Rethinking Infrastructure Project Finance

Rethinking Infrastructure Project Finance
Andrew H. Chen*
Southern Methodist University
e-mail: achen@mail.cox.smu.edu
Current version: November 17, 2004
Abstract
This paper describes some of the difficulties associated
with the current Build-Operate-Transfer (BOT) and Public-Private-
Partnership (PPP) approaches to infrastructure project financing,
and argues that efficient financing of infrastructure projects should
be guided by the invisible hand of the capital market. It further
argues that financial innovations must play a key role in such new
approaches.
________________
*I would appreciate any comments or suggestions on this preliminary draft.

2
Rethinking Infrastructure Project Finance
Andrew H. Chen
1. Introduction
The successful development of large-scale infrastructure projects such as power plants,
roadways, bridges, dams, seaports, airports, and telecommunications networks is vitally
important for economic growth in developed and developing countries alike. Consequently,
ensuring that such projects are successfully undertaken and completed should be a top economic
priority, particularly for the governments of developing countries. In the past decade, the scale
of global capital investment in infrastructure projects has increased rapidly. Esty (2004a) points
out that the total project-financed investments worldwide grew at a compound annual rate of
almost 20% through most of the 1990s and reached $165 billion in 2003. 1
Most large-scale
infrastructure projects, by very nature involving massive capital investment, long completion
times, and high levels of risks, 2 present formidable financial challenges to developing countries
of the world. Infrastructure development cannot succeed without adequate financing, and it is
clear that large-scale infrastructure development in developing economies requires substantial
amounts of technology and capital, primarily from the industrialized countries in North America
and Europe.
Although historically we have witnessed a number of large and successful infrastructure
projects, we have also seen many spectacular failures. For example, the total cost of the 31-mile
Eurotunnel linking the English Channel between the UK and France was estimated to be £4.9
billion in 1987, but the actual cost was £9.7 billion upon its completion in December 1994. Some
1 See Esty (2004a).
2 Some of the project risks are discussed in Fabozzi and Nevitt (2000).

3
problems in the operations phase of the project have made it difficult servicing its massive debt
which has been heavily restructured including the conversion of almost half of the US$12.5
billion debt into equity. 3
And the most recent example is Boston’s Central Artery/Tunnel (CAT)
project which has just recently been completed. The total cost of the newly opened highway
tunnel project was $14.6 billion which was about six times the original cost estimate of $2.6
billion during the design phase in 1985. As CBS News has reported, “The project – the most
expensive highway project in U.S. history – was five years late and billion of dollars over cost,
and has been plagued by allegations of fraud, waste and mismanagement.” 4
The breached wall
caused a flood in Boston in September 2004 and the tunnel is now riddled with hundreds of
smaller leaks that need to be fixed.
In their study of sixty engineering projects with an average size of $1 billion undertaken
around the world between 1980 and 2000, Miller and Lessard (2000) found that almost 40% of
them performed poorly and were either abandoned completely or restructured due to severe
financial crisis. In their more recent study of the performance of large transport infrastructure
projects, Flyvbjerg et al. (2003) found that cost overruns of 50% to 100% and revenue shortfalls
of 20% to 70% were not uncommon.
The common approaches to financing large-scale infrastructure projects around the world
are well-known. Projects are typically financed on a stand-alone basis, giving rise to an
independent project company for the purpose of financing the construction, operation, and
maintenance of a specific public investment project. The project company exists as a legally
independent entity, usually has a clearly-defined lifespan, and most often is financed with equity
capital from one or a small number of project sponsoring firms and with syndicated project loans
3 See “Case study: Eurotunnel – A disaster for lenders” in Fabozzi and Nevitt (2000).
4 See the 11/10/2004 article, “Has Big Dig Become Big Drip?” CBS News, The Associated Press, Boston.

