Strategic Decision Analysis for Transportation Systems

by
Yasuaki Moriyama
B. Eng., Urban Engineering (1995)
The University of Tokyo, Japan

Submitted to the Department of Civil and Environmental Engineering

in partial fulfillment of the requirements for the degree of Master of Science
in Civil and Environmental Engineering
at the
MASSACHUSETTS INSTITUTE OF TECHNOLOGY
June 2003
@ 2003 Yasuaki Moriyama. All rights reserved.

The author hereby grants to MIT permission to reproduce and to distribute publicly paper and
electronic copies of this thesis document in whole or in part.

Author I iý"
Department of Civil and Environmental Engineering
May 9 h, 2003
Certified by
SFred Moavenzadeh
James Mason Crafts Professor
Professor of Systems Engineering and Civil and Environmental Engineering
Director, Center for Technology, Policy, and Industrial Development
Diector, Technology and Development Program
Thesis Supervisor
•' ' __ 1 _ Kc"
by
Certified

Joseph Sussman
JR East Professor
Pro f C9 and Environ/ental Engineering and Engineering Systems
Thesis Co-Supervisor
Accepted by_
if, Oral Buyukozturk
Chaintan, Departmental Committee on Graduate Studies
 Strategic Decision Analysis for Transportation Systems
by
Yasuaki Moriyama
Submitted to the Department of Civil and Environmental Engineering
On May 9t , 2003 in partial fulfillment of the requirements for the
Degree of Master of Science in Civil and Environmental Engineering

Abstract

This thesis examines decision analysis methods of capital investment in transportation systems.
The transportation system is a fundamental infrastructure, and requires strategic thinking in
making decisions on its projects. The current valuation methods, however, do not reflect the
strategic value of a project. We investigate how to make a strategic decision by comprehensively
valuing a project in transportation systems.

This study begins by analyzing the characteristics and risks of a railway project as an example of
the transportation system, and comparing the current valuation methods based on this analysis.
Current methods are useful in valuing a project when we can fairly predict its performance in the
system. However, we have difficulty in valuing a project under uncertainty. Focusing on the
present value of cash flows does not properly measure the significance of a strategic project, but
real options value (ROV) is an effective tool to do it.

We next examine the theoretical correctness of the real options as applied to a transportation
project. The critical question is whether we can create a portfolio that replicates the payoffs of
options even if we cannot trade the options in a market. We answer this question by proposing the
complete market assumption in the real options.

To this end, we analyze a case of a railway project with Binominal Model. ROV quantifies the
strategic value of a project, and consequently improves the investment strategy. We show how it
is critical to recognize options in a project and benefit from them to effectively develop
transportation systems in a competitive market.

Thesis Supervisor: Fred Moavenzaeh

James Mason Crafts Professor
Professor of Systems Engineering and Civil and Environmental Engineering
Director, Center for Technology, Policy, and Industrial Development
Director, Technology and Development Program

Thesis Co-Supervisor: Joseph Sussman

JR East Professor
Professor of Civil and Environmental Engineering and Engineering Systems
 Acknowledgments

First and foremost, I sincerely appreciate my thesis advisors, Professors Fred

Moavenzadeh and Joseph Sussman for their guidance, encouragement, and support. Their
comprehensive knowledge and experience in the construction and transportation fields
gave me the right direction and understanding in this research. I also express my heartfelt
gratitude for their personal attention throughout my academic activities at MIT over the
past two years. It was absolutely one of the happiest and most exciting times in my life.

I am grateful to Professor Stewart C. Myers who gave me valuable advice and

suggestions that have improved the quality of this research. I thank my friends Dr. Mahdi
H. Mattar and I-Tsung Tsai. They gave me a lot of practical advice based on the
experience in their research. I also appreciate Mr. Christopher Gordon at Massport. There,
I learned a lot about the capital investment strategy and the professional mind.

I express my gratitude to my dear friends at MIT. Especially, I am thankful to graduate

students in the Construction Management Program. Studying together and sharing
experience with them was always enjoyable, and greatly broadened my thinking.

I am deeply indebted to East Japan Railway Company for understanding my earnest

desire to study in the U.S., and providing financial supports. I also thank my colleagues
who sent me the project data, which were indispensable to my research.

I very much appreciate the endless support from my wife, Mikako. Her help and concerns
encouraged me with tenderness. I would also like to thank my lovely daughter, Kano. Her
smile gave me the vitality to complete this work. Thank you so much for every thing and
I dedicate this thesis to you all.

Yasuaki Moriyama
May 2003
 Table of Contents

ABSTRACT 3
ACKNOWLEDGMENTS 5
TABLE OF CONTENTS 6
LIST OF FIGURES 8
LIST OF TABLES 9

CHAPTER 1 INTRODUCTION 10

1.1 OVERVIEW 10
1.1.1 HISTORICAL BACKGROUND 10
1.1.2 MOTIVATION AND OBJECTIVES 11
1.1.3 STRUCTURE 13
1.2 RAILWAY PROJECT 15
1.2.1 FUNDAMENTAL FUNCTION OF "RAIL" 15
1.2.2 MARKET FORCE 16
1.2.3 BUSINESS DYNAMICS 20
1.2.4 CAPITAL INVESTMENT 22
1.3 RISK AND UNCERTAINTY 24
1.3.1 RISK 24
1.3.2 STRATEGY 27
1.3.3 UNCERTAINTY 29

CHAPTER 2 DECISION ANALYSIS 33

2.1 LITERATURE REVIEW 33

2.1.1 NET PRESENT VALUE 34
2.1.2 CAPITAL ASSET PRICING MODEL 36
2.1.3 SENSITIVITY ANALYSIS 38
2.1.4 MONTE CARLO SIMULATION 39
2.1.5 DECISION TREE 41
2.2 IMPLICATIONS 43
2.2.1 EFFECTIVENESS 43
2.2.2 LIMITATIONS OF NPV 45
2.2.3 OPTION APPROACH 47
2.3 PROPOSED STRATEGY 48
2.3.1 FLEXIBILITY 49
2.3.2 TIMING 49
2.3.3 VALUATION 50
 S
OPTIO
REAL
3
HAPTER

3.1 THEORETICAL BACKGROUND 52

3.1.1 PROJECT AND OPTIONS 52
3.1.2 VALUE OF OPTIONS 53
3.1.3 NO-ARBITRAGE APPROACH 56
3.1.4 COMPLETE MARKET ASSUMPTION 58
3.2 APPLICATION TO RAILWAY PROJECTS 63
3.2.1 EXPANSION OPTION 63
3.2.2 ABANDON OPTION 65
3.2.3 WAIT OPTION 66
3.2.4 SWITCH OPTION 68
3.3 VALUATION FRAMEWORK 69
3.3.1 BLACK-SCHOLES FORMULA 70
3.3.2 BINOMINAL MODEL 71

CHAPTER 4 CASE STUDY -STATION DEVELOPMENT- 75

4.1 CASE DESCRIPTION 76

4.1.1 COMPANY INFORMATION 76
4.1.2 STATION DEVELOPMENT PROJECTS 79
4.1.3 NPV ANALYSIS 84
4.2 REAL OPTIONS ANALYSIS 90
4.2.1 RESULTS OF THE NPV ANALYSIS 90
4.2.2 VALUATION PROCESS 92
4.2.3 OPTIONS VALUE 97
4.3 RECOMMENDATIONS 106

CHAPTER 5 CONCLUSION 111

5.1 CONCLUSIONS 111

5.2 FUTURE RESEARCH 114

BIBLIOGRAPHY 117
 List of Figures

FIGURE 1-1 THE FIVE COMPETITIVE FORCES THAT DETERMINE INDUSTRY PROFITABILITY ....... 17
FIGURE 1-2 TWO DIFFERENT LOOPS OF INVESTMENT EFFECTS..................................................... 21
FIGURE 1-3 PORTFOLIO OF PROJECTS IN A RAILWAY COMPANY (EXAMPLE)......................... . 23
FIGURE 1-4 MANAGING RISKS .......................... .. ...................... 28
FIGURE 1-5 DIFFERENT RISK .......................................... ..... ....... ....................... 30
FIGURE 1-6 DIFFERENT UNCERTAINTY ............................................ 30
FIGURE 1-7 GEOMETRIC BROWNIAN MOTION MODEL......................... ................................... 31
FIGURE 2-1 A FIRM CAN EITHER INVEST THE CASH IN A PROJECT OR RETURN IT TO
SHAREHOLDERS ................................... ...... 35
FIGURE 2-2 CAPITAL ASSET PRICING MODEL ....................................... 37
FIGURE 2-3 EXAMPLE OF THE RESULT FROM MONTE CARLO SIMULATION ............................... 40
FIGURE 2-4 EXAMPLE OF DECISION TREE ................................................................................ 42
FIGURE 2-5 VALUATION FRAMEWORK .... ................. ............................ .................... 45
FIGURE 2-6 VALUE OF A PROJECT IN DIFFERENT APPROACHES UNDER THE UNCERTAINTY.......... 48
FIGURE 3-1 VALUE OF OPTION (CALL OPTION) ........................................ 54
FIGURE 3-2 ASYMMETRICAL DISTRIBUTION GIVES VALUES TO OPTIONS ...................................... 66
FIGURE 3-3 EXPECTED VALUE OF OPTION .......................................... 73
FIGURE 4-1 OPERATING AREA OF JR EAST ................................................. .......................... 77
FIGURE 4-2 EXAMPLES OF THE STATION DEVELOPMENT.......................... ............... 79
FIGURE 4-3 RANKING AND LOCATION OF PROJECTS A, B, C........................... ....... ................. 81
FIGURE 4-4 CONSTRUCTION AND BUSINESS PLAN OF PROJECT A ..................................... . 82
FIGURE 4-5 CONSTRUCTION AND BUSINESS PLAN OF PROJECT B ........................................... 83
FIGURE 4-6 CONSTRUCTION AND BUSINESS PLAN OF PROJECT C....................................... 84
FIGURE 4-7 MINIMUM LEVEL OF CASH FLOW.................................................. 86
FIGURE 4-8 NPV OF PROJECT A, B, C (WITHOUT SALVAGE VALUE) ............................................ 87
FIGURE 4-9 NEW NPV ANALYSIS (WITH SALVAGE VALUES IN PROJECT A AND B) ................... 88
FIGURE 4-10 IMPLEMENT STRATEGY ............................................. 92
FIGURE 4-11 UNCERTAINTY WILL DECREASE AS THE INVESTMENT PROCEEDS ......................... 94
FIGURE 4-12 VOLATILITY OF THE UNDERLYING ASSET................................ 96
FIGURE 4-13 VOLATILITY IN THE SALES PER UNIT AREA.................................... ............ 96
FIGURE 4-14 FLOW CHART FOR THE DECISION PROCESS IN EXERCISING OPTIONS ................... 100
FIGURE 4-15 EXERCISE TIM ING.......................................................... .................................... 101
FIGURE 4-16 VALUE OF THE PROJECTS (A, B, C) BY IMPLEMENTATION STRATEGIES ................ 107
FIGURE 4-17 MANAGERS CAN CHOOSE EITHER SPEED OR CASH VALUES IN AN IMPLEMENT
STRATEGY ................................................................................................................ . . 108
FIGURE 4-18 ADDITIONAL VALUES OF PROJECTS ........................................ 109
FIGURE 5-1 NETWORK FACTOR .......................................................... 114
 List of Tables

TABLE 1-1 DIFFERENT RANGES OF COMPETITION ......................................... ............... 18

TABLE 1-2 ATTRACTIVENESS OF THE RAILWAY MARKET ...................................... ......... 19
TABLE 3-1 RELATIONSHIP BETWEEN THE OPTION VALUE AND EACH PARAMETER .................... 55
TABLE 3-2 THREE STRATEGIES FOR INVESTMENT IN A RISKY ASSET ..................................... 57
TABLE 4-1 FINANCIAL HIGHLIGHTS IN THE FY 2001 (MILLION DOLLARS).................. 78
TABLE 4-2 PROJECT DATA............................................... 82
TABLE 4-3 SPREADSHEETS FOR THE NPVS OF PROJECT A, B, C ($M)................................ . 89
TABLE 4-4 RESULTS OF ROV ANALYSIS .......................................... ...................... 97
TABLE 4-5 PARAM ETERS ............................................................................................................. 98
TABLE 4-6 BINOMINAL MODEL OF PROJECT A (PERIOD 0-1)................................. 100
TABLE 4-7 OPTION PRICING FOR PROJECT A.................................... 102
TABLE 4-8 OPTION PRICING FOR PROJECT B ........................................ 104
 Chapter 1 Introduction

1.1 Overview

1.1.1 Historical Background

The transportation system is a fundamental infrastructure, which supports our daily life,

economic activities, and the future development of society. Throughout the history of our

nation, we have developed a wide variety of transportation systems with the combined

effort of the public and private sectors. Among them, the railway system, which was born

in the beginning of the 1 9 th century, was one of the greatest innovations of the industrial

revolution both in terms of technology and management, and has contributed to the

development of the transportation industry all over the world.

The railway has a competitive advantage in its efficiency. It can carry a large number of

passengers and freight at a good speed with relatively lower cost and less consumption of

energy than other modes. In the process of industrialization, we have developed railway-

specialized infrastructures, such as inter-city (sometimes international) connections and

commuter systems in metropolitan areas. The railway network is a quite important

element for urban planning and public policy, and can be an excellent business model as

a profit-based company. When we review the history of the U.S., the railway used to be

one of the profitable industries in the 19

th
century. In other countries, we can also see that

venture investors founded many railway companies.

Since the middle of the 20 th century, however, many railway businesses have declined

with the emergence of motorization and air transportation. Some ended in bankruptcy;

others survived while receiving subsidies from the government. The railway is no longer

an attractive business in some countries. Of course, private ownership is not always the

best form for railway organizations, and public ownership and management can be

appropriate in given situations. Still, we recognize the significance of privatization, which

is currently evolving all over the world, and how private management and business

strategy are effective in improving the value of the infrastructure. We are concerned with

how we can revive the railway business in the 2 1st century.

1.1.2 Motivation and Objectives

Why has the railway business declined? Was it just because of the competition with

automobiles, trucks, or airplanes? Are railway companies profitable only if they have

monopoly power in the transportation market? We can say that the railway used to

dominate the transportation market, but lost its share to automobiles, trucks, and air

transportations. The railway business is, however, still profitable in some segments. For

instance, freight railways in the U.S. keep high market share and reasonable profit, and

commuter railways in the Tokyo metropolitan area have developed diversified business

models. Moreover, high-speed railways in Europe, which are partly subsidized, have

captured some demand for travel/business trips again, and demonstrated the potential

success of the business if a railway company could match their service and fare to market

needs.
A primary reason for failures in the railway business is that managers could not make

strategic decisions about how to adjust to their new business environments. This is not

saying that it was the manager's fault, but clarifies how difficult it is to determine

strategies in a large and complex system such as the railway. Managing a transportation

project is difficult because each project does not work individually; rather, it performs as

a part of system. Making decisions in a project may give unexpected impacts on another

project in the complex system. Sussman (2000) advocated, "Transportation is an

example of a broader class of systems we call CLIOS - complex, large, integrated open

systems" . This fundamental nature of the transportation as a system makes it challenging

for managers to predict the performance and impacts of a project, and, thus, determine

their strategy. Therefore, there are recent advances in technologies, logistics, and

management to support decision analysis for transportation systems.

Standing on these accomplishments, this thesis investigates how we can develop a

methodology that can be applied to strategic decision analysis for transportation systems

by taking advantage of recently advanced management theories. Specifically, we will

focus on quantitative approaches to an optimal investment strategy of a railway company

over time. The practical objective of this research is to improve the decision-making

process for investment of railway projects, and enhance the competitiveness of railway

companies in the transportation market.

1 Sussman, Joseph "Introduction to Transportation Systems", 2000, pp. 6-7

1.1.3 Structure

This thesis mainly focuses on methodologies for decision analysis of a transportation

project. In particular, we examine which method is effective for different types of

projects, what will be the problems in applying it, and how to adjust these methods for a

project. As an application, we study a railway project case to confirm the effectiveness of

these methodologies in making strategic decisions in a large and complex transportation

system.

In Chapter 1, we summarize the characteristics and risks of a railway project, which have

impacts on decision-making process. In a transportation project, many studies have been

done to forecast the demand, design an effective investment, and evaluate the feasibility

so that we can deal with "risk", which can be quantified by a probabilistic approach.

However, we still have difficulty in dealing with "uncertainty", in which the probabilities

of potential outcomes are unclear. Therefore, managers periodically review their projects;

this can be a practical way to cope with the uncertainty. However, most infrastructure

investment is irreversible, so we need to recognize this uncertainty even in the decision-

making process. This chapter specifies the target for strategic decision analysis, which we

develop in this research.

