios that might occur, both to protect the contractor and create more favorable conditions for the owner to attract contractors for such projects in the future. Table 4 presents the impacts of several plausible conditions on the value of the project. These boundary conditions are determined based on selected risk factors as follows: (1) delayed project delivery (two years) versus earlier delivery (one year); (2) low demand (only 20 percent of what was originally predicted) versus high demand (reaching the predicted demand in the first year of operation), and (3) high inflation (5% per year) versus low inflation (1% per year). Meanwhile, the base scenario for the comparative analysis is determined based on the actual conditions of the Channel Tunnel. Generally speaking, sensitivity analysis is performed when the project is in the planning stage to help effective project decisions by incorporating possible uncertainty factors in the project analysis. This paper, however, focuses on an analysis based on actual project records, e.g., project delivery, demand, inflation rate, etc., and presents plausible scenarios around the actual project conditions to assess how prevailing risks might have caused the project outcomes to vary. Thus, we set the actual situation of the Channel Tunnel: one year of delayed project delivery, 43 percent of the demand originally predicted, and an inflation rate of 2.75%, as the base scenario. In the case of the project timeline, the value of the project substantially decreases, i.e., -$15,088M down from -$11,836M (base case): corresponding to a 27.5% decrease. This is because if the project were completed two years later, additional construction costs for financing extra two years, land acquisition fees, and engineering fees for two extra years would be added. In addition, two years of revenue and positive social benefit would be lost. On the other hand, the NPV of the project would significantly increase, i.e., - $9,765M up from -$11,836M (base case) corresponding to a 17.5% increase, in the case where the project was completed one year early, i.e., if the project was finished just in time in If system demand turned out to be only 20 percent of the predicted demand, the value of the project

8 Amekudzi and Jeon 8 would decrease notably, i.e., -$13,260M down from -$11,836M (base case) corresponding to a 12% decrease, because of much lower revenues. When system demand reaches the predicted demand, the NPV of the project considerably improves: -$7,653M corresponding to a 35.3% increase relative to the base case. While the NVP of the project decreases by 14% in the case where the annual inflation rate is as high as 5 percent, the value of the project increases notably by 19.4% when the inflation rate is as low as 1 percent. As shown in Table 5, the NPV of the project ranges significantly from 38% less than the actual value to 60% above the actual value if the lower and upper limits respectively of project delivery time, project demand, and inflation are used to craft plausible boundary scenarios. Part of the intent of BOT and BOOT contracts is to determine feasible time frames for the concession holder or contractor to operate the project, in order to recoup the amounts invested into the project, before turning over the project to the government (owner). It is notable that subsequent to the opening of the Channel Tunnel project, the operation period in the BOOT contract has been revised twice to enable the project to achieve economic sustainability, as described in the Project Background section above. Such revisions have caused negative publicity for the project in general, indicating that project finances have not been managed effectively, and calling into question the economic sustainability of the project. The scenario/valuation analysis conducted in this paper is the type of analysis that could be used during the planning stages of a megaproject, such as the Channel Tunnel, to determine alternative plausible values that the project might assume, given prevailing risks. Such alternative values could then be used to calculate feasible operating times that would allow the contractor to recoup their investments made in the project, in a range of favorable to non-favorable conditions. The BOOT contract would thus contain several plausible operation times that correspond to plausible prevailing conditions that could occur after the project was opened for operation. The contractor would then be allowed to select the operation time most closely corresponding with actual project operating conditions -- so that the contractor might recoup the original investment without the need for revising the operating period after the fact (which causes negative publicity and detracts from a positive image for megaprojects). In essence, what is being proposed here is a scenariobased contract of sorts that proactively considers the most dominant risk factors influencing project value, and builds language into the contract to manage such risks. Such contracts would enable contractors to feel more secure in undertaking BOT and BOOT type projects that are laden with uncertainty. Such contracts would also help foster a more positive image for governments in that they would be willing to consider the existing uncertainties in megaprojects up front and proactively accommodate their resulting risks in the BOT or BOOT contract. LIMITATION OF STUDY This study introduces the application scenario planning combined with valuation methodologies to manage major risks of large-scale civil infrastructure projects that are procured using BOT or BOOT schemes. Several assumptions are made to assess the value of the Channel Tunnel. First, the future cash flow for the project is expected to have a regular cycle and repeat the same ups and downs with respect to operations for the 7-year period of operations from 1998 through Second, the operating revenues that would have been projected before the project was started are estimated based on the ratio of the actual demand to the predicted demand for Eurostar. Third and lastly, it should be noted that we could only quantify benefits that come from the operation