4
from a consortium of foreign and domestic banks. The current practice of creating project
companies with the structural attributes of concentrated equity ownership and high leverage with
syndicated bank loans (see e.g., Esty and Megginson, 2003 and Esty, 2004b) has led to Esty’s
(2004a) recent conclusion that “Project finance, as it turns out, just happens to be a particularly
interesting and effective setting in which to illustrate the simple yet critical idea that ‘structure
matters’”. However, I believe that the structural characteristics of project companies in the
existing approaches to financing infrastructure projects are largely to blame for significant
agency problems and for some of the failures of large-scale investment projects around the world.
Existing approaches are based on inherently flawed economic justifications. Therefore, I
disagree with Esty’s (2004a) assertion that “…the growing use of project finance challenges the
Modigliani and Miller’s (1958) ‘irrelevance’ proposition, the idea that corporate financing
decisions do not affect firm value under certain conditions.” If the financing problems under the
current approaches that have repeatedly plagued infrastructure projects are to be solved, we must
consider new approaches.
This paper is organized as follows. Section 2 discusses some of the shortcomings of
existing approaches to infrastructure financing. Section 3 discusses the benefits of financing
infrastructure projects by issuing project securities through the global capital markets and
describes some new innovations in security designs for financing infrastructure projects around
the world. Finally, the paper concludes with a brief summary.
2. Existing Approaches to Infrastructure Financing and Their Shortcomings
Currently, there exist two main approaches to financing infrastructure projects. The first
approach is one that I will call the State-Build-Own-Operate (SBOO) approach. The second
approach is widely known as the Build-Operate-Transfer (BOT) or the Public-Private -

5
Partnership (PPP) approach. Although BOT (or PPP) is unique to the infrastructure project and
to the locale, the basic approach involves awarding a project to a project company that operates
the project for twenty-five to thirty-five years before it is transferred back to the government.
The sole difference between BOT and PPP lies in the two approaches’ output criteria. In the PPP
model, unlike with BOT, the government buys services from project companies at agreed-upon
prices and thus has a greater influence in production decisions and bears a greater share of
project risks; while the project companies are not overly exposed to changing market conditions.
2.1 The SBOO Approach
In many countries, especially those with planned economies, major state projects of
infrastructures are built, owned, and operated by the government. The procedure for a particular
public investment project usually begins with government estimates of the total cost of
undertaking the project, including costs of project construction as well as project operations and
maintenance. The government then evaluates, compares, and chooses from available financing
alternatives. The set of financing alternatives for public projects usually includes public funds
from government budgets, domestic and international bank loans, and privately-placed project
bonds. Not surprisingly, jurisdiction over the aforementioned steps falls under the influence of
domestic politicians. There are several problems inherent in this basic approach. First, bidding
on infrastructure projects is often subject to bribery and corruption. Second, deciding upon a
complex financing structure requires much negotiation and re-negotiation that can cause delays
or even project halts. Third, approval of funding for state-owned projects must often navigate
through lengthy and bureaucratic political processes. And fourth, upon completion of project
construction, actual operation and maintenance are often inefficient by virtue of the state's
ownership of the project. Indeed, we can draw lessons from the numerous privatizations or

6
securitizations of the state-owned infrastructure projects around the world over the past few
decades. We know that such privatizations or securitizations occurred largely because stateowned
enterprises (SOEs) suffered from inefficiency and poor performance stemming from a
lack of proper incentives and reward systems for managers and workers. A recent study by La
Porta and Lopez-de-Silanes (1999) has shown the evidence of benefits of privatization from
Mexico.
It is worth pointing out that the privatizations or securitizations of even some wellfunctioning
publicly-owned infrastructure projects can emerge for other reasons, such as the
government’s budgetary needs. As Baglole (2003) has recently stated concerning privatization
and securitization in Hong Kong: “To help reduce a record budget deficit, the government plans
to sell partial stakes or securitize revenues from major assets over the next 18 months, from its
airport to its tunnels and even its housing department. The program could raise anywhere from
$14 billion to $40 billion in coming years, depending on what goes on the block, and would put
Hong Kong at the forefront of privatization efforts globally.” 5
Experience with such privatizations or securitizations suggests that creating immediate
private ownership of a public investment project among a diverse group of investors, both
domestic and from abroad, may lead to a more efficient and more successful type of
infrastructure financing. In later sections, I will illustrate how this more rapid and efficient
financing of infrastructure projects may be accomplished by issuing project securities with new
financial innovations through the capital markets.
2.2 The BOT or the PPP Approach
5 See Joel Baglole, “Hong Kong Plans Another Handover,” Wall Street Journal, 10/17/2003.