Chapter 2 first reviews current literature for decision analysis theories, and next compares

their effectiveness in terms of application to a railway project. Among them, Net Present

Value (NPV) seems to be a current paradigm for many companies in making decisions on

capital investment. This methodology, however, heavily depends on the projected cash
flows, and is quite sensitive to unexpected events. Theoretically, the NPV can consider

the risk of a project, but has difficulty in fully dealing with uncertainty. This chapter

introduces a concept of "real options" which can value a project under uncertainty by

taking advantage of recently developed financial option techniques.

We examine the details of real options in Chapter 3. The key idea is "options" in the

decision-making process. If the future is difficult to predict, it is better to provide some

flexibility in a project than make every decision in advance. Using the real options, we

can value the flexibility and allow managers to decide when they implement each

investment over time and with new information. This chapter suggests how we can utilize

real options for a specific type of railway projects.

In Chapter 4, we analyze a strategy of station development projects as a case study. This

case describes a strategic decision of East Japan Railway Company (Tokyo, Japan) who

explores business opportunities by developing the stations in the Tokyo metropolitan area.

The company recognizes the potential advantage of stations, which are located at the

centers of downtowns and have many passengers. The managers, however, have

difficulty in making an investment decision by using the NPV, which does not reflect the

strategic value of this project. We examine how real options provide a rational answer for

this decision-making problem.

Chapter 5 concludes this thesis by summarizing the results of this research and proposing

further research to expand the application of real options in transportation systems.

1.2 Railway Project

This chapter introduces fundamental concepts of a railway project. We summarize

railway-specific characteristics and classify projects to explain the focus of the strategic

decision analysis. This process provides the basis of arguments in the following chapters.

1.2.1 Fundamental Function of "Rail"

The definition of railway can be that it has "rail" in its system components. The existence

of rail has two significant meanings when we discuss its characteristics. One is the

advantage of "efficiency" both in terms of technology and management. The other is the

disadvantage of "rigidity" due to its fixed cost and fixed location of the facilities.

The use of steel rail and wheel drastically decreases the physical friction and loss of

energy compared to other surface transportation systems. Combined with the invention of

internal-combustion engine, the railway gave us the ability to transport a large number of

passengers and freight with low cost and high speed. The fixed route enables us to

operate the vehicles efficiently. Using advanced signal and control technologies, we can

realize high transport density. The minimal impact of the railway on the environment is

also derived from its efficient consumption of energy in this system.

The fixed cost and fixed location of facility, on the other hand, is the major disadvantage

of rail systems. Vehicles have to operate on the rail, and cannot freely choose their route,

unlike automobiles or trucks. Therefore, the railway requires a certain amount of facilities
regardless of the level of demand and operation density. In a certain situation, fixed

facilities cost more than the operating cost, and, sometimes, more than the revenue. We

continue to discuss further aspects of a railway project, which affect decision analysis,

but this fundamental tradeoff is the basis of these arguments.

1.2.2 Market Force

Porter (1998) explains the long-term profitability of a firm based on the five forces in the

market value chain. Figure 1-1 describes how the combination of these five forces will

determine the dynamics of competition, and how a firm will assess its position in a

competitive market.

Porter's model explains the demand from the market, which puts pressure upon each

firm's performance. Every firm has to improve its service (product) while raising the

quality and lowering the cost throughout the market value chain. The firm is exposed to

direct competition with its rivals to keep/increase its share in an existing market segment

(rivalry among existing competitors). If the market segment is attractive enough, another

rival, who is not currently a competitor, will come in to try to take the customers away

(threat of new entrance). The firm needs to compete with existing rivals while building

the entry barriers against potential competitors. In addition, the firm has to compete with

another market segment such as new products or services which can replace theirs (threat

of substitutes). Customers are often in a position to choose from competitors or negotiate

the price (the bargaining power of buyers). The firm's suppliers also put pressure on the

firm's strategy to improve their inbound logistics (the bargaining power of supplier).

Threat of New Entrants

Bargaining Power Rivalry Among Bargaining Power

of Suppliers Existing Competitors Sof Buyers

Threat of Substitutes

Figure 1-1 The Five Competitive Forces that Determine Industry Profitability 2.

The nature of the railway business, however, makes it difficult for managers to fully deal

with these five forces. As described in this chapter (1.2.1 Fundamental Function of Rail),

the railway system needs fixed facilities to achieve its low operating cost. This fact

naturally results in economies of scale, which resulted in a regional monopoly in some

cases. Therefore, a railway company does not always have severe competition directly

with other railway companies. Vast initial investment, government regulation, and the

potential advantage of a predecessor imply high entry barriers to potential competitors.

However, other transportation modes (e.g., auto, air, truck) can eventually take their

customers away as substitutes. This is actually what happened in the 2 0 th century. Also,
other communication methods such as the telephone or Internet can be effective

substitutes for this market in a long time scale. Moreover, we can consider the customer's

preference in spending time as a substitute for this segment (e.g., customers may prefer

staying home to traveling). These different ranges of competition add complexity for

managers as to their market position. Table 1-1 summarizes these different ranges of

competition. It is easy for managers to recognize direct competitors (1 st range) and take

measure to cope with the situation. Indirect competition (2 nd- 4 th range) proceeds in

relatively long time-scales, and thus it is difficult to determine the market position if

managers focus on short-term performance. The railway business often faces these

indirect competitors more than direct ones, and managers are required to strategically

compete with these various rivals in different time scales.

Table 1-1 Different ranges of competition

Time
Range Competitor Scale Driver
Scale
1st Other railway companies Level of Service
Short v
Availability
nd Infrastructure
tr
2 nd Other transportation modes
Purpose of trip

3 rd Communication technologies Technological innovation

Communication technologies
Business style
Attractiveness
4th Preference in spending time Long Sense of value
Family structure

2Porter, Michael E., "Competitive Advantage", The Free Press, 1998, pp. 5
Furthermore, customers usually have little bargaining power against providers in terms of

price. They can choose a transportation mode based on the level of service of their

options. Customers are, however, rarely in the position to deal with the price or service

individually. This bargaining power is exercised by the government as a regulator.

Managers sometimes misunderstand who the customer is for their service. Some railway

companies are not customer-oriented, but government-oriented. The supplier sometimes

has strong bargaining power when the entry barrier (e.g., railway-specific standard)

discourages the new entries in the inbound industries, and decreases the number of

competitors. In this case, the supplier may have monopoly power, which prevents the

improvement of inbound logistics and technical innovations of a railway company. We

can also include labor union issues in this category of suppliers. Table 1-2 summarizes

the long-term attractiveness of this market.

Table 1-2 Attractiveness of the railway market

Market Force Threat Factors

Rivalry Among Existing
Low Economies of Scale
Competitors
Vast initial investment
Threat of New Entrants Low
Regulation
Other transportation modes
Threat of Substitutes Communication technologies
(Gradually)
Preference in spending time
Customer - Low
Bargaining Power of Buyers Middle
Government - High
Manufacturers
Bargaining Power of Suppliers Middle
Labor unions
(Note: Low level of threat = High profitability of the market)
 1.2.3 Business Dynamics

Managers have to deal with these market forces in a railway business. The complex

system and business environment of a railway company, however, may lead managers to

improper decisions. Although their principal goal is to create and maintain a competitive

advantage in the market, managers tend to insist on short-term profits and operation

policies due to the different time scales of their system output. For example, safety and

environment related investment does not provide direct financial returns, unlike more

frequent operation or high-speed projects. These projects pay, however, in the long run

through the improvement of their reliability in service and healthy development of the

land use or community. The lack of direct competitors may lead managers to save the

capital investment and maintenance expenditure, which lower their competitiveness in

the future. Also, managers have to optimize their strategy in different time scales. Since

most infrastructure investment is usually irreversible, the optimal design of a project in a

short term may have negative impacts on the potential opportunities in the future.

Sterman (2000) built system thinking and a modeling framework for a complex world,

arguing that our own actions have unanticipated side effects and cause many problems in

the future 3 . Therefore, our effort to solve important problems may not only fail but also

make it worse or create new problems. To deal with the growing dynamic complexity, we

need to expand the boundaries of our mental model and develop strategic thinking in the

decision-making process.

3 Sterman, John D., "Business Dynamics", Irwin McGraw-Hill, 2000

 Level of Servic

Investment Cash Out

Delay

Attractiveness R B

Cash Flow Debt Service

+
SCompetitive
Advantage

4
Figure 1-2 Two different loops of investment effects

Figure 1-2 indicates two different loops of investment effects in a firm. The right loop

shows a short-term effect of the investment on the financial status. The incremental

investment will increase the cash out, and debt payment, and finally decrease the source

of investment (cash flow). This loop works to balance (B) the financial status through

investment activities. The left loop indicates how the investment gives returns through

the sequential increase of the level of service, attractiveness, and competitive advantage,

and cash flow. This loop works to reinforce (R) the investment direction in both positive

and passive ways. The point is that these two loops work in different time scales.

Investment decisions based on the present financial status may result in non-optimal use

of capital resources and consequently insufficient investment effects. Managers have to

maintain their strategic views in making decisions on investment. Standing still or

focusing on short-term return is not a good option in a business.

4This diagram is originally drawn for this thesis.

1.2.4 Capital Investment

We now discuss the types of investment. In general, there are two broad categories in

railway investment. One is the operating investment, which sustains the current

performance of the system. The other is the capital investment, which has strategic

objectives to enhance the performance of the system. This thesis mainly focuses on the

capital investment since it is a critical decision, which enables the railway company to

strategically compete in a complex environment.

Of course, maintenance can be an incrementally important factor for the management of a

railway company. It is, however, a kind of requirement to sustain the current operating

performance and the level of service. This can be somewhat adjusted by managers

according to the situation, but cannot be drastically changed. Therefore, we will focus on

the capital investment in which managers have ability to design their investment.

The capital investment enhances the performance of systems to achieve the goal and

mission of a company. This investment is an important management decision. It is a

difficult decision not only because it is a large investment, but also because it can carry

extra risks to the company. The capital investment is actually productive activity, which

tries to increase the value of the company. Its usually irreversible nature and strong

impacts on the operation and maintenance activities, however, create additional risks as a

result.
 The railway company
Figure 1-3 shows the portfolio of projects in a railway company.
has synergistic effects with
can diversify its portfolio by exploring other business, which
thesis, however, we limit the
railway activities (e.g., real estate, hotel, resort). In this
1-3) so that we can examine
scope of the portfolio to the following major projects (Figure
requires sophisticated
the strategy of railway companies. Although every project
investment) need more
decisions on its investment, projects in shaded segments (capital

strategic thinking to mitigate the risks in this portfolio.

Figure 1-3 Portfolio of projects in a railway company (example)

1.3 Risk and Uncertainty

In Figure 1-3, we classified railway-specific projects based on their scale and risk.

Managers have to recognize the importance of risk assessment to cope with the different

levels of risk in their projects. In this section, we identify the risks in a railway project,

suggest strategies to cope with the risks, and define the difference between risk and

uncertainty to manage a strategic project.

1.3.1 Risk

Lessard and Miller (2001) successfully classified the types of risk faced in large

engineering projects, and consolidate them into three major risk categories; (1) Market

Risk, (2) Technical Risk, and (3) Institutional Risk. These categories are useful for us in

identifying the risks of a railway project. Although the type of risk differs according to a

project, these categories are properly applicable to a railway project based on the

arguments, which we have made in this chapter. Therefore, we examine the risks based

on these categories. However, the risk does not work separately; or rather, it is often

accompanied by many related risks, which affect the performance of the project in a

complex way. Hence, managers are required to pay comprehensive attention to the risks

in order to mitigate and avoid the managerial losses incurred when the some of the risks

turn into reality.

Market Risk

The demand, finance, and operations risks are in this category. In a business, the market

is always important but uncertain. The profitability of a railway business depends on the

demand of the market. This market risk is related to the ability to forecast the demand.

Inaccuracy in the forecast may lead to shortfall in the cash flows, or inadequate capacity

of facilities. Moreover, in many transportation projects, the demand has a fundamental

tradeoff relationship with the level of service. Adjusting the fare to balance the demand

may even worsen the situation in a competitive market. In general, demand is very

difficult to predict and hence constitutes a great income risk. Also, urban transportation

projects, which can expect high demand, face the high cost of projects because of

expensive land acquisitions and strict environmental constraints.

On the other hand, operations risk is relatively low for a railway project once it recovers

the initial investment. The railway project has reliable daily cash flows from passengers.

Moreover, the demand, specifically in the commuter segment, is stable regardless of

economics cycles. In addition, the low operating cost of the railway enables it to keep

earning cumulative profits through the operating period. In general, the volatility of the

market risk is very high in the beginning, and rapidly stabilized through the life of a

project.
Technical Risk

The technical risk is mainly related to the complexity of the system. Since a railway

project involves many types of different software and hardware, both individual

performance of each system and the integration of the whole system can be concerns. In

particular, the integration between newly installed technologies and the existing railway

system can present a challenge for a railway project. Developing a new technology is

based on numerous assumptions that cannot be verified until the operation starts. When

we put a new technology to practical use, it may not only fail to perform well, but also it

can disorder the correct operations of existing systems. However, due to the social

responsibility as an infrastructure, managers are required to keep operating railway

systems, and have to avoid such troubles. To launch a new technology or modify an

existing system involves complex risk. Managers have to maintain sufficient

technological expertise to deal with this complexity.

Institutional Risk

The regulatory and political risks are concerns to a railway project. As an infrastructure,

the regulator sometimes has strong authority for providing a permit (concession) to a new

railway project. The manager has to consider the regulatory frameworks, urban planning,

and public policies surrounding the project in the entry strategy.

The feasibility of a project also depends on the law and regulations that govern the

pricing rule, liability, property rights, and contracts of the project. The railway project
faces regulatory and political risks, which may impose unanticipated restrictions on the

project. We also have to consider the possibility that a project will meet opposition from

the residents, local groups, interest parties, and politicians.

Furthermore, if the expected return from the project is insufficient, and managers have no

alternative to increase the return, they usually take an exit strategy. In a railway project,

however, it is often difficult to stop providing its service or sell its assets. In this case, the

project loses its ability to limit substantial loss, and managers have to cope with this

institutional risk even in taking the exit strategy.

1.3.2 Strategy

Lessard and Miller (2001) argue, "Successful projects are not chosen but shaped with risk

resolution in mind. "6This statement implies how it is critical for managers to recognize

the risk in advance and prepare proper solutions to mitigate its negative impacts. In order

to manage the risk, first of all, we need to identify the investment-related risk. Next, we

also need to evaluate their impacts on investment activities. The objective of this process

is to reduce the exposure of the investment to the identified risk. Figure 1-4 shows the

proposed strategy based on the type and the extent of control over risk.

6Lessard D. & Miller, R., "The Strategic Management of Large Engineering Projects", MIT Sloan Working Paper,
2000
 -o Diversify
portfolios of projects Embrace the residual risk

C, o
x ;; Direct allocation and mitigation Shifting of the indeterminate risks

High Control <- [Extent of control over risk] -> Low Control

Figure 1-4 Managing Risks 7

If the risk is specific to a project and controllable, try to identify the source and the

impacts of the risk, and reshape the investment so as to allocate and mitigate the negative

impacts. This is the basic case in risk management. If the risk is specific to a project but

outside the control of managers, however, they have to shift the risk by using contracts or

markets and prepare for unexpected events. This can be the risk management on a project

level. In contrast, when the risk is broad but controllable, the effective approach is to

diversify the portfolio of projects and reduce the impacts of the risk. This approach has to

be done not on a project level, but on a corporate level. Finally, when the risk is difficult

to specify and control, we have to embrace this risk. Managers have to decide if the risk

is worth taking.

7 Lessard D. & Miller, R., "Understanding and Managing Risks in Large Engineering Projects", 2001 p. 11
1.3.3 Uncertainty

In this thesis, we have been using "risk" to consider the possible impact of unexpected

events in a project. In order to practically manage a project, however, we have to

distinguish "uncertainty" from "risk", and recognize the different natures and effects of

the two states.

Lessard and Miller (2000) explained risk as "the possibility that events, the resulting

impacts, the associated actions, and the dynamic interactions among the three may turn

out differently than anticipated."8 Here, we understand that risk can be quantified in

statistical terms, and we can take a probabilistic approach to cope with unexpected events.

However, "uncertainty characterizessituations in which potential outcomes are not fully

understood." The common concern are unexpected events in the future, which impacts on

a project, but the difference is whether we can estimate the probabilities of the

unexpected events. To make the difference clear, we define the risk and uncertainty in the

following manner:

Risk: The possibility of unexpected events with knowable probabilities, and

considering the impact of various events.

Uncertainty: The distribution of possible outcomes with unclear probabilities.

[Note: We use "risk" when we simply consider the possible impact of unexpected events without

distinguishing the probability issue (e.g., 1.3.1-2)]

 For instance, Figure 1-5 shows two projects, which have different risks. Although both

projects have the same expected value, project A has higher risk since it is more likely to

fail. Probability matters in risk analysis, and we can decide whether we will accept the

risk or not based on the expected value of a project, the probabilities of possible

outcomes, and our risk preferences.