9 Amekudzi and Jeon 9 of the Tunnel excluding any other social or environmental benefits such as a higher quality of life, and better air quality from decreased usage of other modes of transportation, etc. SUMMARY In both developing and developed economies, several governments are considering alternative delivery methods such as design-build (DB), build-operate-transfer (BOT), and build-ownoperate-transfer (BOOT) schemes for multibillion-dollar mega infrastructure projects. These schemes come with various political, financial, and legal risks in both socioeconomic settings. The intent of this paper is to demonstrate how scenario planning methodology can be combined with valuation techniques to manage risks associated with a large-scale civil infrastructure projects procured through alternative delivery methods. The Channel Tunnel is used as a case study. Plausible scenarios are developed to capture the boundary conditions that might occur based on various risk factors, (e.g., on-time project delivery versus delayed delivery, high demand versus low demand, and high inflation versus low inflation), and valuation methods are applied to assess the impacts of these risks on the project value, the net present value (NPV) of the project. In particular, for BOT or BOOT projects, such scenarios can be used to evaluate feasible times for recouping project investments (i.e., the operation period) before the facility is transferred to the owner. The case study analysis shows that the operate period would be longer in less favorable conditions i.e. conditions with project delays, low system demand, and high inflation, for example. With a better handle on the impacts of these risk factors on the project value, contracts can be written to address different plausible scenarios that might reasonably occur -- and the scenarios that most closely match the actual outcome of the project can be called into effect to protect both the contractor and government owner of the project. Such scenario-based or dynamic contracts can insulate the contractor better and allow host governments to provide more favorable conditions for infrastructure systems delivery. ACKNOWLEDGEMENT This work was supported by the National Science Foundation (NSF) under Grant No : Application of Portfolio Theory and Sustainability Metrics to Civil Infrastructure Management. Any opinions, findings, and conclusions or recommendations expressed in this material are those of the authors and do not necessarily reflect the views of the NSF.

10 Amekudzi and Jeon 10 REFERENCES [1] ASCE Design-Build Contracting, Continuing Education Series Pamphlet, Accessed in July [2] Institute for Transportation and Development Policy, Privatization: Will It Lead to Sustainable Transport? A Paper Prepared for the Conference on Transport and Environment, Manila, the Philippines, December 10-12, [3] Flyvbjerg, B., Bruzelius, N., Rothengatter, W., Megaprojects and Risks: An Anatomy of Ambition, Cambridge Univ. Press, Cambridge, UK, [4] Schaufelberger, John E. Risk Management on Build-Operate-Transfer Projects, Construction Research Congress 2005, ASCE. [5] Web Definition for Design-Build, Accessed in July [6] Schaufelberger, John E. Use of Design-Build on Mass Transit Rail Projects, Construction Research Congress 2005, ASCE. [7] United Nations Industrial Development Organization (UNIDO). Guideline for the Development through Build-Operate-Transfer (BOT) Projects, Vienna, [8] "Eurotunnel: Au Revoir Alastair," The Sunday Times (London), October 6, [9] Website of Eurotunnel, Accessed in June [10] Definition of NPV (Net Present Value), n.htm, Accessed in June [11] Accessed in June [12] Winch, Graham M. The Channel Fixed Link: Le Projet du Siecle, Case Study, MEng 450, The University of Manchester, United Kingdom, November [13] Eurotunnel, Eurotunnel 2004 Results, April 2005, Accessible at E C69/0/Annualresults2004UK.pdf. [14] Accessed in June [15] Blank, L. and Tarquin, A. Engineering Economy, Fifth Edition, McGraw-Hill, [16] Accessed in June 2005.