7
In order to tap investment capital from private sectors while improving the efficiency of
public investment projects, governments in many countries have been widely urged by the
international organizations such as World Bank and IMF to apply BOT or PPP 6 approaches to
encourage the participation of private sectors in infrastructure developments. The original
purposes of both BOT and PPP models are threefold: 1) to attract capital funds from private
sectors to assist public investment development; 2) to increase the efficiency of public
investment projects; and 3) to share project risks by government and private sectors. However,
as noted earlier, a lion share of project risks is borne by governments in a PPP model of
infrastructure project financing. Therefore, there has been a new trend of preferring BOT model
over that of PPP in many countries. For example, on October 27, 2004, the Planning
Commission of the Government of India announced its preference for the toll-based BOT to the
annuity-based PPP for the development of major highways in the country and asked the
Government to review the incentives granted to PPPs. 7
Nevertheless, the use of the BOT and PPP models for infrastructure project financing is
widespread in many countries. As of May 2004, there were 240 BOT projects in East Asia and
240 in South America. 8 As for the PPP approach, it is reported that since the early 1990s, the
UK has signed over 569 contracts with an overall value of 52.6 billion pounds having been spent
on PPP projects. 9 Compared to the SBOO approach, the newer BOT and PPP approaches can
undoubtedly improve the efficiency of infrastructure financing. However, as I will argue below,
6 See Bradford (2001). The PPP approach outgrew from the UK government’s Private Finance Initiative (PFI)
program launched in the early 1990s that aimed at bringing private equipment, services, and funds into the field of
public utilities. The PFI structures have grown and now represent more than 15% of the UK government’s total
equipment expenditures.
7 Source: Business Line, 10/28/04.
8 Source: Financial Times, 5/12/2004.
9 Source: China Post, 10/3/2003.

8
both the BOT and the PPP approaches still suffer from some serious shortcomings that can
hinder the performances of infrastructure projects around the world.
2.2.1 Inefficient Bidding Processes
In BOT and PPP project bidding, domestic and foreign project companies usually submit
bids for the entire project to government agencies, which make the final selection. Therefore, the
process is not immune from the influence of domestic politicians and concomitant bribery and
corruption. For example, Mr. Estrada, the former President of the Philippines, stated that it was
his administration’s policy not to grant BOT projects any government guarantee. He confirmed
that Mr. Jimenez, a former businessman, offered him $14 million for a direct government
guarantee on the multibillion dollar Caliraya-Botocan-Kalayaan (CBK) hydropower project—an
offer which Mr. Estrada had declined. Mr. Estrada further stated that Mr. Ramos, the former
President of the Philippines, and Mr. Abrille, the former Philippine ambassador to Argentina,
had lobbied for IMPSA (Industrias Metalurgicas Pescarmona Sociedad Anonima) for the CBK
project, and that IMPSA contract was not formally consummated in his two and a half years in
office. Nevertheless, the new administration which followed Mr. Estrada’s took only a few days
to conclude the CBK deal. 10 It should be pointed out that the BOT and PPP laws in some
countries explicitly prohibit government guarantees of infrastructure projects. An important
question is whether project companies would be willing to bear all the project risks without any
implicit or explicit government guarantees under either the BOT or PPP models of project
financing. On November 4, 2004, it was reported that Siemens (the German train manufacturer)
had threatened to withdraw from the tender of the $1.8 billion Tel Aviv light railway BOT
project unless the Israeli government increased its guarantee of the project from around 75% to
10 Source: BusinessWorld (Philippines) 1/15/2003.