Success: $20 (Prob. = 0.4) Success: 10$ (Prob. = 0.8)

Project A (expected value = $8.0) Project B (expected value =$8.0)

Failure: $0 (Prob. = 0.7) Failure: $0 (Prob. = 0.3)

Figure 1-5 Different Risk

However, we do not always understand the clear probabilities of events, and have to

consider uncertainty in a project. Figure 1-5 indicates two projects, which have different

uncertainty. Both projects have the same expected value, but project C is more uncertain

because it has more volatility. We have to consider the range of the possible outcomes,

and volatility matters in dealing with uncertainty.

Probability
AK Expected Value Expected Value

I
I
I
I

\I~ll~~

0 Proiect C 0 Project D

Figure 1-6 Different Uncertainty

8 Lessard, D. & Miller, R., "The Strategic Management of Large Engineering Projects", MIT Sloan Working Paper,
2000
The Geometric Brownian Motions Model, which is frequently applied to describe the

movement of financial variables (e.g., stock price, interest rate), expresses the significant

difference between risk and uncertainty in its diffusion process. According to this model,

the value of a financial asset evolves based on the following formula:

The Brownian Motions Model:

dV= a -V- dt + r -V- dz

Where:

V: Value of a financial asset

a : Expected annual growth rate
dt: Time interval
a : Standard deviation of V
dz: increment of standard Wiener process 9 (= e·4 dt)
e: normally distributed random variables with zero mean and a standard deviation
of 1.0

Value

SPossible
range of Vt

Vo

Time

Figure 1-7 Geometric Brownian Motion Model

9 The Wiener process describes a random motion of a variable.

In this model, the value of an asset varies based on the expected value and deviation on

the time axis. Therefore, we have to analyze the risks of an asset to estimate its growth

rate and expected value. However, the deviation from expected values also increases, so

we have to consider the possible error of the estimation. This deviation indicates

uncertainty of an asset where the probability of the movements is unclear.

This difference between risk and uncertainty is critical in choosing an appropriate

valuation method in a project. Some methods can take a probabilistic approach to cope

with risk. Uncertainty, however, requires managers to take a different approach, and

imposes an ambiguous task. We have to keep this fundamental difference in mind, and

take an effective strategy based on the characteristics of a project and availability of

information. In Chapter 2, we are going to examine the effectiveness of decision analysis

methods to value a project under these conditions, and discuss how we can determine

investment strategies for transportation systems.

 Chapter 2 Decision Analysis

We summarized the characteristics of a railway project and identified its associated risks

in Chapter 1. That was the first step in developing an investment strategy for a railway

company. What we have to do next is to examine the effectiveness of decision analysis

methods to value a project under these conditions. Managers need to evaluate their

investment in terms of its impacts on the business activities. Eventually, managers try to

reduce the exposure of their projects to the identified risks, and conclude with a certain

benchmark for their investment decision-making. Since there is no definitive measure for

valuing a project, we have to understand the effectiveness and limitation of each

valuation methodology, and use the effective methods according to the nature of a project.

2.1 Literature Review

Among the various project valuation methods, Net Present Value (NPV) seems to be a

current paradigm for many companies in making decisions on capital investment. Brealey

and Myers (2000) successfully verified the theoretical correctness of NPV and its overall

superiority to other valuation methods of a project (e.g., Pay Back Period, Internal Rate

of Return) in the finance literature1 o. In this section, we review the NPV, its practical

drawbacks, and its applications, which are currently taken to support an investment

decision-making.

10 Brealey, Richard A. & Myers, Stewart C., "Principle of Corporate Finance", Irwin McGraw-Hill, 2000, pp 93-114
 2.1.1 Net Present Value

The Net Present Value of a project is

T
NPV= Z CFt/(l+r) t - Io + ST/(I+r) T
t=1

where:

CFt: expected cash flows of a project at time "t"

r: the discounted rate reflecting the cost of capital and the associated risks
Io: initial investment
T: Economic life of a project.
ST: Salvage value of a project (if applicable)

The NPV will be calculated by:

1. Forecasting the cash flows generated by the project over its economic life.

2. Determining the appropriate opportunity cost of capital, which reflects both the time

value of money and risks involved in the project.

3. Using this opportunity cost of capital to discount the future cash flows.

4. Subtracting the initial investment from the sum of the discounted cash flows (present

value), and adding the salvage value if the asset has cash value at the end of a project.

If the NPV>O, the investment financially makes sense.

The excellence of NPV is its reflection of the investor's expectations. Paying dividends to

shareholders is always a firm's alternative to investing in a project. If the expected return

from a project is lower than the investor's expectations (NPV<O), a firm can pay

dividends to its shareholders, who are also looking for another opportunity in the market.
 Cash

Investment
Investment Firm Shareholder (Financial Asset)
(Real Asset)Dividend
Project Dividend

n
Figure 2-1 A firm can either invest the cash in a project or return it to shareholders

Some criteria such as Benefit/Cost (B/C) and Internal Rate of Return (IRR) can be used

with NPV, so as to compare the values of projects, and decide which project to choose.

However, these criteria have the same theoretical background of NPV, which focuses on

the cash flows and time value of money. Therefore, we categorize them in the same

group in terms of a valuation method.

There are, however, two major drawbacks in using the NPV as a valuation method for a

project. One is the difficulty of determining an appropriate discount rate, and the other is

the ambiguity as to the future cash flows. We review the discount rate problem in 2.1.2

(Capital Asset Pricing Model), and the future cash flows issue in 2.1.3-5 (Sensitivity

Analysis-Decision Tree) in order to examine whether we can avoid these drawbacks.

1 Brealey, Richard A. & Myers, Stewart C. "Principle of Corporate Finance", Irwin McGraw-Hill, 2000, pp 95
2.1.2 Capital Asset Pricing Model

In the NPV, theoretically, a discount rate depends on the risk of a project. Using

Weighted Average Cost of Capital (WACC) as a discount rate, we can evaluate the risk

of a project based on the firm's evaluation in the security market. In this case, there is an

assumption that the project's risk is the same as the firm's average risk.

The Weighted Average Cost of Capital of a firm is

WACC = re (e/v) + rd (d/v)

where

d, e = current market value of its debt and equity

v = sum of the debt and equity value (d + e)
rd = current rate of its borrowing
re = current expected rate of return on its stock (including earning and growth)

The railway company, however, has a wide variety of projects, ranging from small-scale

and low-risk projects to large-scale high-risk projects 12 . Therefore, this assumption is

theoretically inaccurate and leads managers to a wrong decision. Simple use of the

WACC will favor the risky projects, while having bias against less risky projects than the

average. Our focus is capital investment, which involves relatively high risks. Therefore,

we need to consider these risky cash flows by adjusting the discount rate.

12 Figurel-3
The most commonly used method for determining the risk-adjusting discount rates is

Capital Asset Pricing Model (CAPM), which shows how we can value the individual risk

premium in relation to its covariance to the market by using a simple linear formula.

The Capital Asset Pricing Model is:

R = Rf + / (Km - Rf)

where

R: expected return rate on a project

Rf: rate of a risk-free investment (e.g., treasury bonds)

Kin: expected return rate for the market portfolio.

f : volatility of a project

3 measures the volatility of a project, relative to the market portfolio. The above

equation shows that investors require the higher expected return to compensate the higher

expected risk. The formula determines the project-specific risk premium as a linear

function of 3. Therefore, if we know a project's 3 , then we know the value of R, which

can be the risk premium for the project.

ProDosed Proiect

Figure 2-2 Capital Asset Pricing Model

The practical drawback of using the CAPM is how to determine the " / " for a specific

project. Unlike the stock market where the CAPM has been developed, we do not have

enough historical data to determine the 3 for capital investment. Also, historical data

from other projects does not correctly reflect the project-specific risks since each project

has its unique risks. Using the CAPM gives us insight as to determining the risk premium,

but cannot fully solve the discount rate problem.

2.1.3 Sensitivity Analysis

Since the valuation of a project depends on the forecast of its cash flows, we need to

examine our premises by changing some critical parameters to see how the NPV will

vary according to this change. There are risks associated with the error in the forecast of

the initial cost, construction period, revenue, etc. This process provides not only the

feasibility of a project in a specific situation, but also the opportunity to re-shape the

project based on the sensitivity of each parameter. If the project is very sensitive to a

specific parameter, we can try to mitigate the potential risk by modifying the project so as

to reduce the exposure to risk.

In practice, managers usually vary input data and model coefficients within a certain

range (e.g., ±10%) based on their experience or guidelines. Another simple approach is

using the range from "optimistic" case to "pessimistic" case, or taking "reasonable

extremes" assumptions to compare outcomes. The underlying concept of this

methodology is that more things can happen than will actually happen, and we need to
examine each possibility in dealing with uncertainty. This sensitive analysis helps

managers to identify the possible range of outcomes, consider the impacts, and make

decisions on investment.

On the other hand, the result is ambiguous. Since the adjustment range is wide, and

possible combinations of the sensitive analysis are many, managers often have difficulty

finding the rationality of their decisions. In addition, the interrelation among parameters

often makes this analysis too complex to use in practice. Furthermore, each situation does

not equally happen, but the sensitivity analysis simply shows the result of each case, and

does not consider the probabilistic distribution of the outcomes in the simulations.

2.1.4 Monte Carlo Simulation

The sensitivity analysis enables managers to consider the effect of changing one

parameter or limited number of combinations of data at a time. Using Monte Carlo

simulation, however, we can consider all plausible combinations of data, and examine the

entire distribution of possible outcomes. We simulate a project in a computer by using the

assumed distribution of each variable.

In the Monte Carlo simulation, we first have to build a model of the project. This model

requires a set of equations and constraints for each of the variables. Second, we have to

determine the distribution of our forecast error in each variable. The forecast error has an

expected value of zero, and a range of a certain percent (not always symmetric). Finally,
the computer samples from the distribution, and calculates the resulting cash flows

(probabilistic distribution of outcomes) as shown in Figure 2-3.

13
Figure 2-3 Example of the result from Monte Carlo simulation

The advantage of using the Monte Carlo simulation is to allow managers to acknowledge

the uncertainty of their project. The manager can gain a better understanding as to how

uncertainty financially impacts on their project.

However, the Monte Carlo simulation has some drawbacks. The first one is the modeling
and
problem. It is difficult to determine the model of a project with the interrelationships
these
probabilistic distributions for each variable. Since the result depends on

assumptions, it is possible that the result will be biased. Also, in practice, a simulation

model that tries to be completely realistic will be too complex. Managers have to validate
in fully
the models to verify the result of simulations, but may have difficulties
is the
understanding the meanings of a complex model. The second drawback
can use the
interpretation of the outcomes due to the lack of specific benchmarks. We

13 http://www.decisioneering.com/spotlight/
Monte Carlo simulation to understand the possible distribution of outcomes, but it may be

difficult to use it for a decision-making if these drawbacks are critical for managers.

2.1.5 Decision Tree

Both the sensitivity analysis and Monte Carlo simulation provide managers with an

understanding of the nature of a project, specifically the possible range and distribution of

outcomes. These methods tend to underestimate, however, the value of a project because

they do not consider the manager's ability to cope with unexpected events during the life

of a project. If the project environment (e.g., market demand) is different from the initial

assumption, managers usually abandon the project to limit their loss (in case of bad

events), or chose an expansion strategy to increase the benefit (in case of new

opportunities). An investment decisions is not simply an accepted/rejected decision that

has to be done at the beginning of the project. In reality, subsequent decisions are tied

together during the entire period of a project. In this sense, Decision Tree (DT) can be an

effective tool to analyze a project that involves sequential decisions.

In Figure 2-4, the square represents the decision point where managers have to choose

what to do, while each circle represents probability points where the decision faces

different outcomes. The key idea is the combination of the decision and probability points.

One decision will reveal the consequential events. Managers can see what will happen,

and make decisions at the next step. The DT can make the underlying scenarios clear by
showing the links between each stage of a project, and help managers to find the strategy

with the highest value, which is also the expected value of a project.

Figure 2-4 Example of Decision Tree

The DT is particularly useful when we have discrete time stages for information and

decisions. Dixit and Pindyck 51 (1994) introduced the continuous version of the DT

(dynamic programming), which is available for problems with continuous flow of

information and decisions with the same theoretical framework as the DT.

Theoretically, the DT process will turn the possibility into reality, and reduce risk by

narrowing the range of potential outcomes. Therefore, we can use a lower discount rate to

recognize the change of risks. The DT, however, does not count this value, and uses the

'5 Dixit, A.K. & Pindyck, R.S., "Investment under Uncertainty", Princeton University Press, 1994
same discount rate throughout the life of a project. It simply sums up the cash flows in

each scenario as the expected NPV. This assumes that the project's risk is constant

regardless of the stage of a project. Another serious drawback is the probability of events.

Based on the definition of risk and uncertainty in Chapter 1, we may deal with risk in the

DT, but cannot deal with uncertainty, in which the probability of unexpected events is not

fully understood.

2.2 Implications

We have reviewed the current decision analysis methods to compare the different

approaches in valuing a project. The next step is to categorize these methods so as to

specify an appropriate valuation method for a railway project. Since every method has its

own advantages and drawbacks, managers cannot simply rely on a single way of valuing

a project, and each method should be regarded as a part of the information about the

value of a project. Therefore, it is useful to examine the effectiveness of each approach,

and examine how we can apply it to a railway project.

2.2.1 Effectiveness

In general, we can summarize the comparison of valuation methods with Figure 2-5. The

basic method is the NPV, which reflects the investor's expectations in a project. It is a

reliable valuation method if the project is stable and foreseeable in a financial sense. We

can use the NPV for a project that has very low risk and uncertainty. Also, the CAPM is a

technique to adjust the discount rate of NPV with the project-specific risk. The CAPM
can consider the risk premium on top of the firm's average risk, but has difficulty in

dealing with high risk and uncertainty due to the possible error in the forecast of cash

flows.

If the risk is high, we can use the DT. We can calculate the expected NPV based on the

probabilistic outcomes in the DT. We only have to determine the first step of an

investment decision, and make another decision according to the consequential outcomes.

Eventually, we can approach the optimal strategy, in which we can expect the highest

return from the project. If uncertainty dominates a project and the probability is unclear,

we can apply Sensitivity Analysis or Monte Carlo Simulation to consider the potential

range and distribution of outcomes.

When we face both high risk and high uncertainty, however, we will be at a loss in

dealing with them. Simply combining the above methods is not a good option in this case.

Each method has its own practical drawbacks, which may be amplified in combination.

In addition, the various criteria confuse managers in dealing with the problems.

Furthermore, managers need clear benchmarks for their investment decisions in terms of

the accountability to shareholders. The problem we are facing now is critical because

capital investments of a railway company 16 typically fit this segment in this matrix. We

need to build an additional valuation methodology in addition to these methods.

16 Figure 1-3
 Decision Tree How do we value?

T0
0 -------------f--------------- I

Yý 4 NPV
CAPM
Sensitive Analysis
Monte Carlo Simulation

Low -- [Uncertainty] -- High

Figure 2-5 Valuation Framework

2.2.2 Limitations of NPV

We investigate the effectiveness of NPV again because the above valuation framework is

based on the NPV method, and others basically expand the practicality of NPV in dealing

with risk and uncertainty. If the NPV has an inevitable limitation when valuating a

railway project with the existence of high risk and uncertainty, we have to consider

another approach to value a project.

The fundamental assumption of NPV is that money has a time value, and we can measure

it by using a discount rate. Here, reliable cash flows are more valuable than risky cash

flows. Therefore, risk always gives negative impacts on the valuation of a project through

the higher discount rate. The underlying value of a project is always the expected cash

flows, which can be adjusted by considering its risk. In practice, the following two

reasons highlight why we have difficulty in using the NPV as a valuation method of a

railway project where high risk and uncertainty are main concerns.
 1. The discount rate is too ambiguous to value a project under uncertainty.

The result of NPV depends on the choice of a discount rate as we discussed. Although we

have some guidelines for the choice of a discount rate (e.g., WACC, CAPM), these

approaches do not consider the dynamics of a project. Managers constantly obtain

information and make decisions during the whole process of a project under uncertainty.

Finally, we try to choose the best strategy by taking advantage of the newly revealed

situations. Using a constant discount rate ignores this reality and has a bias against long-

term projects and strategic decisions, which we focus on in this thesis.

2. The probability of outcomes is not clear in a complex system.

The NPV deals with risk as an adjustment factor, which is accounted for by a

combination of probabilistic scenarios and change of discount rates. In the real world of

uncertainty, however, the probability of each scenario will probably differ from what

managers originally expected. The effectiveness of the expected NPV, which depends on

this probability, is difficult to prove due to the reliability of this assumption. As we

discussed at the beginning of this thesis 17, one decision on a project may create

unanticipated side impacts on another project in a complex transportation system. This

fundamental characteristic makes this problem too hard for managers to simply use the

NPV in a railway project.