11 Amekudzi and Jeon 11 LIST OT TABLES AND FIGURES TABLES TABLE 1 Cost Overruns, Less-than-predicted Revenues, and Contract Type of Selected Megaprojects TABLE 2 Predicted and Actual Records of the Channel Tunnel TABLE 3 Predicted and Actual NPVs of the Channel Tunnel TABLE 4 Impacts of Risk Factors on the NPVs of the Channel Tunnel TABLE 5 NPVs for Lower and Upper Boundary Scenario TABLE 1 Cost Overruns, Less-than-predicted Revenues, and Contract Type of Selected Megaprojects TABLE 2 Predicted and Actual Records of the Channel Tunnel TABLE 3 Predicted and Actual NPVs of the Channel Tunnel TABLE 4 Impacts of Risk Factors on the NPVs of the Channel Tunnel TABLE 5 NPVs for Lower and Upper Boundary Scenario FIGURE 1 Costs, Revenues, and Operating Profit of Channel Tunnel

12 Amekudzi and Jeon 12 TABLE 1 Cost Overruns, Less-than-predicted Revenues, and Contract Type of Selected Megaprojects Project Cost overrun (%) Actual demand (%) of the forecast, opening year Type of private financing Channel Tunnel, UK, France BOOT Boston CA/T, USA 196 N/A N/A Denver International Airport N/A 55 N/A Boson-Washington-N.Y. rail 130 N/A N/A Mexico City metro, Mexico N/A Shinkansen Joetsu rail, Japan 100 N/A N/A Malaysia s North-South Highway 146 N/A BOT [Adapted from Megaprojects and Risks (2003) and Schaufelberger (2005)]

13 Amekudzi and Jeon 13 TABLE 2 Predicted and Actual Records of the Channel Tunnel Risk Factor Projected Actual or % of the predicted Project Delivery April, 1993 May, month Demand Rail Passenger 2.9 (1995) 18.0% (million) 15.9 (1994) 6.9 (2001) 43.0% (million gross tonnes) Rail Freight 1.3 (1995) 18.0% 7.2 (1994) 2.4 (2001) 33.0% Investment Cost (1997 prices, Billion and $) (conversion to US$ at in 1997) $7.04 $ % Financing Cost (1997 prices, Billion and $) $7.8 $ % Operating Profit by the year 2004 (Billion and $) 4.31 (EST) 0.35 $6.87 (EST) $ % Sources: Megaprojects and Risks (2003), Eurotunnel: Au Revoir Alastair (1997), and Winch (1998).

14 Amekudzi and Jeon 14 TABLE 3 Predicted and Actual NPVs of the Channel Tunnel Year Inflation-adjusted interest rate = 10.15% Predicted Cash Flow (Million $) Actual Cash Flow (Million $) NPV ($ Million) ,836.01

15 Amekudzi and Jeon 15 Billion$ ('97prices 2, ,000-4,000-6,000-8,000-10,000-12,000-14,000 '94 '95 '96 '97 '98 '99 '00 '01 '02 '03 '04 '05 '06 '07 '08 '09 ' Year Costs Revenues Operating Profit FIGURE 1 Costs, Revenues, and Operating Profit of Channel Tunnel Source: Eurotunnel Annual Accounts

16 Amekudzi and Jeon 16 TABLE 4 Impacts of Risk Factors on the NPVs of the Channel Tunnel Base Case NPV -11,836.0 (Million $) Inflation-adjusted Interest Rate Project Timeline Travel Demand Inflation Rate Completed 2 yrs later Completed 1 yr early Only 20% of predicted demand Reach at the predicted demand 5% (high) 1% (low) 10.15% 10.15% 10.15% 10.15% 12.56% 8.27% NPV (Million $) -15,088.0 (-27.5%) -9,764.8 (+17.5%) -13,260.1 (-12.0%) -7,653.2 (+35.3%) -13,480.7 (-13.9%) -9,537.5 (+19.4%)

17 Amekudzi and Jeon 17 TABLE 5 NPVs for Lower and Upper Boundary Scenario Lower Boundary Scenario Base Case Upper Boundary Scenario Inflation-adjusted Interest Rate NPV (Million $) Scenario with demand at 20% of the predicted, project completed 2 years later than predicted, inflation at 5% Scenario with demand reaching at the predicted, project completed 1 year earlier than predicted, inflation at 1% 12.56% 10.15% 8.27% -16,369.5 (-38.3%) -11, ,713.1 (+60.2%)