9
100%. 11 The result of Siemens’ action could jeopardize the important infrastructure project in
Israel.
While it is true that using public tenders in BOT and PPP project bidding processes can
increase efficiency in project cost reduction, the governments of some countries often impose
restrictions on bidding that work against their own best intentions. Instead of allowing open
tenders for projects, some governments, based upon their criteria, negotiate and directly assign
project contracts to closely-tied investment companies. According to the World Bank’s mid-
1999 reports on the impact of the Southeast Asian financial crisis, a number of governments and
enterprises in the region had to incur debts of billions of U.S. dollars due to bad negotiations in
BOT investment projects. For example, Indonesia and the Philippines incurred some US $10
billion and US $6 billion in bad debt, respectively, from BOT investments. 12
The negotiations of large-scale infrastructure projects in BOT or PPP bidding processes
usually take a huge amount of efforts and time to complete. As Pope (2004) points out, in
February 2004 when the $3.6 billion financing for the Baku-Tbilisi-Ceyhan (BTC) pipeline
project which had been negotiated for more than ten years was closed it required 208 finance
documents and 17,000 signatures from 78 parties. 13 This does not seem to be an efficient way to
finance a large-scale investment project, even if the project could be proven successful in the
next several decades to come.
It is worthwhile to note that some governments now employ new measures to promote
and increase capital investment via the BOT format. For example, it was reported that the
Taiwan Cabinet’s Public Construction Commission finalized a plan of allocating yearly quotas
11 Source: Israel’s Business Arena, 11/4/2004.
12 Source: Financial Times, 3/6/2003
13 Source: Wall Street Journal,

10
for the number of BOT projects each government agency should sign up with private investors,
and offering monetary awards to government employees with superb performance in winning
BOT contracts. 14 Again, the presence of political influence on infrastructure developments under
the BOT format is quite clear.
2.2.2 Imperfect Project Contracts
Asian countries in particular have frequently invited firms and contractors from abroad to
invest in their infrastructure projects under the BOT and PPP approaches. Because these foreign
firms and contractors often possess advanced technologies and vast amounts of capital, domestic
politicians may seek to provide enticements via the “red carpet,” i.e., financial incentives
allowing the creation of tax-free bonds for infrastructure; or multi-year tax holidays for
infrastructure investment. However, a red carpet, if not carefully maintained, can be easily
transformed into so much red tape. For example, in the early 1990s the Hong-Kong based
Hopewell Holdings secured an ambitious road and train system BOT project in Bangkok for a
30-year operation with concession of taxes and other benefits. The total worth of this road and
rail project was approximately $4 billion. However, in January 1998 the Thai government, citing
major delays and cost overruns, ordered the disengagement of Hopewell from Bangkok’s BOT
project. In the end, the Thai government was only willing to reimburse Hopewell a fraction of
the $600 million the infrastructure company had spent on the project. 15
Another instance of an unsuccessful BOT infrastructure project involved the U.S. firm
Congentrix Energy Inc.. Congentrix had been invited by the Indian government to join with
Hong Kong’s China Light and Power Ltd. as a partner in the project. In 1992, the two
companies were to invest in and develop a $1.6 billion Mangalore power project in the South
14 Source: Taiwan Economic News, 2/20/2003
15 Source: Financial Times, 11/4/98.

11
India state of Karnataka. The project was designated as a “fast-track” project, which entitled it to
federal financial guarantees. Mangalore Power Co. (MPC), the company set up by Congentrix
and China Light and Power to develop the project, waited for over 7 years to get clearance for
executing the project. During those years, MPC spent more than $20 million and fought several
public interest litigations for two years, eventually prevailing. However, there was a pending
two-year-old suit alleging the payment of $13.8 million in bribes to Indian officials. After a
lower court ordered an investigation into the allegations of corruption, MPC appealed to the
Supreme Court in April 1998, and the Justices reserved the case for judgment in January 1999.
MPC waited for almost a year without a verdict; as a result, Cogentrix quit the project on
December 9, 1999. 16
On December 13, 1999, four days after Cogentrix announced its
withdrawal from the power project, the Supreme Court of India gave a judgment in favor of
MPC. 17 These examples illustrate that sometimes a very lengthy amount of time is required to
conduct negotiations and renegotiations among project companies consisting of foreign and
domestic project sponsors, foreign and domestic lending banks, and politicians of the host
country. The project sponsors engaged in an infrastructure project with concentrated equity
shareholdings will typically bear almost all of the risks—both financial and political—yet will
lack adequate diversification. Therefore, countries in which the political processes of project
approval are mired in red tape will have difficulty in attracting foreign project companies for
infrastructure projects.
The PPP approach to project finance is also often subject to political risks. For example,
Enron Corporation built a $2.9 billion power plant in Maharashtra, India that resulted in
16 Source: Wall Street Journal, 12/10/99.
17 Source: Financial Times, 12/26/99.