17 1.1.2 (Motivation and Objectives)

2.2.3 Option Approach

Conversely, if uncertainty is a major concern in a railway project, we can value such

phased projects as will confirm whether the next phase is worthwhile. The project has

strategic value on top of its cash flows when future is revealed. Consequently, flexibility

also has value if it gives opportunity to cope with unexpected events. As a result, the

value of these projects is actually higher than their expected NPV under uncertainty.

The key insight of this converse approach is that projects can be considered as kinds of

"options," which can increase the value of a firm when exercised with a proper timing. In

this approach, uncertainty or volatility can actually increase the value of a project,

contrary to the NPV approach. Previously, risk or uncertainty was a sort of barrier to a

project, but using the option thinking, which can consider risk and uncertainty as positive

factors, we can add value to a project. The manager can flexibly decide when and how to

invest for every potential project while avoiding the negative outcomes and taking

advantage of new opportunities. The value of this option will increase where the

uncertainty increases. This strategic approach enables managers to utilize uncertainty

instead of trying to mitigate it.

 U

°1
0
cI
o
>~

Uncertainty -+

Figure 2-6 Value of a project in different approaches under the uncertainty's

(Note: This figure is a conceptual model, and does not reflect a certain value of projects)

2.3 Proposed Strategy

The option approach can be an effective valuation method for capital investment of a

railway company, which has a large-scale and high-risk investment under uncertainty.

The option approach is not a totally innovative idea; or rather, it is trying to illustrate the

logic of experienced managers who have been acting to amplify good fortune while

mitigating loss in investment decisions without explicitly referring to "options." They

sometimes decide to invest in a project whose NPV is zero or minus, recognizing the

potential opportunities that may be realized by the investment. To understand the logic of

such a decision, we have to examine what characteristics of a project can increase its

value under uncertainty and provide rationale in valuing a project under uncertainty.

18This figure is originally drawn in this thesis

2.3.1 Flexibility

Managerial flexibility has its value under uncertainty. In dealing with uncertainty, it is

effective to build flexibility into the strategy and adjust decisions according to outcomes
19
that will occur after the initial decision. In this sense, Trigeorgis and Mason (1987)

advocated the concept "expanded NPV" in which a flexible project has an additional

value to the usual expected NPV due to its ability to approach the optimum strategy:

Expanded NPV = expected NPV + value of managerial flexibility

This equation explains the significance of the concept shown in Figure 2-6. When

uncertainty is not a concern, we can rely on the expected NPV. If uncertainty is critical to

a project, however, the expected NPV does not reflect the strategic value of the project,

and the value of the managerial flexibility becomes essential for the project.

2.3.2 Timing

One of the challenges of a railway project is the optimization problem in different time

scales as we discussed in Chapter 120. An option can be exercised only once, and

exercising an option may eliminate other potential options. We need to think strategically

to capture the long-term benefits while accommodating the short-term objectives.

Therefore, we have to consider the timing of investment and its opportunity value. The

opportunity value changes according to the economic condition in which a project is done.

19Trigerous, L. & Mason S. P., "Valuing Managerial Flexibility", Midland Corporate Finance Journal -Spring-, 1987
20 1.2.3 (Business Dynamics)
Managers have to meditate on the long-term design, while exercising the investment only

if the condition is favorable to the project. To the contrary, if the situation is not

promising, we should wait until the opportunity value improves, and finally make a

decision on the investment. The value of a project varies according to the timing of the

investment, and thus timing has value when we consider a project with the consequential

information and opportunities.

2.3.3 Valuation

The question here is how to value these strategies. If the flexibility is favorable in

managing a project, how much can we invest to obtain this advantage and measure the

changing value of opportunities? Real Options Value (ROV) provides the answer.

The ROV represents an application of the financial option theory to a real investment

project, which has option-like characteristics. If a project has flexibility to accommodate

the actual situations, the flexibility gives option-like ability to avoid bad situations while

taking advantage of new opportunities. This ability leads managers to recognize the real

value of flexibility under uncertainty, and thus demonstrates substantial improvements in

the investment strategy. The definition of "option" is the right but not the obligation to

take an action some time in the future, usually for a predetermined price and a given

period. In a project, a firm has rights to invest certain amount in exchange for the

intended assets at the most preferable timing just as put/call options could be exercised to

buy the underlying asset at the strike price.

The application of ROV to a railway project enables managers comprehensively to

consider the essential features of a railway project, which were described in Chapter 1. In

addition, ROV overcomes the limitations of NPV in valuing an investment opportunity,

such as the subjective probability, choice of discount rate, etc. We examine the

theoretical background and practical application of ROV in Chapter 3 so as to utilize this

technique for the strategic decision analysis for transportation systems.

 Chapter 3 Real Options

This chapter first reviews the theoretical background of real options, and discusses how a

railway company can apply real options to its capital investment strategy. We also

examine the valuation frameworks of real options to use it in a real project. In Chapter 2,

we pointed out the critical roles of flexibility and timing in a project, and introduced the

real options value (ROV) to measure their values. In Chapter 3, we go through the details

of ROV, which can be an effective tool for investment strategies under uncertainly where

the NPV does not work well neither as a valuation method nor decision analysis tool.

3.1 Theoretical Background

3.1.1 Project and Options

In the financial market, options give the owner the right to buy (call) or sell (put) a

financial asset (e.g., stock) at a predetermined price within a certain period of time,

without the obligation to do so. This right provides the owners with unlimited potential

gain in the future, while the lack of obligation protects them by limiting the downside risk.

They can exercise options and take a gain if they can expect enough returns. If the option

is not exercised, the only cost is the price of the option. This asymmetrical payoff gives

values to options.

Projects often contain option-like characteristics, and we can treat investment decisions
as exercising of options. Managers have options to invest in projects, but do not have to

exercise their options in every situation. In addition, they may wait for more information

and make decisions later. If the investment is irreversible, there is an opportunity cost of

investing now rather than waiting. Moreover, they can design a project with managerial

flexibility to adjust to unexpected events during the operating period. The value of these

option-like characteristics will increase when the potential outcomes of a project become

more uncertain.

This idea is quite similar to financial-options methodology. Managers identify options in

a project, and estimate the cost (exercise price) and cash flows (underlying asset) of the

project. If the situation looks promising, the manager exercises the option and invests in

the project; if it does not seem feasible, the only loss is the cost to acquire the option.

This is different from the traditional approaches (e.g., NPV) we reviewed in Chapter 2,

and allows a more strategic approach to projects.

3.1.2 Value of Options

The real options provide us with unlimited gain and limited loss for a project, and this

asymmetrical payoff is the source of the value of options. Figure 3-1 shows the value of

options (call-option) where "value of asset" is the expected cash flows, and "exercise

price" is the initial investment in a project.

 Value of Option (C)

t (S)

v
V lUk., ',J
l
f A
.• L •,\,.

T Exercise Price (K)

Figure 3-1 Value of Option (Call Option)

If the value of asset exceeds the exercise price, we can invest in the project and take

positive returns (payoff = S-K). When the value of asset is below the exercise price,

however, we do not have to exercise the option, and the payoff remains zero. Therefore,

the payoff of options can be the maximum of either 0 or S-K.

However, immediate payoffs may not reflect the full value of options because the option

value exceeds the payoff due to its expectation in the future. A higher payoff might be

obtained if we wait to invest until situations become more advantageous while the worst

outcome is limited to zero. Therefore, managers always have potential opportunities to

increase the returns on top of the immediate payoff. This reasoning explains why the

options value is not equal to the payoff line in Figure 3-1.

Here are the parameters, which actually determine the value of options:

S = asset price
K = exercise price at which the asset can be bought (call) or sold (put)
T = time remaining until option expires
13 = standard deviation of returns for asset (volatility)
R = risk-free rate of interest

In terms of the relationship to the option value, as the greater the asset price (S) relative

to the exercise price (K), the payoff increases, and the option value thereof increases.

Similarly, the greater the exercise price (K) relative to the asset price (S), the lower the

option value. Also, an option with a longer term to expiration is the same as an option

with a shorter term plus additional time. Therefore, the longer the term to expiration (T),

the higher the option value. Volatility of the underlying asset is critical since options have

a limited (zero) downside and a unlimited (positive) upside returns, and increased

volatility expands the potential range of expected returns, and increases the option value.

The risk-free rate is applied to discount the future cash flow in ROV, and a higher interest

rate will decrease the option value.

Table 3-1 Relationship between the option value and each parameter

Asset Exercise Time to Stnadard Risk-free

Parameters
Price (S) Price (K) Expires (T) Deviation (B) Rate (R)

Correlation + - + +
3.1.3 No-Arbitrage Approach

We review the underlying concept in the option-pricing problem. The critical concept

behind the ROV is "no-arbitrage approach" where two different assets with the same

payoff must have the same price. In other words, if we replicate an asset with another

portfolio that has the same payoffs (cash flows), the values of these assets are equal to

each other. Therefore, we can determine the price of an asset by simply summing up the

value of parts in the replicated portfolio. We will illustrate this concept with a simple

example21

Assume a firm has an opportunity to invest in an asset that requires an initial investment

of $100 now and yields uncertain payoffs one year later. The payoff is $125 in the "up"

state and is $80 in the "down" state. The firm can invest $100 and receive either $125 or

$80, but the probability is unknown. In this situation, one effective strategy is to buy a

call option, which gives the right to buy the asset in $100 next year. Then, the payoff is

$25 in the up state and $0 in the down state. Suppose another strategy is to build a

portfolio with the asset (X%) and $99 risk-free security (Y%), the price of this portfolio

is $100X-$99Y ( ). When the security becomes $100 after a year, the up state payoff is

$125X-100Y (2 ) and the down state payoff is $80X-$100 (3). If the two strategies have the

same value, (2)=$25 and (3)=$0. Solving these equations yields X=5/9, Y=4/9, and

(1)=$11.56.

21 Lecture on Finance Theory 2 (15.402 at MIT) by John Chalmers, November 18,2002

 Table 3-2 Three strategies for investment in a risky asset

Year 0 Year 1 (Up State) Year 1 (Down State)

Directly investing -$100 $125 $80

$25 ($125-$100) $0 ($80-$80)
Using call option -C (Price of option)
Exercise Not Exercise

Partly financing by risk-

$100X-$99Y $125X-$100Y $80X-$100Y
free security

The obvious result is that we can determine the price of an option by building a portfolio

that replicates the payoff of the option. In addition, the option price is independent from

the probability of outcomes, but depends on the possible range of outcomes (volatility).

Furthermore, since there is no risk in this transaction, we can use the risk-free rate as an

appropriate discount rate in valuing the cash flows.

In Chapter 2, we examined the effectiveness of NPV, and concluded with its two critical

drawbacks (ambiguous discount rate, and unclear probability). This no-arbitrage

approach is an essential concept in valuing options since it adjusts a situation so that the

risk-free discount rate applies and there is no need to concern about the probability of

outcomes.
 3.1.4 Complete Market Assumption

The rationale to apply a replicating portfolio in the no-arbitrage approach requires an

important assumption of "complete market." This assumption indicates that capital

market is sufficiently complete so that we can build a portfolio whose value is perfectly

equal to that of a project. In other words, we can exchange real options for a project with

financial securities as if the project could be treated as stocks. Therefore, we can apply

the financial options pricing model to the ROV method.

Dixit and Pindyck (1995) developed the real options approach with this complete market

assumption 22. However, every real option is not traded in a market. For instance, a

development right of an oil field can be traded, but an expansion right of a railway is

difficult to sell or buy. Myers (2003) explained this problem and argued that assuming

complete markets is enough to value real options23 . Since the purpose of valuing real

options is not actually to sell or buy the options but to measure the value in managing

investment activities, managers can measure the value of real options as if they were

traded in a market, and price them based on the no-arbitrage concept. We can understand

this reasoning when we review the concept of NPV, in which we assume that non-traded

assets (projects) can be valued by using the same discount rate of traded assets

(securities) that have equivalent risk and returns to the projects.

This is a critical assumption in applying the ROV to transportation systems where real

options are not traded. We examine this assumption to clarify the applicability of ROV

22 Dixit, A.K. & Pindyck, R.S., "The Options Approach to Capital Investment",
Harvard Business Review, 1995
to this research. We explain the theoretical correctness of this assumption by the

following two steps.

1. We can price non-traded real options by the value of comparable traded securities.

This assumption is satisfied when using a pricing theory in which the value of an asset is

estimated by comparing it with similar assets with known prices. For instance, we

generally consider the price of a similar deal when we try to value a used car although we

do not actually trade the two cars. Likewise, when managers value a potential project,

they can consider a similar financial investment (security), which resembles the project in

risk and returns even if they do not sell/buy the project by the securities. This

approximation pricing is actually one of the most traditional and widely accepted theories

in the finance field. Theoretically, the difference between traded assets and non-traded

assets does not matter in pricing the value in this process.

The typical application of this theory is the CAPM 24 by which we adjust the discounted

rate of a project based on the relationship between the project and market portfolio. The

relationship is characterized by the parameter 13 (covariance). The CAPM actually prices

the non-traded asset (project) based on the comparable values of traded securities in the

market. The CAPM is currently applied to many theoretical discussions in valuing a

project. If we use the CAPM in a real project, we can apply the same pricing theory to

ROA and we can price real options by market securities.

23 Lecture on Real Options Seminar (15.975 at MIT) by S.C. Myers, February 10, 2003
The CAPM was, however, originally built to price a market asset although it is currently

applied to a real project. Therefore, there is a practical problem in using the CAPM in a

project. According to the CAPM, the price of an asset is determined by its /3:

/3: cov (asset, market portfolio)/ 2 market portfolio

However, the / is difficult to determine in a project as we discussed in Chapter 1.

Luenberger (1998) proposed Correlation Pricing Theory (CPT) to practically use the

concept of the CAPM in a project, and verified its theoretical accuracy and correctness in

the financial literature 25 . The CPT has a similar form as the CAPM, but we do not have to

estimate the 3 of a project. We can regard the CPT as a more convincing example of the

approximation pricing theory in real options problems.

The formula of the Correlation Pricing Theory is:

V=1/R{E(y)-cov(y,m)[E(m)-Pm-R]/ u m2

where

V: value of a non-traded asset "y"

R: 1 + r (r: risk-free rate of interest)

E (y), E (m): expected payoffs of "y", and security "m"

24 CAPM=Capital Asset Pricing Model (2.1.2)

25 Luenerger, David G., "Investment Science", Oxford University Press, 1998
 Cov (y.m): covariance between the "y" and "m"

Pm: current price of "m"

a m: variance of "m"

In this formula, "m" is a security portfolio (not a market portfolio), which is the most

closely correlated with "y". The correctness of the CPT is shown by the fact that it gives

precisely the same result as the CAPM if both were exactly calculated, so there is no-

arbitrage between the CPT and CAPM. The only difference is that the market asset in

CPT does not have to be the market portfolio. The form of CPT is more convenient by

avoiding the 3 , and showing the theoretical process in pricing a non-traded asset with

known prices of traded securities. The CPT can price a non-traded asset by adjusting the

price of comparable marketed assets. Appling the CAPM and CPT in the valuation

problem of a project indicates the correctness of pricing non-traded real options by

market securities, which replicates the payoff of real options.

2. We can build a security portfolio, which completely replicates the payoff of real

options.

This assumption is readily satisfied in any linear pricing model, such as the CAPM or

CPT. The linearity of the CAPM and CPT can be shown by the fact that they use the

projection of payoffs into a norm linear space, known as a Hilbert space. The random

payoffs of an original set of "n" assets can be replicated as corresponding "n" vectors in a
 Hilbert space whre every vector (v) can be written with the norm vectors (ei) and their

coefficients (ai) as:

v= Z (ai.ei)
i=1

This linearity of projected payoffs is an underlying concept of the CAPM and CPT. The

important result of the linearity is that the value of an asset can be stated by a number of

norm securities rather than single optimal portfolio. This linearity enables us to price a

collection of assets by using a combination of securities. If the non-traded asset is

approximated by a linear combination of securities, the asset is assigned a price equal to

that of the corresponding linear combination of securities.

The number of distinct securities is the number of risks and payoffs, so any combination

of risk and payoff can be matched in countless ways. Providing completeness with

respect to any conceivable state variable is not possible but, still, we can ask the

common-sense question: "Does the existence of real options allow investors to expand

the range of attainableportfolio risk characteristicsin any significant way? If the answer

is no, we can replace real options with existing securities, so the market is complete"26

26 Myers, Stewart C., April d,

2 3r 2003
3.2 Application to Railway Projects

Brealey and Myers (2000) classified the common and important real options found in

capital investment projects in the following manner27: expansion option, abandon option,

wait option, and switch option. In this section, we examine how we can use these options

based on the characteristics of a railway project, which we have discussed in Chapter 1.

3.2.1 Expansion Option

This option makes follow-on investments if the immediate project succeeds. Managers

often invest in a project for strategic reasons. Such a project gives not only its own cash

flows but also expansion options, which can take advantage of new opportunities and

provide additional gain in the future.

This option is very critical for a railway company since the railway system often has an

optimization problem in different time scales 2 8. Managers are required to consider the

long-term plan when they invest in a project. In another word, managers have to design a

project so as to leave options to expand the system in the future. Considering the long-

term plan in a project may increase its cost. However, we can obtain expansion options

by paying this additional cost in the initial investment. In this case, the price of options

can be the additional cost, and the underlying asset is the potential cash flows in the

future projects. Using the real options approach, managers can recognize if the additional

cost is worth paying in this case.

27Brealey, Richard A. & Myers, Stewart C., "Principle of Corporate Finance", Irwin McGraw-Hill, 2000 pp.619
 Moreover, a railway project has significant side effects on the land use. Investing in a

project may bring another opportunity to expand the business in the future. For example,

building a new station or line will encourage regional development through the new

transportation service. Therefore, the population around the railway system may increase,

and the incremental demand enables us to expand the railway system by investing in

another project. Thus, we can acquire expansion options through the initial investment. In

this case, the price of this option is the initial investment of the immediate project.

Managers can reasonably value the initial investment by considering the value of

expansion options in addition to the cash flows of a project.

Another example is the business model that combines the real estate with the railway

system. If a railway company purchases the land around the proposed new station or line

before the project, the company can expect capital gain by developing the land later. Thus,

the company can reserve the extended benefits of the project within the company to

recover a part of the railway project. In this case, the railway project provides expansion

options to develop the real estate, and managers can invest in a railway project expecting

potential profit from these options. Furthermore, the difference between the real estate

and the railway in terms of the sensitivity to the market fluctuation helps managers to

diversify their business activities under uncertain business circumstances.

28 1.2.3 Business Dynamics

3.2.2 Abandon Option

A project has additional values if managers can abandon it depending on to the situation

during the operating period. As we need to consider the value of expansion options in

case of success, we have to recognize the value of abandon options, which limit the

negative impacts in case of failure.

In a railway project, however, it is usually difficult to exercise this abandon option due to

the social responsibility as infrastructure. In Chapter 1, we discussed institutional risks in

a railway project, and argued the difficulty in using an exit strategy even if the demand is

below the forecast. In addition, the railway system requires high fixed cost to realize its

low operating cost, so cannot drastically reduce its total cost. These facts make it difficult

to limit the downside impacts once the operations start.

This is a critical difference between a railway project and other capital investments where

abandon options can be obtained with a certain cost or contracts, so as to manage the risk

exposure under uncertainty. Therefore, in a railway project, managers have to make

effective use of other options to manage the risks. For example, dividing a project into

several phases costs more than investing in the whole project at a time. This strategy,

however, gives the similar effect to abandon options since not entering into next phases

can abandon the rest of the project when the initial project fails. This is actually using

expansion options to recover the disadvantage of not holding abandon options in a project.
3.2.3 Wait Option

This option enables us to wait and learn before the investment. The value of options

exceeds the immediate payoffs due to the asymmetrical distribution of the future in

Figure3-2. Using options captures the positive results while avoiding the negative

outcomes. Although a project has zero or negative payoffs, it still has a certain value if

we can apply the option to wait. We can exercise the option only when the situation

changes and positive payoffs are available. Similarly, we do not have to invest in a

project only because the project makes positive returns. If we wait on the investment, the

project may have higher profit in the future than investing now.

However, projects provide cash flows just as stocks give dividends. The right not to

invest now has values, but we lose the potential cash flows, which could be gained while

waiting. Therefore, managers have to compare the value of wait options with the cost of

holding (or buying) such options.

Value of Option (C)

Value of Asset (S'~

T Exercise Price (K)

Figure 3-2 Asymmetrical distribution gives values to options

In a railway project, wait options work to determine the timing of investment. For

example, station development projects create new business opportunities for a railway

company. Developing a business (e.g., retail shop, restaurant) in a station can provide

more convenient service to passengers, while considering them as potential customers.

However, creating a business space in a station is expensive because the station was

originally designed only to accommodate facilities required for operating railway systems,

and changing the design has various constraints in order to maintain the current

performance of systems. The business has to be highly profitable to effectively use such

an expensive space. Therefore, managers have to analyze the market situation with wait

options, and invest in projects only if they can expect enough profit from the business. In

general, station development is advantageous to railway companies because most stations

are their own properties where competitors usually cannot enter. This wait option allows

managers to determine the proper timing and maximize such advantages in investment.

Another example is a modernization project. Managers try to enhance the performance of

systems or reduce the maintenance cost by replacing the existing facility with an

advanced one. It is critical for a railway company to introduce new technologies to

increase its competitive advantage in the market. However, new technology always has

uncertainty as we discussed technology risks in Chapter 1. This wait option allows

managers to wait and see the effectiveness of the new technology, while holding the right

to use the technology and modernize their system. This wait option gives a certain

benchmark to managers who are facing the investment-timing problem in a project.

3.2.4 Switch Option

The switch option provides the ability to vary the output or production methods

according to the situation. If a firm can flexibly vary its products by exercising this

switch option, the firm can be less vulnerable to the market fluctuation due to its

adaptability to the market trends. Similarly, a firm with dual production methods is more

valuable than with a single method due to its ability to stabilize operating costs. This

built-in flexibility has substantial values under uncertainty where the market demand or

resource cost is frequently changing. However, the flexible design makes the system

complex and expensive. Managers have to value the switch option in making a decision

on whether the flexibility is worth investing in.

In a railway project, there exist high market risks and the fixed system of railways

prevents managers from easily adapting their system to the market. Therefore, this option

is critical in designing a railway system. The railway has the efficiency in its operation in

exchange for the fixed cost and location of facilities. Therefore, building flexibility in a

railway system is effective in taking the benefit while avoiding the negative side of its

characteristics.

For example, the switch option is useful in designing a railway network. Managers often

face difficulty in deciding if they should build additional tracks or points in their network.

Although it costs, having such extra facilities gives flexibility to the network, which

allows the railway company to operate additional trains in a peak period or take

alternative routes in case of accidents.

Another example is the design of rolling stock. If we design rolling stock so as to drive

both on high-speed and conventional rails (e.g., dual gauge car), we can flexibly operate

such rolling stock to accommodate the fluctuation in trip demands. Moreover, passenger

rail, especially high-speed rail, has different customer segments such as business trip and

family travel. Flexible design can easily capture the preferences of different types of

customers by switching the interior design while using the same car. Recognizing switch

options in these examples gives practical benchmarks in making an investment decision.

3.3 Valuation Framework

We reviewed the theoretical background of the real options value (ROV), and how we

can make practical investment decisions by applying the ROV to a railway project. We

now examine the valuation framework of ROV. The attractiveness of ROV is its

independence from a choice of discount rates and subjective probabilities. A portfolio of

securities that replicates the cash flows of a project quantifies the value of a project as

well as the optimal timing of investment to maximize the value. The ROV is an

application of the financial options theory, and there are two major valuation frameworks

to value financial options: Black-Scholes Formula and Binominal Model. We compare

the two methods and build a valuation framework to answer the ROV problems in a

railway project.
3.3.1 Black-Scholes Formula

Black and Scholes developed a pricing method of options as the following compact

formula. Although the formula mainly deals with only European calls 29 (without

dividends), it has been refined since then and widely used in the financial market.

The Black-Scholes Formula is:

Co = So-N (dl) - X-e-rT'N(d 2)

di = [ln (SO/X) + (r+ a 2/2) T] / ( a T1/2)

d2 = dl-( -T1/2 )
where

Co: current option value

So: current stock price
N(d): probability which will be less than "d" in a normal distribution
x: exercise price
e: base of the natural log. (=2.71828)
r: risk-free interest rate
T: time to maturity of the option in years
In: natural log function
a : standard deviation of annualized continuously compounded rate of return on
the stock.

This formula is similar to C = f (asset price) - g (loan). This formula represents the

concept in which an option has value equal to the replicating portfolio (buying a stock by

borrowing) where the standard deviation gives a measure to the possible range of payoffs.

29European options can be exercised only at the expiration while American can be exercised anytime before the
expiration.
This valuation method is based on the statistical movement of stocks, and the movement

is assumed to be random around the trend line.

The practical problem of the Black-Scholes Formula is, however, the focus on European

options model where we can exercise options only at the expiration. As we described, a

variety of options can be applied in a railway project, and many of them should be

exercised whenever the situation is optimal for the investment, even prior to the

expiration. the Black-Scholes is a compact and effective tool to price an option, but we

need to apply another model in pricing options in a railway project.

3.3.2 Binominal Model

The Binominal Model assumes many periods in exercising an option. This model varies

the value of underlying assets based on the volatility in a given period, and rolls back the

values from the expiration. For each period, the model compares the value of immediate

payoff and holding of the option, and decides whether the owner should exercise the

option. Thus, this model determines the optimal strategy of exercising options, and the

value of options at the time of investment by summing up the values throughout the

whole period.

Value of holding the call option for the next period is:

Ch = [p*Cu + (1-p)*Cd]/(1+r)

where
 p: risk-neutral probability
cu: value of option at end if up
cd: value of option at end if down

The risk-neutral probability is a key idea for connecting sequential periods under

uncertainty. We can calculate this probability by applying the no-arbitrage concept.

According to this concept, we can replicate the payoffs of an option as if we bought an

asset by borrowing a risk-free loan.

Assuming the asset share is X, and loan share is Y, the replicated payoffs are:

X-Su + Y-S(I+r) = Cu

X-Sd + Y-S(I+r) = Cd

where

u: parameter to determine the S at end if up (e-exp ( a A t))

d: parameter to determine the S at end if down ( e-exp (- a f At))

r: risk-free interest rate

When we solve the equations, the option price (= value of the portfolio) is:

Ch = [(1+r-d)Cu+(u-(1 +r))/Cd](1 +r)/(u-d)

We can rewrite the above formula by using a factor "q"

Ch = [qCu+(1-q) Cd]/(l+r)
q = (1+r-d)/(u-d)
This process leads us to adjust the investment situation as if we could calculate the

expected value of options with binominal probabilities q and (1-q), which are risk-neutral

probabilities.

Finally, the value of options is the maximum of their immediate exercise, holding for

another period, or zero:

C = max {s-k, [p'cu + (l-p) 'Cd]/(1+r), 0 }

where

s: value of asset
k: exercise price

C = max{ s-k, Ch} q

Ch
Cd
(1-q)

I- discounted by "r"

Figure 3-3 Expected value of option

The above figure is similar to a decision tree, and we can go backward from the last to

first period to value C. In additions, we can use risk-neutral probabilities instead of the

assumed probability, and apply risk-free interest rates as discount rates.

The Binominal Model approximates the movement of the value of underlying assets as a

sequential increase and decrease. This model actually applies the concept of a replicating

portfolio over many periods, and provides a flexible and transparent method to option

valuation in a railway projects.

It is hard to argue that one or another is a better decision analysis method, and we need to

apply an appropriate valuation method based on the characteristics of a project, the

availability of data, and the purpose of analysis. In this sense, we proposed the

effectiveness of ROV in capital investment projects of a railway company, and explained

the excellence of the Binominal Model in pricing the options. In the next chapter, we will

analyze a real case to illustrate how we can make a strategic decision in investment

problems by using the ROV.

 Chapter 4 Case Study -Station Development-

This chapter analyzes station development projects as a case study. As we classified

railway projects in Chapter 1, a station development project is located at the "high-risk"

and "mid-scale" segment in the project portfolio 30 . Developing several stations,

specifically in downtown areas, can be a large-scale investment program, and requires

managers to make strategic decisions. We proposed an example of real options to a

station development project in Chapter 3. This project is an interesting case as an

application of strategic decisions analysis, which we have developed in this research.

In addition, this project plays an important role in developing transportation systems.

Stations are key facilities for the railway system, and effective use of stations is critical

for a railway company to properly operate the system and provide better service to

customers. Stations also have potential for retail business because of their advantage in

locations where many passengers gather. Furthermore, stations are essential facilities in

urban planning. Combining a station development with an urban renewal project may be

an effective investment to achieve a revival of a downtown and real estate development.

These aspects of a station make the project attractive, but various barriers (e.g., high

development cost, design constraints) lead managers to difficult investment decisions.

This case describes a strategic decision, which a Japanese railway company is now facing.

The company considers investing in station development projects in the Tokyo

metropolitan area, and trying to value them with the NPV analysis. Managers

acknowledge the strategic importance of the projects, but also have difficulty in

explaining their reasoning with the NPV analysis. We examine how real options provide

a rational answer for this problem.

4.1 Case Description

4.1.1 Company Information

In accordance with the provision of the Law for Japanese National Railways

Restructuring, the Japanese National Railways (JNR) was privatized into six passenger

companies and one freight company on April 1, 1987. The East Japan Railway Company

(JR East) is the largest passenger railway company in Japan and serves the Tokyo

metropolitan area and the eastern part of the main island of Japan. JR East operates 70

railway lines (including three high speed lines), 1709 stations, and serves more than 16

million passengers daily.

30 Figure 1-3
 he

Figure 4-1 Operating Area of JR East31

Table 4-1 shows the financial highlights of JR East. The company inherited a part of the

large debt of JNR in exchange for the operating assets when it was privatized. This debt

was so huge that repaying the interest (around $1,600M annually) is one of the major

financial burdens. Therefore, it is critical for the company to reduce the debt by

effectively controlling its investment activities. However, as we discussed in Chapter 1,

capital investment is a source of cash flows in the future, and indispensable to compete in

the market. Therefore, managers are required to make a strategic decision on their

investment activities based on a solid understanding of the business dynamics of their

projects.
 Table 4-1 Financial Highlights in the FY 2001 (million dollars)32

Income Statement
Operating revenue 19,123
Operating income 2,378
Net income 358
Depreciation 2,421
Free Cash Flow 33 2,627
Balance Sheet
Total Asset 52,799
Long-Term Debt 32,931
Shareholder's equity 6,998

The company is involved not only in railway operations, but also in other business such

as retail sales, shopping centers, hotels, real estate development, etc. These businesses

have synergistic effects with the railway business, and help the company to diversify its

portfolio.

JR East recently announced a medium-term business plan, whose completion is targeted

for the fiscal year of 2006. One of the main strategies to achieve the plan is "station

development", which enhances the business activities at stations by optimizing the

allocating of facilities both from the railways and other businesses' points of view.

Although the Japanese economy is still uncertain, and not fully recovered from the

recession in the late 1990's, the company is finding opportunities in the currently

evolving deregulations and restructuring of industries, and decided to begin with three

projects in 2003.

31Corporate Guide of JR-East in 2001

32 Annual Report in 2001, JR East
4.1.2 Station Development Projects

These projects change stations to accommodate both railway and business activities.

Previously, most stations allocated their space only for railway facilities. However, the

railway system has a long history, and advances in technology and management now

enable the system to operate with smaller space and less people. Therefore, there is an

opportunity to review the layout of stations and create business space, which can generate

cash flows in addition to the railway business. Moreover, reviewing the layouts and

adjusting them to the current operations make the system more efficient. These projects

are effective to enhance the performance both of the railway and related businesses.

Figure 4-2 is an example of the projects. One approach is to renew the interior of the

station, remove the unnecessary facilities, and create a shopping mall (a). Another

approach is to construct a building over the station (b), which is an effective way of

developing real estate in a limited area such as downtown. These projects actually change

the station from a space to pass through into a place to gather.

4
Case (a) - before
Case (a) - after
Case (b)
Figure 4-2 Examples of the station development

33 Cash flows from operating activities and cash flows from investing activities.
JR East has already developed some suburban stations by taking one of those methods,

but the new series of projects, which JR East is now planning, more drastically change a

station by combining various approaches and making large-scale investment in its hub

stations in downtowns. These projects are very important to achieve the medium-term

business plan of JR East, but also quite uncertain due to their complexity and lack of

precedents. Managers are required to make a strategic decision so as to mitigate the risk

and capture the benefits while moving fast and creating future opportunities under

uncertainty.

At this time, JR East is considering the development of those stations, which have more

than 200,000 daily passengers in the Tokyo metropolitan area. There are around 30

stations, which meet this requirement, and managers narrowed the potential projects

down to three as the initial investment in this strategy. Managers recognize the

importance of this investment, which opens up the potential of this strategy. However, the

uncertain business environment, and relatively expensive construction costs make it

difficult for them to make an investment decision with the NPV analysis. The following

figures and tables represent the information for the proposed projects A, B, and C.

Figure 4-3 indicates the ranking (in terms of the number of passengers) and locations of

the three projects. Among the targeted stations, the proposed projects are the 8th , ,
1 7th 2 7 th

largest stations, and located in the northern, western, and eastern areas in the Tokyo

metropolitan area. Providing the top five stations are so large and complex that each

project can be very specialized, projects A, B, and C are good representatives of the
 it is
potential projects in terms of their scale and location in this company. Therefore,

very critical for managers to succeed in the three projects, and move on to the potential
initial
projects by applying similar plans, whose effectiveness can be verified by the

success.