12
considerable controversy in 2001. Enron’s plan was the first of the "fast-track" PPP power
projects signed in 1995 and the only one to have been successfully implemented in India. The
Dabhol Power Company (DPC), in which Enron held a 65 percent share, and its sole customer,
the Maharashtra State Electricity Board (MSEB), had been in dispute on the issue of tariffs.
MSEB owed DPC $45 million in unpaid bills; and DPC insisted the tariff—three times higher
than that typically levied by Indian power producers—reflected high capital costs, rising fuel
costs, and the depreciation of the rupee since the contract was signed in 1995. DPC had issued a
preliminary notice to cancel its contract with MSEB. As a result, in June 2001 the Indian and
foreign banks decided to force the closure of the Enron power plan for a year to give the two
partners in the project a chance to discuss and reconcile their differences. 18
The above examples underscore the fact that any project company in BOT or PPP
financing which engages in direct investment on infrastructures invariably must deal with
bureaucratic red tape and political risk because, ultimately, the key decisions rest in the hands of
host-country politicians.
In addition, changes in political administration can have an ex-post impact on the
viability of agreed-upon BOT or PPP contracts. For example, at the urging of the World Bank
and the International Monetary Fund, in 2000 the government of Turkey adopted the BOT
approach to its electricity sector. However, in 2003 the new Turkish government licensed 72
new power plants with outputs ranging up to 1,120MW, and at the same time it moved to cancel
30 BOT projects with a combined capacity of 2,846MW due to high output costs. 19
It is also important to point out that the enforceability of contracts agreed upon by the
project companies and the governments under BOT or PPP laws is often subject to the judicial
18 Source: Financial Times, 6/8/2001.
19 Source: Platts Energy Economist, March 2003.

13
interpretation of the courts. For example, on January 21, 2004, the Supreme Court of the
Philippines issued a resolution affirming its May 5, 2003 decision declaring as null and void
Piatco’s (Philippine International Air Terminals Co.) BOT contracts in the NAIA-3 (Terminal-3
at the Ninoy Aquino International Airport) project, which were valued at $650 million. 20 The
ruling had a significant impact on various parties involved. Pistco’s German partner, which has
written off some $313 million in its investment in the Terminal-3 project, vowed to pursue its
claims on the facility. 21
It has also been argued that the propriety of security arrangements for
the investments in the project, particularly those provided by international financing institutions
(e.g., the Asian Development Bank and World Bank’s International Financing Corp., which had
approved a $440-million loan package for the project) should not be left to the court’s decision. 22
2.2.3 Lack of Diversification and Liquidity in Project Finance
The number of participants in project finance under the BOT or PPP approaches is
usually very small, giving rise to substantial illiquidity and risk concentration. For instance,
some governments usually allow private firms to apply either singly or through the formation of
a consortium that is restricted to four or five project sponsoring firms for tenders. The number of
leading banks in the syndicated project lending consortium is also usually quite small. 23
Therefore, the BOT and PPP approaches do not provide the benefits of risk-sharing of project
risks for equity-holders and creditors of the project companies. For instance, the Taiwan
government launched a BOT project for a high-speed rail link between Taipei and the CKS
20 Source: BusinessWorld (Philippines), 2/2/2004.
21 Source: Manila Standard, 5/7/2003.
22 Source: BusinessWorld (Philippines), 2/2/2004.
23 As Esty (2004b) points out “Lending syndicates resemble pyramids with a few arranging banks (arrangers) at the
top and many providing or participating banks (providers) at the bottom. Prior to closing a loan, the arranging (or
mandated) banks meet with the borrower, assess the credit quality, negotiate key terms and conditions, and prepare
an information memorandum for providing banks.” He provides evidence showing the importance of legal and
financial systems on the composition and the pricing of syndicated project loans.