1,000 people
1,600

1,400

1,200
1,000

800

600

400
200

0
3 5 7 9 11 13 15 17 19 21 23 25 27 29
1
ranking

Figure 4-3 Ranking and location of projects A, B, C

Table 4-2 summarizes the data for projects A, B, and C. The common strategy is to create

a business space in a station by reviewing its layout and constructing buildings over or

adjacent to a station. However, the accessibility to the site (= station) significantly

impacts on the cost and period of the construction work, and the local market influences

the expected income of the new business. We explain the details of the three projects in

the following plans.

 Table 4-2 Project Data

Project A Project B Project C

Development Area (yd 2) 4,770 8,556 2,050
Retail shops Retail Shops Retail Shops
Restaurants Hotel Caf6 & Restaurants
Construction Cost ($M) 39.6 40.7 11.4
Construction Period (years) 3.0 2.5 1.0
Expected Net Cash Flow ($M/year) 3.2 3.2 1.4

Project A has the largest number of passengers among the three projects. JR East will

create two business areas in this station (Zone 1 and 2), and operate retail shops and

restaurants there. The advantage of this project is the large local market, by which the

new business expects high sales compared with other projects. However, the construction

work has to be done over many busy tracks, and will be very expensive.

I I Tracks I
Zone 1 Retail Shops
Restaurants I I I Area 4,040yd2
1F:730 yd2 I I Sales/yd 2
$17,624
Initial Cost I I I
. I I I I Margin 4.13%
~L.3M
.3M
Existing Station Net Cash Flow (/y) $2.94M
IF: Track
2F: Station
~:::: Zone 2 Restaurants
:·:·:·ed·:·:

Retail Shops Area 730yd 2

":~'''' 2F: 1720 yd2 Sales/yd2 $8,096
3F: 2320 yd 2
Initial Cost Margin 4.32%
S$37.1M Net Cash Flow (/y) $0.26 M
I I I I Annual Net Cash Flow
I I I I
$3.20M

Figure 4-4 Construction and Business Plan34 of Project A

Project B is located at the center of the eastern area of Tokyo, and one of the major

stations for changing commuter trains. JR East will operate retail shops in zone 1, and a

hotel in the zone 2. The hotel space is very limited, and its construction work has to be

done by using the zone 1 area. Therefore, the construction of the two zones has to be tied

together and difficult to separate. JR East now operates small-scales hotels, which benefit

from high operating rates and margins due to their specialized service for business

customers, and considers a similar hotel in zone 2.

Zone 1 Retail Shops

Retail Shops Area 4,470yd2
2F: 1990 yd 2 2
3F: 2180 yd2 Sales/yd $12,579
4F: 300 yd 2 C\1
\i Margin 3.20%
inlll U•.U
$28.0M - Net Cash Flow (/y) $2.36M
Zone 2 Hotel
Existing Station
1F: Track Area 4,096yd2
2F: Station Rooms 130
Avg. rate $130
Exp. Vacancy 15%
Hotel ..............
2
340 vd /F Margin 12.26%
Initial Cost \ \ Net Cash Flow (/y) $0.84M
$12.7M Tracks
Annual Net Cash Flow
$3.20M

Figure 4-5 Construction and Business Plan of Project B

34 The forecast of sales, margin, and net cash flow are based on market surveys of a similar business in the local market
Project C is the smallest project. It is, however, located in an area, which has many

offices and community facilities around the station. Marketing surveys indicate that the

number of customers is not so large, but the purchasing power per person is very strong

in this area. Furthermore, its construction constraints are not so severe as the other two

projects, and JR East can save construction costs in this project.

Retail Shops
Retail Shops Area 1,170yd2
2
1F: 600 yd
2F: 570 yd 2 Sales/yd 2 $21,091
Caf6& Restaurants Margin 4.30%
2
3F: 480 yd Net Cash Flow (/y) $1.06 M
4F: 400 yd2
Existing Station -1
Initial Cost IF: Track F- Caf6 & Restaurants
4a
$11 .4M 2F: Station Area 880yd2
Sales/yd2 $9,455
Margin 2.91%
Net Cash Flow (/y) $0.31 M

Anrnrnl nPt cnah flnP 'l1 37M

Figure 4-6 Construction and Business Plan of Project C

4.1.3 NPV analysis

In JR East, managers used to consider the payback period of a project to make an

investment decision. However, they has recently decided to use the NPV as the primary

benchmark for valuing an investment in accordance with the new accounting system,

which regards the cash flow as an important criterion for a corporate and investment

activities. Managers, however, still have difficulty in using the NPV for station
development projects. In addition to the drawbacks of NPV, which we pointed out in

Chapter 1, the company has the following practical problems in using the NPV:

1. Applying the weighted average cost of capital (WACC) to determine a discount rate

is difficult because the railway business is still dominant in the corporate portfolio of

JR East. Managers have much experience in the railway business, and it is more

stable than a new business. Therefore, using the WACC, which reflects the average

risk of a firm, underestimates the risk in a new business. Also, determining the risk-

premium to adjust the WACC is challenging due to the lack of historical data and

complexity for the project. In addition, JR East is now trying to reduce long-term debt,

and the capital structure will drastically change during the project period. These facts

result in the difficulty in using the cost of capital to determine the discount rate in the

proposed project, and force managers to take another approach.

2. Forecasting the cash flows over a long period is difficult because the new business is

quite sensitive to market conditions, and the forecast error usually expands during the

project period. Managers can forecast the cash flows for a few years based on the

market survey and their experience in other projects. However, the long-term forecast

is too ambiguous to use in valuing the project.

In brief, managers can properly estimate neither the expected cash flows nor the risk

premium rate of a station development project under the present circumstances. To solve

these problems, therefore, they decided to conservatively estimate the cash flows and use
the risk-free rate in calculating the NPV. In other words, they consider the project-

specific risk not in the discount rate but in the forecast of cash flows. In practice,

managers are using the following policy to value station development projects with the

NPV analysis:

1. Reduce the expected cash flows to the minimum level. In other words, estimate

conservative cash flows that managers can expect to be achieved even if the situation

turns out in failure. Moreover, managers do not expect the growth of the cash flow; or

rather, they assume the minimum level will continue during the whole project period.

Cash Flow

I Possible range of cash flow

+-Minimum Level of Cash Flow

Time

Figure 4-7 Minimum Level of Cash Flow

2. Use 3% discount rate, which is the interest rate on the long-term debt (20 year bond)

for JR East. Repaying the debt is always a risk-free alternative of the investment, and

reducing the cash flows to the minimum level and discounting it by a risk-free rate

has a similar effect to using a risk-premium rate to discount the expected value of

cash flows.
 1), and
3. Assume 15 years as the project life in case of a business within a station (zone
is the
20 years when adjacent to the station (zone 2). The reason for this difference
additional
possibility of reviewing the layout of a business space due to the
zone 1 has
requirement of the railway operations in the future. Therefore, business in

slightly shorter project life than that in zone 2.

to the
4. Do not consider the salvage value of a project. These assets are closely related

railway operations, and are not easily converted into cash in a market. Therefore,

managers only consider the cash flows of a project to avoid overestimation.

Using these steps, managers calculated the NPV for each project as seen in the following
figure35
figure

M NPV (cash value)

6
5
4
3
2
1
0
I
-1 1

Figure 4-8 NPV of Project A,B, C(without salvage value)

The obvious result of the NPV analysis is that only project C makes sense in terms of the

financial returns on investment. Managers, however, intuitively recognize the potential of

35 For details, please see the spreadsheets at the end of this section
 these projects, and need to explore the opportunities to achieve their medium-term

business plan by 2006. Therefore, managers discussed the NPV analysis,

and tried to
increase the value of projects by quantifying their intuitive expectation.
They finally
concluded by adding salvage values of projects although the investment policy
of JR East
does not usually consider the salvage value of a similar project.

Managers estimated the salvage value based on the discounted book value
(after
depreciation) of the business assets in projects A and B. Figure 4-9 shows the result
of
this new analysis.

NPV ( Cash Value + Asset Value)

Figure 4-9 New NPV analysis (with salvage values in project A and B)

As the result, the NPVs of project A and B tuned out to be positive, and managers could

provide the rationale for investing in the three projects. The following spreadsheets show

the calculation of the NPVs in this analysis.

 Table 4-3 Spreadsheets for the NPVs of project A, B, C ($M)

PRJECT A
Zone 1 .1 .- I .,I .~
Year 0 1 2 3 4C 5 6 71 8I 9 10V 111 12 141
144
151 161 171

Cash Row
71O ). 8
2.94 2.94 2.94 2.94 o2A4 o2a9 2.94 2.94 2.94 2.94 2. . 2.
294 .H1 2y4.9
4

Asset -5.30 -10.60 -2120 I I I I 13.

2.69 ,,, 1.
1.941 1.89b 1.8;: 10.11
.,.I ,,,I
PV -5.30 -1029-19.98 2391
2692 32 225 2.19 2.12 - .232 1.94 1.83
NPV 6.91 Cash Value -2.00 Savage VauE 8.91

Zone 2
Year 0 1 2 3 4 5 6 8 S 10V 11 121
12103131
111 026R 13, ^^^ 17 181 19 ~UI
20 21l
026
Cash Flow 026 026 026 026 026 026 O2 026 f
036
VLV
•R
V·LV VLVI VLV
()0i6
026
(02615
026 026 026
t
1 1.64,
_
6 620

I I I I I I I
Asset 2.50 0.91I -. I - • U.P
WI U. 14
1 1.02
-

PV 0.0 -2.43 024 023q 023 022 021 020 020( 0.19 0.181 0.181 0.17' U.15I U.15I 0.15 U.14

NPV 221 Cash Vaue 1.30 Savage Vaue 0.91

ICash Vlue -0.70 Sdvae Vau 9.821

PROJECT B
Zone 1
Year 0 1 '
LW~t
4 51 0.18 71 8 9 10 117 I
'

I
'

_r
15 1
Cash Row 2.36 2.36 2.362.36 236 236 236 2.36 I 1 r
Asset -933 - 18.67 -I
I
..,.

I I 1
I2-
11.26
PV -9.33 - 18.12 .V- 2.10 2.04 1.981.92 1.86 1.81 1.76 1.70 1.51 8.49
NPV 7.45 Cash Value 0.18 Salvage VauE 7.27
7j~m• )
ZM 2 ,, I
71
Year 0 1 2 v1 I 8
I
9I i10
..........
1-·
1 121 131 141 151 161
-
171 19 20 21
Cash Row 0.84 0841 0841 0841 0.8410.8410.8410.8410.841
08 0.84I 0.841 0.84 0.841 0.84 0.841 C.84
^ ^.,1 I I I I I I I I I I I I 33.29
L AssetL -12.71 I I I I I I I I I I, I, I 1 ... - .. . I ri-. 2.22
I DUI I -cn
10"•n7 0 :6I A66
, A064• R62 n mi, 059I 0571 0551 0.541 0.521 0.511
- ' "2 O 05 04 0..46
0.481 0.4UI

NPV 1.55 Cash Value -0.27 Salvage Vau 1.82

Cash Vaue -0.09 Svage Vdu 909

PROJECT C
Year 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
Cash Flow 1.37 1.37 1.37 1.37 1.37 1.37 1.37 1.37 1.37 1.37 1.37 1.37 1.37 1.37 1.37
Asset -11.40
PV -11.40 1.33 1.29 1.26 1.22 1.19 1.15 1.12 1.08 1.05 1.02 0.99 0.96 0.94 0.91 0.88

NPV 5.00

Note 1: The "Cash Flow" indicates the net cash flows from the business operations, and "Asset"

means the cash flows from initial investments and salvage assets.

Note 2: The initial investment disperses during the construction periods.

Note 3: Zone 1 has 15-years operation, while zone 2 has 20-year.

92 20
4.2 Real Options Analysis

4.2.1 Results of the NPV analysis

The obvious result of the NPV analyses is that all projects have positive NPVs, but only

project C makes sense if managers do not count the salvage value and simply consider

the cash flows in valuing a project. Projects A and B are difficult to invest in when

managers focus on the cash values of a project as JR East usually does.

However, managers are going to launch all three projects based on the decision supported

by the NPV analysis, which will consider the salvage value of a project as an exception to

their valuing procedures. The reason managers value the salvage value is that they find

differences between their intuition and the NPV results, and modified the analysis so as

to justify their decisions. This happens because NPV does not reflect the strategic value

of a project, and cannot be a practical tool in making a decision on strategic projects,

specifically under uncertainty. Subjective use of NPV, however, makes the analysis

meaningless due to the lack of the logical consistency, may lead managers to wrong

decisions, and result in loss of their competitive advantage in the market.

The following two points are concerns about the NPV analysis, which JR East is

currently applying to the projects.

1. Project A and B value their salvage assets although JR East does not usually consider

the salvage value of a similar project. The salvage value is actually a part of the NPV
 according to finance theory, but the risk aversion policy of JR East usually forces

managers to focus on the cash value to avoid overestimating a project. One of the

reasons is that the expensive construction cost of a station project increases the book

value of assets, so the market value tends to be less than its book value. Also, some

assets are related to the railway operations, and difficult to sell. Therefore, managers

cannot always expect the same cash flows as the book value of an asset. Whether a

firm considers a salvage value, of course, depends on the investment policy and the

type of projects. However, projects A and B are exceptions to the valuation policy in

JR East, and managers have to justify their reasoning for making exceptions.

2. NPV does not reflect the strategic value of the projects. The reason that managers

counted the salvage value might be their expectation of the project beyond the cash

flows. Since there are many proposed projects, it is reasonable that the company can

benefit from future projects if the initial projects succeed. Mangers, however, need to

explain their reasoning by using an appropriate and consistent tool and to value their

expectations.

At this time, managers have difficulty in explaining the rationale for their investment

decisions although they intuitively think they should invest in all three projects. The

problem is the limitation of NPV analysis in valuing these projects under uncertainty. As

we discussed in this research, real options value (ROV) is an effective tool to make a

decision for these strategic projects.

4.2.2 Valuation Process

The first step of ROV is to recognize options in the project. According to the

classification of options 36 , wait (timing) options can be practically applied in station

development projects. Managers do not have to invest in all the uncertain projects at the

same time; or rather, they can invest in one of the projects, examine the results, and make

decisions on other investments later. The initial project has values as a pilot project,

which verifies the effectiveness of the general strategy for these kinds of projects. In this

case, project C is appropriate as the first project since it has the positive cash NPV

($5.OM without the salvage value), and requires the smallest amount of investment and

the shortest construction period. Project C creates options to wait and see the results of

the new strategy (station development). Managers can decide whether and when they

exercise options and invest in projects A and B based on the results of the project C.

Figure 4-10 illustrates the difference of the two processes, and options in the investment

strategies.

Original Project A, B, C Follow on investments

Strategy Initial Cost: $91.7 M L (27 projects)

Confirm the results of

three strategies (scale, area)

Proposed Project A Project B, C Follow on investment

process Initial Cost: $11.4 M Initial Cost $80.3M (27 projects)

S Confirm the results of

Confirm the results of three strategies (scale, area)
overall strategies
(station development)
Strategy
Figure 4-10 Implement
Figure 4-10 Implement Strategy

36 3.2 Application to Railway Projects

At this time, JR East has the details (e.g., construction and business plan) for projects A,

B, and C. However, there is uncertainty in these projects because they do not have

precedents, which help managers to forecast the cash flows. Therefore, managers can

invest in project C, examine the actual cash flows, and confirm the effectiveness of their

strategy. This investment will reveal the potential of station development projects, and

resolves the uncertainty related to the effectiveness of overall strategy. Managers can

adjust the NPVs of projects A and B based on the result of Project C.

Although the success of initial project (=C) does not necessarily indicate the success of

the next projects (=A and B), managers can improve the accuracy of their analysis, and

consequently make a better decision due to this additional information. If the project C

succeeds, the underlying asset of their options (= expected cash flows) will increase, and

projects A and B will more likely succeed. Managers can expect higher NPV and make

investment in projects A and B, which also resolve the uncertainty related to the scale and

area of projects.

Project-specific uncertainty remains to be unsolved even in this phase, and the

uncertainty may be high in a large and complex project. Managers, however, can focus

on the project-specific uncertainty to analyze the 27 projects, and their decision will be

more accurate because other uncertainty is already resolved.

 Success

Success

------- Expected Value of a

Failure
' Do not invest

Step 1 (Pilot Project =C) Step 2 (Projects in different area/scale = A&B) Step3 (Other Projects)

Source of uncertainty

Overall Strategy -' Different Area/ Scale E[ Project-specific condition I

Different Area/ Scale Project-specific condition
Project-specific condition

Figure 4-11 Uncertainty will decrease as the investment proceeds

The next step is to determine the parameters, which price the options. We have already

obtained several parameters from the project information:

S (asset price) = Present value of cash flows

K (exercise price) = Construction cost

Assuming the project period is 15 years and we use the Binominal Model, which we

explained in Chapter 3, one year is reasonable to avoid the complexity while keeping the

37 We can not make decisions on other projects at this time because we do not have project information (e.g.,
construction and business plan)
model precise. In this case, managers can make investment decisions every year based on

the performance of project C.