14
International Airport in 1997. But the two contractors, Evertransit International Co. and BES
Engineering Corp., both abandoned the project citing financial problems and technical
difficulties. As a result, the government scrapped the project in April 2003 after six years of
planning. 24
In addition to problems of illiquidity under the BOT and the PPP approaches, any
contract that permits private firms to operate projects for twenty-five to thirty-five years and then
transfer them to the government creates severe agency and management incentive problems. As
one can easily see, managers of project companies under either a BOT or a PPP contract have
ample incentives over the length of the project to engage in managerial misbehavior and selfdealing
behavior before turning over the project to the government. Since most of the existing
large-scale infrastructure projects around the world under BOT or PPP format were launched less
than a decade ago, it is too early to witness any managerial incentive problems associated with
the existing large-scale infrastructure projects.
Motivated by these observations regarding the shortcomings of the current BOT and PPP
approaches, I now propose a revolutionary new approach—one that brings with it correct
financial tools and proper economic incentives—that holds out the promise of improving the
critical process of infrastructure project financing in developed and developing countries around
the world.
The new proposal in this paper is a simple one. Successful financing of major
infrastructure developments is too important and too vital to the rapid economic growth of many
countries in the world to be simply left to the hands of politicians. Instead, project finance
should be guided by the capital market's invisible hand. And a key part of such financing must
come from financial innovations. The fundamental challenge of infrastructure financing, then, is
24 Source: Global News Wire, 10/3/2003.

15
how to match the massive demand for capital investments in the world with a supply of capital
from millions of private investors through project securities available on a global scale.
3. A New Approach to Project Financing through Capital Markets
Since the widely-used BOT and PPP approaches have not produced satisfactory results in
terms of the performance of large-scale infrastructure projects around the world in the past
decade, a new approach is needed. The capital markets provide fundamental opportunities for
risk-sharing among millions of private investors through financial innovations. As I have noted
in Chen (2002), motivations for financial innovation are no doubt complex and diverse, but they
can be best understood by the simple acronym TRICKME, in which: ‘T’ stands for reduction of
taxes and transactions costs; ‘R’ for circumvention of bad regulations; ‘I’ for reduction of
informational asymmetry; ‘C’ for market completion; ‘K’ for knowledge advancement; ‘M’ for
marketing efforts for new products and new processes; and ‘E’ for financial engineering. 25
This
acronym captures the idea that newly-developed financial instruments can be used to enable
financing infrastructures in a way that will reduce agency costs arising from bureaucratic
regulations and informational asymmetries. It is my view that the operation of the capital
market’s invisible hand will prove to be the most efficient and most successful way to achieve
desired results in infrastructure development. While I think the principle of Kane’s (1977)
“regulatory dialectic” applies perfectly to a relationship between government regulators and
managers of project companies, it is my belief that the invisible hand of capital markets will
25 It should be pointed out that Miller (1986) claims that the prime motivators for financial innovations in the past
are frequent and unanticipated changes in regulatory and tax codes.