T (time interval) = 1 year

The interest rate of a Treasury bond is usually applied as a risk-free rate, and the average

rate of the Japanese Treasury bond (10 year) in the past five years (1998-2002) is 1.5 %38

R (risk-free rate of interest) = 1.5%

We use the movement of sales per unit area in the commercial industry so as to measure

the volatility of the underlying assets in the projects. The underlying asset is the cash

flows, which come from given areas in these projects. Since these projects have physical

constraints and cannot increase the business area, sales per unit area determine the cash

flows in projects. Actually, the margin rates slightly vary according to the sales, but the

data neither for the net income per unit area nor the margin rates are available at this time.

Therefore, we pragmatically use the fluctuation of the sales per unit area in the

commercial industry to measure the volatility in this ROV analysis. Figure 4-12 shows

the flow of this process.

38 http://www.mof.go.jp/jgb.htm
 (Volatility)

iber (Project Information)

,--- Underlying Asset

I
I

SNetCashFlow

Figure 4-12 Volatility of the underlying asset

The sales per unit reflect the attractiveness of the new business through the purchasing

power of customers (passengers). Since the business area is limited by the project plans,

the unit sales determine the total sales of the business, and consequently determine the

cash flows (underlying asset) from the project by using the given parameters.

hc]x Sa~bs peruh•t i the camme:ailhduswty

350 08
0.7
300
06
250 05

200 04

03
150
02
100

50 0

-01
0
1983 1986 1989 1992 1995 1998 2001

Figure 4-13 Volatility in the sales per unit area

 Figure 4-13 indicates the historical data of the sales per unit area in the commercial

industries in Japan 39 (index =100 in 1983). The average rate of the annual fluctuations

determines the possible range of the movement of an estimate, and consequently

measures the volatility of the projects. Although the drastic increase of sales in 1987 (due

to the economic boom) seems to be exceptional, similar situations may occur during the

long periods (15-20 years) of the projects. Therefore, we consider the past 20 years and

include the data from 1987 to determine the volatility in this analysis.

a (volatility) = 14.4%

Finally, we can value the projects based on the options to wait by using these parameters

in spreadsheets (Binominal Models).

4.2.3 Options Value

The results of ROV analysis are shown in the table 4-4:

Table 4-4 Results of ROV analysis

The value of Project ROV (w/ wait option) NPV (w/o options) ROV-NPV
Project A $4.2M -$0.7M $4.9M
Project B $4.9M -$0.1M $5.OM

39 http://www.mizuhocbk.co.jp/pdf/industry/1001_18.pdf
Project C provides options to examine the effectiveness of station development strategy,

and make investment decisions in projects A and B. The total value of the projects A and

B increases by $9.9M with the wait options, which are actually almost twice the NPV of

project C. We can recognize that project C is more significant as a pilot project than for

its cash flows. In this case, project C has positive NPV, so managers can make

investment even if they do not recognize the value of options. However, the ROV

indicates that managers can invest in project C even if the NPV is slightly negative (NPV

+ options value > 0) when they focus on long-term returns from their projects. It depends

on the investment policy whether managers invest in a project, which has negative NPV

and high options value. However, managers can consider the results of real options

analysis to match their decision-makings with the investment strategy.

Here is the structure of the spreadsheets in the case of project A. Based on the given

parameters and the Binominal formulas (3.3.2 Binominal Model), we calculate the

following numbers, which we use in this analysis.

Table 4-5 Parameters

Asset Value (Initial) S 38.900 Up Factor u 1.155

Strike Price K 39.600 Down Factor d 0.866
Risk-free Rate R 1.5%
Volatility (year) a 14.4% Up Probability p 0.516
Time interval (year) At 1.000 Down Prob. 1-p 0.484

Discount Factorl DF 0.985

 Note: S, K, r, a , and A t = given numbers in 4.2.2 (valuation process)

u = e-exp( a f- At)) = e.exp(0.144f 1) = 1.155

d = e-exp(- a - A t)) = e-exp(-0.144f 1) = 0.866
p = (1+r-d)/(u-d) = (1+0.015-0.866)/(1.155-0.866) = 0.516
DF = e-exp(-r- At) = e-exp (-0.015 -1) = 0.985

Using these numbers, we can build the spreadsheets as in Table 4-6. The asset value

varies based on the volatility (via up and down factors) in each period. The exercise value

is the immediate payoff from project A. Managers can invest in project A, and capture

the exercise value whenever the situation is favorable to the project. The holding value is

the expected value of the project in the next period. Managers have options to invest even

in the next period, and can compare the values of exercise and holding, and make a

decision as to when they exercise the option to invest in the projects.

Option values (= project value) will be determined by this decision. If managers invest,

the option value is equal to the exercise value in the period, while the option value is the

holding value in the case of not investing. We can calculate the option value by

expanding the same process to the final periods where the holding value is zero due to the

expiration. Figure 4-14 is a flow chart, which explains the logic in this model.
 Table 4-6 Binominal Model of Project A (period 0-1)

Period 0 1

Asset Value 38.900 44.925

Exercise V. (0.700) 5.325
Holding V. 4.228 8.232
Option V. 4.228 5.325
Exercise? NO YES 4 °

Asset Value 33.683

Exercise V. (5.917)
Holding V. 3.192
Option V. 3.192
Exercise? NO

(unit: million dollars)

Figure 4-14 Flow chart for the decision process in exercising options

We next have to define the timing of exercising the investment options. Theoretically,

managers have to exercise their investment options only when the exercise value exceeds

the holding value. If the payoff is sufficiently large, however, they are tempted to

exercise the options and capture the benefit. In this analysis, if managers wait for the

40 The investment option is exercised because the holding value is less than the exercise value plus annual cash flow
($8.23<$5.3M+$3.69M). Please read the following page for details.

100
exercise value to exceed the holding value, they cannot invest until the last period 41.

However, managers have to generate cash flows from these projects, and achieve their

medium-term business plan by the year of 2006. Therefore, managers can decide to

invest when they expect sufficient cash flows to meet their objectives. In projects A and

B, we assume that managers exercise their investment options when the future

expectation (= Holding Value - Exercise Value) is less than the annual cash flow in a

certain period.

Finally, we obtain the whole spreadsheet through repeating the calculation and decision

processes in each period as shown in Table 4-7 and 4-8.

Value

(S)
of Asset
Value

ý- -1
= PV of cash flow
Holding Exercising

Figure 4-15 Exercise Timing

41Please compare the values in Table 4-6 and 7

101
 Table 4-7 Option Pricing for Project A

Period 0 I I 1 I I 2 I I 4 1 1 5 6 7
Asset Value 38.900 44.925 51.883 59.919 69.200 79.917 92.295 106.590
Exercise V. (0.700) 5.325 12.283 20.319 29.600 40.317 52.695 66.990
Holding V. 4.228 8.232 12.867 20.902 30.181 40.898 53.275 67.568
Option V. 4.228 5.325 12.283 20.319 29.600 40.317 52.695 66.990
Exercise? NO YES YES YES YES YES YES YES
Asset Value 33.683 38.900 44.925 51.883 59.919 69.200 79.917
Exercise V. 5.917) (0.700) 5.325 12.283 20.319 29.600 40.317
Holding V. 3.192 4.171 8.198 12.867 20.902 30.181 40.898
Option V. 3.192 4.171 5.325 12.283 20.319 29.600 40.317
Exercise? NO NO YES YES YES YES YES
Asset Value 29.166 33.683 38.900 44.925 51.883 59.919
Exercise V. (10.434) (5.917) (0.700) 5.325 12.283 20.319
Holding V. 2.247 3.072 4.099 8.153 12.867 20.902
Option V. 2.247 3.072 4.099 5.325 12.283 20.319
Exercise? NO NO NO YES YES YES
Asset Value 25.254 29.166 33.683 38.900 44.925
Exercise V. (14.346 (10.434) (5.917) (0.700) 5.325
Holding V. 1.438 2.073 2.920 4.004 8.090
Option V. 1.438 2.073 2.920 4.004 5.325
Exercise? NO NO NO NO YES
Asset Value 21.867 25.254 29.166 33.683
Exercise V. (17.733) (14.346) (10.434) (5.917)
Holding V. 0.805 1.235 1.856 2.721
Option V. 0.805 1.235 1.856 2.721
Exercise? NO NO NO NO
Asset Value 18.935 21.867 25.254
Exercise V. (20.665) (17.733) (14.346)
Holding V. 0.371 0.612 0.992
Option V. 0.371 0.612 0.992
Exercise? NO NO NO
Asset Value 16.395 18.935
Exercise V. 23.205 (20.665)
Holding V. 0.126 0.225
Option V. 0.126 0.225
Exercise? NO NO
Asset Value 14.196
Exercise V. (25.404)
Holding V. 0.024
Option V. 0.024
Exercise? NO

102
 8 9 10 11 12 13 14 15
123.100 142.166 164.185 189.615 218.983 252.900 292.070 337.307
83.500 102.566 124.585 150.015 179.383 213.300 252.470 297.707
84.076 103.140 125.156 150.583 179.948 213.861 253.027 0.000
83.500 102.566 124.585 150.015 179.383 213.300 252.470 297.707
YES YES YES YES YES YES YES YES
92.295 106.590 123.100 142.166 164.185 189.615 218.983 252.900
52.695 66.990 83.500 102.566 124.585 150.015 179.383 213.300
53.275 67.568 84.076 103.140 125.156 150.583 179.948 0.000
52.695 66.990 83.500 102.566 124.585 150.015 179.383 213.300
YES YES YES YES YES YES YES YES
69.200 79.917 92.295 106.590 123.100 142.166 164.185 189.615
29.600 40.317 52.695 66.990 83.500 102.566 124.585 150.015
30.181 40.898 53.275 67.568 84.076 103.140 125.156 0.000
29.600 40.317 52.695 66.990 83.500 102.566 124.585 150.015
YES YES YES YES YES YES YES YES
51.883 59.919 69.200 79.917 92.295 106.590 123.100 142.166
12.283 20.319 29.600 40.317 52.695 66.990 83.500 102.566
12.867 20.902 30.181 40.898 53.275 67.568 84.076 0.000
12.283 20.319 29.600 40.317 52.695 66.990 83.500 102.566
YES YES YES YES YES YES YES YES
38.900 44.925 51.883 59.919 69.200 79.917 92.295 106.590
(0.700) 5.325 12.283 20.319 29.600 40.317 52.695 66.990
3.873 7.998 12.867 20.902 30.181 40.898 53.275 0.000
3.873 5.325 12.283 20.319 29.600 40.317 52.695 66.990
NO YES YES YES YES YES YES YES
29.166 33.683 38.900 44.925 51.883 59.919 69.200 79.917
(10.434) (5.917) (0.700) 5.325 12.283 20.319 29.600 40.317
1.577 2.447 3.681 7.847 12.867 20.902 30.181 0.000
1.577 2.447 3.681 5.325 12.283 20.319 29.600 40.317
NO NO NO YES YES YES YES YES
21.867 25.254 29.166 33.683 38.900 44.925 51.883 59.919
(17.733) (14.346) (10.434) (5.917) (0.700) 5.325 12.283 20.319
0.400 0.700 1.208 2.043 3.363 7.534 12.867 0.000
0.400 0.700 1.208 2.043 3.363 5.325 12.283 20.319
NO NO NO NO NO YES YES YES
16.395 18.935 21.867 25.254 29.166 33.683 38.900 44.925
(23.205) (20.665) (17.733) (14.346) (10.434) (5.917) (0.700) 5.325
0.047 0.092 0.181 0.355 0.699 1.376 2.707 0.000
0.047 0.092 0.181 0.355 0.699 1.376 2.707 5.325
NO NO NO NO NO NO NO YES
12.293 14.196 16.395 18.935 21.867 25.254 29.166 33.683
(27.307) (25.404) (23.205) (20.665) (17. 733 (14.346) (10.434) (5.917)
0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000
0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000
NO NO NO NO NO NO NO NO
10.644 12.293 14.196 16.395 18.935 21.867 25.254
(28.956) (27.307) (25.404) (23.205) (20.665) (17.733) (14.346
0.000 0.000 0.000 0.000 0.000 0.000 0.000
0.000 0.000 0.000 0.000 0.000 0.000 0.000
NO NO NO NO NO NO NO
9.216 10.644 12.293 14.196 16.395 18.935
(30.384) (28.956) (27.307) (25.404) (23.205) 20.665
0.000 0.000 0.000 0.000 0.000 0.000
0.000 0.000 0.000 0.000 0.000 0.000
NO NO NO NO NO NO
7.980 9.216 10.644 12.293 14.196
(31.620) (30.384) (28.956) (27.307) (25.404
0.000 0.000 0.000 0.000 0.000
0.000 0.000 0.000 0.000 0.000
NO NO NO NO NO
6.910 7.980 9.216 10.644
(32.690) (31.620) (30.384 (28.956
0.000 0.000 0.000 0.000
0.000 0.000 0.000 0.000
NO NO NO NO
5.983 6.910 7.980
(33.617) (32.690) (31.620)
0.000 0.000 0.000
0.000 0.000 0.000

103
 Table 4-8 Option Pricing for Project B

I Period I I 1 I 2 1 1 3 1 1 4 1 1 5 6 7
Asset Value 40.610 46.900 54.164 62.553 72.241 83.431 96.353 111.276
Exercise V. (0.090) 6.200 13.464 21.853 31.541 42.731 55.653 70.576
Holding V. 4.923 9.159 14.064 22.452 32.139 43.327 56.248 71.170
Option V. 4.923 6.200 13.464 21.853 31.541 42.731 55.653 70.576
Exercise? NO YES YES YES YES YES YES YES
Asset Value 35.164 40.610 46.900 54.164 62.553 72.241 83.431
Exercise V. (5.536) (0.090) 6.200 13.464 21.853 31.541 42.731
Holding V. 3.716 4.857 9.119 14.064 22.452 32.139 43.327
Option V. 3.716 4.857 6.200 13.464 21.853 31.541 42.731
Exercise? NO NO YES YES YES YES YES
Asset Value 30.448 35.164 40.610 46.900 54.164 62.553
Exercise V. (10.252) (5.536) (0.090) 6.200 13.464 21.853
Holding V. 2.616 3.577 4.772 9.066 14.064 22.452
Option V. 2.616 3.577 4.772 6.200 13.464 21.853
Exercise? NO NO NO YES YES YES
Asset Value 26.364 30.448 35.164 40.610 46.900
Exercise V. (14.336) (10.252) (5.536) (0.090) 6.200
Holding V. 1.674 2.414 3.400 4.662 8.994
Option V. 1.674 2.414 3.400 4.662 6.200
Exercise? NO NO NO NO YES
Asset Value 22.829 26.364 30.448 35.164
Exercise V. (17.871) (14.336) (10.252) (5.536)
Hdding V. 0.937 1.438 2.161 3.168
Option V. 0.937 1.438 2.161 3.168
Exercise? NO NO NO NO
Asset Value 19.767 22.829 26.364
Exercise V. (20.933) (17.871) (14.336)
Holding V. 0.432 0.712 1.155
Option V. 0.432 0.712 1.155
Exercise? NO NO NO
Asset Value 17.116 19.767
Exercise V. (23.584) (20.933)
Holding V. 0.147 0.262
Option V. 0.147 0.262
Exercise? NO NO
Asset Value 14.821
Exercise V. (25.879)
Holding V. 0.028
Option V. 0.028
Exercise? NO