16
prove to be more efficient in achieving the desired results in infrastructure developments around
the world. 26
The core of the new approach to financing infrastructure is the idea that the economic
costs and benefits of infrastructure projects can and should be determined by placing the project
financing in the hands of millions of global private investors from the start. In other words, I
advocate using the invisible hand—the capital market—to determine the economic value of an
infrastructure project and to provide the necessary capital for its construction, operations and
maintenance. By utilizing initial public offerings (IPOs) of project securities for any new largescale
infrastructure projects, the true “private participation” of the infrastructure projects for
economic developments in a nation can be achieved immediately. The new approach
substantially reduces the lack of diversification and lack of liquidity faced by both shareholders
and bank creditors of project companies in the BOT or the PPP approaches. Before discussing in
more detail some financial innovations relevant to IPOs of project bonds, project stocks and
project preferred stocks, I first would like to highlight some of the key benefits of the proposed
approach:
1) Issuing project securities for a new infrastructure project in both domestic and foreign
capital markets would attract funds from millions of investors at home and abroad, 27
thereby helping to ensure ample funding and strong interest in and awareness of the
infrastructure project on a global scale. As a result, the “globalization” of the
infrastructure project can be achieved.
26 Kane (1977) states that “introducing political power into economic affairs initiates a dialectical process of
adjustments and counter-adjustments. In what resembles reflex actions, markets rechannel regulatory power, as
regulators short-circuit regulator intentions…by finding and exploiting loopholes.”
27 The project securities can be issued at relatively small unit prices to attract both domestic and foreign investors.
The project investors will include individual investors, investment banks, commercial banks, mutual funds,
insurance companies, and even governments.

17
2) Sufficient numbers of bonds and equity shares of the project company can be issued
to establish reasonable unit prices of project financing instruments, thereby ensuring
broad participation in the domestic as well as foreign capital markets. As Miller and
Puthenpurackal (2004) have recently shown, Global bonds’ fungibility allows them to
be placed simultaneously in bond markets around the world and to lower the cost of
debt capital to the issuers. Therefore, it is important to increase the liquidity of project
securities by making large IPO issues of project securities in different equity and
bond markets around the world and by making them more attractive to foreign
investors through financial innovations to reduce the informational asymmetry about
the project. It is also important to note that a high degree of community interest
within the domestic country is necessary for the ultimate success of the infrastructure
project. 28
3) Placing ownership immediately in private investors’ hands fosters efficiency and
liquidity in the market for claims on the future cash flows of the infrastructure
projects. Transparent economic information of an infrastructure project and due
diligent analysis of the project’s future cash flows would establish the market values
of project securities with different claims on the future cash flows that can attract
various types of investors at the IPO stage as well as in the subsequent secondary
markets. In case some infrastructures have the characteristics of natural monopoly,
certain types of public services commissions can be set up to determine the prices of
28 The Thai people used to call the Hopewell’s road and train system BOT project in Bangkok the “hopeless” project.
Clearly, lack of community interest in the project was one of the key factors contributing to its failure.

18
services or products to allow a “fair rate of return” for the project stockholders in
accordance to the Hope case 29 in that the Court states “…the return to the equity
owner should be commensurate with returns on investments in other enterprisewould
be allowed to earn a fair rate of return
4) Publicly-traded project stocks would provide a foundation for granting stock options
to construction workers and project operation employees that could help to improve
incentives and productivity at the construction phase as well as at the operational
phase. Consequently, the currently widespread agency problems that create cost
overrun and inefficiency can be reduced.
5) The governments in the host country could be allocated some units of project
securities, so a true public-private-partnership can be achieved at the very beginning
of any new infrastructure project, which will in turn reduce a great deal of agency
problems and agency costs we have mentioned before.
3.1 Issuing Project Bonds with Event Risk Provisions
In general, the private ownership of infrastructure through the BOT or PPP format
involves both shareholders as well as creditors of project companies. However, they are few and
lack diversity. The shareholders of project companies are usually represented by a small number
of domestic and foreign project sponsoring firms. Similarly, the small numbers of creditors are
typically represented by domestic and international lending banks. Without a large number of
diverse private investors that hold project equity shares and project bonds, the benefits of risk-
29 In the Hope case the Court states “… the return to the equity owner should be commensurate with returns on
investments in other enterprises having corresponding risks. That return, moreover, should be sufficient to assure
confidence in the financial integrity of the enterprise, so as to maintain its credit and to attract capital.” See the
Supreme Court’s decision on the landmark Federal Power Commission v. Hope National Gas Company, 320 U.S.
591, case.