104
 8 9 10 11 1 12 13 14 15
128.511 148.415 171.402 197.950 228.609 264.017 304.909 352.135
87.811 107.715 130.702 157.250 187.909 223.317 264.209 311.435
88.403 108.305 131.289 157.834 188.490 223.894 264.781 0.000
87.811 107.715 130.702 157.250 187.909 223.317 264.209 311.435
YES YES YES YES YES YES YES YES
96.353 111.276 128.511 148.415 171.402 197.950 228.609 264.017
55.653 70.576 87.811 107.715 130.702 157.250 187.909 223.317
56.248 71.170 88.403 108.305 131.289 157.834 188.490 0.000
55.653 70.576 87.811 107.715 130.702 157.250 187.909 223.317
YES YES YES YES YES YES YES YES
72.241 83.431 96.353 111.276 128.511 148.415 171.402 197.950
31.541 42.731 55.653 70.576 87.811 107.715 130.702 157.250
32.139 43.327 56.248 71.170 88.403 108.305 131.289 0.000
31.541 42.731 55.653 70.576 87.811 107.715 130.702 157.250
YES YES YES YES YES YES YES YES
54.164 62.553 72.241 83.431 96.353 111.276 128.511 148.415
13.464 21.853 31.541 42.731 55.653 70.576 87.811 107.715
14.064 22.452 32.139 43.327 56.248 71.170 88.403 0.000
13.464 21.853 31.541 42.731 55.653 70.576 87.811 107.715
YES YES YES YES YES YES YES YES
40.610 46.900 54.164 62.553 72.241 83.431 96.353 111.276
(0.090) 6.200 13.464 21.853 31.541 42.731 55.653 70.576
4.510 8.887 14.064 22.452 32.139 43.327 56.248 0.000
4.510 6.200 13.464 21.853 31.541 42.731 55.653 70.576
NO YES YES YES YES YES YES YES
30.448 35.164 40.610 46.900 54.164 62.553 72.241 83.431
10.252 (5.536 (0.090) 6.200 13.464 21.853 31.541 42.731
1.837 2.849 4.285 8.710 14.064 22.452 32.139 0.000
1.837 2.849 4.285 6.200 13.464 21.853 31.541 42.731
NO NO NO YES YES YES YES YES
22.829 26.364 30.448 35.164 40.610 46.900 54.164 62.553
(17.871) (14.336) (10.252) (5.536) (0.090) 6.200 13.464 21.853
0.465 0.815 1.406 2.378 3.915 8.346 14.064 0.000
0.465 0.815 1.406 2.378 3.915 6.200 13.464 21.853
NO NO NO NO NO YES YES YES
17.116 19.767 22.829 26.364 30.448 35.164 40.610 46.900
(23.584) (20.933) (17.871) (14.336) (10.252) (5.536) (0.090) 6.200
0.054 0.107 0.210 0.414 0.814 1.602 3.151 0.000
0.054 0.107 0.210 0.414 0.814 1.602 3.151 6.200
NO NO NO NO NO NO NO YES
12.833 14.821 17.116 19.767 22.829 26.364 30.448 35.164
(27.867) (25.879) (23.584) (20.933) (17.871) 14.336 (10.252) (5.536)
0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000
0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000
NO NO NO NO NO NO NO NO
11.112 12.833 14.821 17.116 19.767 22.829 26.364
(29.588) (27.867) (25.879) (23.584) (20.933) (17.871) (14.336)
0.000 0.000 0.000 0.000 0.000 0.000 0.000
0.000 0.000 0.000 0.000 0.000 0.000 0.000
NO NO NO NO NO NO NO
9.622 11.112 12.833 14.821 17.116 19.767
(31.078) (29.588) (27.867) (25.879) (23.584) (20.933)
0.000 0.000 0.000 0.000 0.000 0.000
0.000 0.000 0.000 0.000 0.000 0.000
NO NO NO NO NO NO
8.331 9.622 11.112 12.833 14.821
32.369) (31.078) (29.588) (27.867) (25.879)
0.000 0.000 0.000 0.000 0.000
0.000 0.000 0.000 0.000 0.000
NO NO NO NO NO
7.214 8.331 9.622 11.112
(33.486) (32.369) (31.078) (29.588)
0.000 0.000 0.000 0.000
0.000 0.000 0.000 0.000
NO NO NO NO
6.246 7.214 8.331
(34.454) (33.486)1 1(32.369)
0.000 0.000 0.000
0.000 0.000 0.000

105
4.3 Recommendations

1. Recognize the value of wait options, and make phased investment

It is not a wise strategy to invest in the three projects at the same time; or rather,

managers should launch only project C as a pilot case. They can expect about $5.OM

NPV from the investment, and acquire options to wait and see the results of their strategy,

which are worth $9.9M in total. Managers can examine the reaction from customers in

project C, and make decisions on projects A and B based on it.

It is risky to invest in a large project under uncertainty. If the project turns out in failure,

the negative impact can be serious to JR East. Therefore, it is essential to make phased

investment, and limit the potential loss of this project while holding options to make the

rest of the investment. Thus, phased investment provides us with flexibility to cope with

uncertainty about the project. The value of this flexibility critically increases the whole

value of a large-scale investment under uncertainty. Figure 4-16 shows how the same

project has different value according to its implementation strategy. We can appreciate

the role of flexibility and importance of phased investment in a strategic project. The

ROV is an effective tool to reveal these values, which we could not recognize in the NPV

analysis.

106
 PzDct Va~le
15

10

-5

Figure 4-16 Value of the projects (A, B, C) by implementation strategies

It seems clear that we can benefit from the phased investment if the option is free.
their
However, waiting options actually cost because of the time constraints of

implementation strategy. As we described in the company information, a station

development project is one of the main strategies of JR East to achieve its medium-term
to
business plan by 2006. Therefore, it is essential for managers to move fast while trying

find the optimal investment timing, which maximizes the values of project. Managers

could not rationally solve the trade-off problem with the NPV analysis, but the ROV

clearly provides quantified benchmarks for this strategic decision based on the value of

wait options. If the additional value of the projects is worth waiting for, managers should

make phased investment; otherwise they can invest in all three projects at the same time.

107
 Move Project A,B,C
Fast Speed or Value9

Period: 3 years Target Year: 2006

Cash Value: $4.2M

Optimal
Optimal Project A Project B, C
Timing
(wait option)
Period: At least 4 years
Cash Value: $16.8M

Figure 4-17 Managers can choose either speed or cash values in an implement strategy

2. Recognize the value of options to quantify the expectations.

In the NPV analysis, managers assumed salvage values in project A and B so as to

provide positive NPVs and support their decisions to invest in the projects. They

modified the NPVs because this method could not fully explain the values of projects,

which experienced managers intuitively recognized.

The ROV provides a clear answer to this vague expectation. That is an "option" to

expand their project, and generate incremental cash flows in follow-on projects. Since the

company has an additional 27 potential projects in the target area, the cumulative benefit

from these projects is so significant that the initial projects are more important as

pioneers of this strategic investment than as cash flows.

108
 may choose to invest
Although we recommend making phased investments, the company
their business plan. In this
in all three projects at the same time, and move fast to achieve
which can be created by
case, managers have to recognize the value of expansion options,
to value the other 27
the three projects. At this time, we do not have enough information
to later invest in a
projects. If we assume, however, that each project creates an option
is close to the total
similar project in terms of its scale and area, the value of the options
should try to
amount of the salvage values in the NPV analysis (Figure 4-18). Managers
options, and
estimate the value of follow-on projects, consider the value of expansion
method
justify the rationale for their investment decisions not with an exceptional
to apply an
(salvage value) but with consistent analysis (ROV). Managers have

appropriate tool, which can effectively explain their thinking.

Figure 4-18 Additional Values of Projects

109
Although the ROV uses some assumptions to determine the value of investment options,

this approach reflects the strategic value of projects, which managers are now considering

under uncertainty. Therefore, we recommend them to use the ROV and strategically think

in capital investment in transportation systems based on the discussions and case study in

this research, which we summarize in next chapter.

110
 Chapter 5 Conclusion

In this thesis, we examined decision analysis methods of capital investment in

transportation systems. The business trend, which regards cash flows as an important

criterion of a corporate performance, forces managers to maximize the cash value of a

project to meet the shareholder's expectations. The characteristics of transportation

systems, however, require a comprehensive approach to optimize the performance over

time, and thus, managers need to apply an appropriate tool to measure the value of a

project to effectively develop the systems. Here are the conclusions of this research.

5.1 Conclusions

Conclusion 1: The strategic value of a project is critical for developing transportation

systems.

Focusing on the NPV of a project does not necessarily result in the optimal investment

for developing transportation systems. The fundamental nature of CLIOS 42 makes it

difficult to fully predict the performance and impacts of a project in transportation

systems. Therefore, managers need to recognize the strategic value of a project and create

options in addition to the NPV in their investment activities so as to flexibly adjust to

unexpected events in the future.

42 CLIOS : complex, large, integrated open systems.

from Sussman, Joseph "Introduction to Transportation Systems", 2000, pp. 6-7

111
Applying Porter's model and System Dynamics 43 to the railway business, we can

recognize the importance of capital investment, which contributes to long-term

profitability in the competitive market. The capital investments are, however, typically

located in the risky segments of the project portfolio in a railway company, so the

investment activities often create unexpected system results. Therefore, managers are

required to think strategically so that they can capture the present benefit while enhancing

opportunities and avoiding negative results under uncertainty.

Conclusion 2: Real options value (ROV) is an effective tool to measure the strategic

value of a project.

Combining the NPV with another method (e.g., decision tree) measures the cash value of

a project, and provides benchmarks in investment decisions when the future is fairly

stable or probabilistically predictable. However, these methods are insufficient for

valuing a project under uncertainty where the potential outcomes are not probabilistically

understood, and the estimated value of the strategic options becomes critical to making an

investment decision. The ROV clearly values the options, and provides the strategic value

of a project under uncertainty.

It is useful to be aware of the merits and limitations of each valuation method. In this

sense, the merit of ROV successfully overcomes the limitation of other methods that we

reviewed in this thesis. Previously, even if managers recognized the strategic value of a

43 1.2.2 Market Force & 1.2.3 Business Dynamics

112
project, they did not have an effective tool to quantify their intuitive judgment. Using the

realistic evaluation tool can guide managers toward exploiting the opportunities, and

creating competitive advantages in the market.

Conclusion 3: The ROV is applicable to transportation systems, and the Binominal

Model is a practical tool to measure the ROV in a project.

We verified the theoretical correctness of the ROV as applied to transportation systems

by proposing the complete market assumption where we can trade options. This complete

market enables us to use the no-arbitrage approach, which is an underlying concept of the

options pricing theory. We also demonstrated the practicality of ROV and the binominal

model through a case study of a railway project, and showed substantial improvements in

the investment strategy.

The ROV quantifies the strategic value of a project, and consequently improves the

investment strategy. We can increase the value of a project by recognizing the importance

of options, looking for opportunities to build them into a project, doing the valuation, and

making a "strategic" decision. This approach is critically effective when uncertainty is a

major concern, and sequential decisions play an important role in managing a project.

Thus, the ROV is quite useful to create a strategic decision in transportation systems.

113
5.2 Future research

We also suggest additional research to apply real options to transportation systems based

on the results of this research.

First, we have to consider the role of the "network" in a system, and expand the option-

based investment strategy to the portfolio management. This thesis focused on an

individual project in applying the ROV to transportation systems. However, the network

is a key concept when we study an investment strategy in transportation systems.

Individual projects do not work effectively by themselves; rather, they affect each other

and develop into integrated systems as a whole. One investment may have side effects

and have impacts on the value of another project in the transportation network, and we

need to examine the role of this factor as it affects strategic decision-makings.

t2
ne)

Project 1 impacts on the existing traffic flows

Figure 5-1 Network Factor

114
Based on the understanding of this factor, we need to investigate how we can value the

portfolio of investment options and how the value changes according to investment

decisions under uncertainty. The result of this work allows managers to make decisions

not only on the project level but also on the corporate level. Just as financial managers

build a portfolio of securities to manage risks and benefits of their investment, this

approach gives us substantial improvements in managing the corporate strategies in the

transportation field.

Next, we need to examine the past transportation projects and create a database and

application to use the ROV in practice. In this thesis, we applied the ROV in a railway

project. However, we simplified the analysis to avoid the complexity and to emphasize

the comparison between the ROV and conventional methods. For instance, volatility will

not be determined by a single factor; rather, we need to combine several factors. To make

the analysis more realistic and practical, we have to review past railway projects,

examine the historical data, and build models and applications, which managers can

practically use to make an investment decision.

As we discussed in this thesis, experienced managers intuitively use option thinking to

some extent in valuing a project. The ROV offers a way to quantify the intuitive

management with a coherent process that more closely approximates the real-world

behavior in a dynamic environment.

115
This thesis helps to improve the investment strategy in the transportation industry.

Options theory caused drastic changes in the financial market by providing numerous

ways for individuals and corporations to hedge risks and improved the performance of

their portfolios. Using real options has further potential to innovate in managing an

investment in the fast-moving world of technology and today's uncertain business climate.

Managers have to appreciate the benefit of this strategic approach to a project, and the

application of real options to transportation systems including railway systems leads real

benefit. "Real options are real."4 4

44Tom Copeland, 15.975. SSIM-Real Options-, April 7. 2003

116
 Bibliography

Abken, Peter A.& Madan Dilip B., "Pricing S&P 500 index options using a Hilbert space
basis", Federal Reserve Bank of Atlanta Working Paper,1996

American Road and Transportation Builders Association, "Public/Private Ventures in

Transportation", 1990

Amram, Martha & Kulatilaka, Nalin, "Real Otions", Harvard Business Press, 1999

Arnold, Tom, "Value Creation at Anheuser-Busch: A Real Options Example", Journal of

Applied Corporate Finance, Summer 2001

Bookstaber, Richard M., "Option Pricing and Investment Strategies 3Rd Edition", Probus
Publishing Co., 1991

Brealey, Richard A. & Myers, Stewart C., "Principle of Corporate Finance", Irwin
McGraw-Hill, 2000

Brealey, Richard A. & Myers, Stewart C., "Capital Investment and Valuation", Irwin
McGraw-Hill, 2003

Copeland, Tom, "Real Options: A Practitioner's Guide", Monitor Corporate Finance,

2000

Dixit, A.K. & Pindyck, R.S., "Investment under Uncertainty", Princeton University Press,
1994

Dixit, A.K. & Pindyck, R.S., "The Options Approach to Capital Investment", Harvard
Business Review, 1995

Hevetrt, Kathleen, "Real Options Primer: A Practical Synthesis of Concepts and

Valuation Approaches", Journal of Applied Corporate Finance, Summer 2001

Holloway, Charles A., "Decision Making under Uncertainty: Models and Choices",
Prentice-Hall Inc., 1987

JR East, "Annual Report 2001"

Karpark, Birsen & Zionts Stanley, "Multiple Criteria Decision Making and Risk
Analysis", Springer Verlag Berlin Heidelberg, 1987

Kulash, Damian & Clinger, Jenifer, "Economic Returns From Transportation

Investment", Eno Transportation Foundation, Inc., 1996

117
Kulatilaka, Nalin, "The value of Flexibility: The Case of a Dual-Fuel Industrial team
Boiler", Financial Management Association, 1993

Lessard, D. & Miller, R., "Understanding and Managing Risks in Large Engineering
Projects", MIT Sloan Working Paper 4214-01, 2001

Lessard, D. & Miller, R., "The Strategic Management of Large Engineering Projects",
MIT Sloan Working Paper, 2000

Lifson, Melvin W., "Decision and Risk Analysis", Cahners Books, 1986

Luenerger, David G., "A Correlation Pricing Formula", Stanford University Working
Paper

Luenerger, David G., "Arbitrage and universal pricing", Stanford University Working
Paper

Luenerger, David G., "Investment Science", Oxford University Press, 1998

Luenerger, David G., "Optimization by Vector Space Method", John Wiley & Sons, 1997

Luenerger, David G., "Projection Pricing", Stanford University Working Paper

Millar, John B., "Principles of Public and Private Infrastructure Delivery", Kluwer
Academic Publishers, 2000

Ministry of Transportation, Japan, "Cost/Benefit Analysis Manual for Railway Projects",

1999

Moel, Alberto, "Bidding for the Antamina Mine: Valuation and Incentives in a Real
Options Context", Harvard Business School Working Paper, 1998

Myers, Stewar C., "Valuing Investments in Research and Development", MIT Real
Options Seminar, 2002

Myers, Stewart C. & Howe, Christopher D., "A Life-Cycle Financial Model of
Pharmaceutical R&D", MIT Working Paper, 1997

Neufville, Richard de, "Dynamic Strategic Planning for Technology Policy", MIT-TPP
working paper, 2000

Neufville, Richard de, "Real Options: Dealing with Uncertainty in Systems Planning and
Design", MIT-TPP working paper, 2001

118
Ng, Francis & Bjornsson, Hans, "Evaluating a Real Option in Material Procurement",
Stanford University Working Paper, 2001

Nichols Nancy A. "Scientific Management at Merck", Harvard Business Review, 1994

Porter, Michael E., "Competitive Advantage", The Free Press, 1998

Siegel, Daniel R., "Valuing Offshore Oil Properties with Option Pricing Models",
Northwestern University Department of Finance, Working Paper, 1985

Sterman, John D., "Business Dynamics", Irwin McGraw-Hill, 2000

Suijs, Jeroen, "Cooperative Decision-Making Under Uncertainty", Kluwer Academic

Publishers, 2000

Sussman, Joseph, "Introduction to Transportation Systems", Artech House Inc., 2000

Trigerous, L. & Mason S. P., "Valuing Managerial Flexibility", Midland Corporate

Finance Journal -Spring-, 1987

U.S. Department of Transportation, Urban Mass Transportation Administration, "The

UMTA Rail Modernization Program", 1979

Wilson, Rosalyn A., "Transportation America", Eno Transportation Foundation, Inc.,

1994

Yu, Po-Lung, "Multiple-Criteria Decision Making", Plenum Press, 1985

119