19
sharing and liquidity cannot be obtained. Therefore, in addition to project stocks, a large
number of project bonds and project preferred stocks should also be issued at initial public
offerings. The project bonds should have tax-free status in order to attract private domestic and
foreign bond-investors. Furthermore, various types of project bonds can be structured to have
different maturities and priorities in order to attract different kinds of private bond-investors
worldwide.
Just like corporate bonds, project bonds may be subject to some unforeseen adverse
shocks such as wars, political upheaval, tax law changes, earthquakes, oil price shocks,
exchange-rate fluctuations, etc. Occurrence of such event risks would affect the total value of
infrastructure and consequently the value of project bonds and hence bondholders’ wealth.
Typically, the corporate bondholders in the United States have protected their wealth from
certain types of event risks in the late 1980s and the early 1990s by demanding that new bond
issues should include “poison puts” or “event-risk provisions” 30 in their indentures. Also,
Standard & Poor’s Corporation started to add the “event risk” rating to its credit risk rating of
corporate bonds in the beginning of the 1990s. The categories of event-risk rating range from
the strongest protection against event risk (E1) to the insignificant or no protection against event
risk (E5). 31
Let me now describe two financial innovations that have been incorporated in the
covenants of corporate bonds first introduced in the early 1990s. These innovations in corporate
bonds can be applied to the project bonds in financing infrastructure projects as I have discussed
earlier. For convenience, the first kind of innovation in the covenants of bonds will be called
30 They have been dubbed “poison” because they may make a company indigestible for a hostile bidder.
31 The event-risk bond covenant ratings are an outgrowth of the Standard & Poor’s Event Risk Task Force (1989)
which investigated bondholders’ protection against event risk due to takeovers or recapitalizations.

20
“protective put-type” bonds with “bearish” event-risk provisions. Bonds with “bearish” event
risk provisions are similar to regular puttable bonds. However, there is a key difference between
these two types of bonds, namely, that the exercise of a put option attached to the former is
triggered by the occurrence of an external risk event, while the exercise of the put option
attached to the latter is determined by bondholders. We can refer to the second kind of
innovation in the covenants of corporate bonds as “conversion call-type” bonds that have
“bullish” event-risk provisions. Bonds with “bullish” event risk provisions are similar to regular
convertible bonds. Again, the difference between the two depends on the trigger point for the
conversion. 32
3.1.1 Project Bonds with Bearish Event-Risk Provisions
Project bonds with bearish event-risk provisions should protect bondholders from wealth
loss due to the occurrence of some designated events—such as wars, terrorists attacks, changes
in political systems, changes in world oil prices, and changes in foreign exchange rates—that
might cause a significant decline in the economic value of an infrastructure project and hence, a
drastic downgrade in the rating of project bonds. If such an event has occurred, then the project
bond has effectively matured and the bondholders have the right to ask the issuer or guarantors
for immediate payment of the principal amount of the bond and the accrued interest. In this way,
a project bond with bearish event-risk provisions can be viewed as a puttable bond and should
have a value greater than that of the straight project bond. The bearish event-risk provisions
should therefore serve to make investing in the project more attractive to both domestic and
foreign private bond investors.
3.1.2 Project Bonds with Bullish Event-Risk Provisions
32 The contingent-claim analysis has been applied to formulate the valuation models for bonds with various eventrisk
provisions in Bicksler and Chen (1992).

21
Project bonds with bullish event-risk provisions can provide bondholders with
opportunities for wealth gains by converting debt into equity upon the occurrence of some
designated event. These bullish events can include favorable changes in exchange rates, tax laws,
and energy prices that have resulted in a significant increase in the economic value of an
infrastructure project, especially the value of equity shares. If such an event occurs, then the
holders of project bonds with bullish event-risk provisions can immediately convert the debt into
equity shares.
3.1.3 Valuation of Risky Project Bonds with Event-Risk Provisions
This section presents a general valuation model for risky project bonds with both types of
event-risk provisions. The market value of the infrastructure project of the bonds with event-risk
protection is described as a mixture of both diffusion and jump processes. The total change in
the value of the infrastructure is assumed to consist of two types of changes:
(1) The small and “normal” changes in the market value of the project due to the
financial developments in the project. This component is modeled by a standard
geometric Brownian motion with a constant variance per unit time and it has a
continuous sample path.
(2) The large and “abnormal” changes in the market value of the project that are due
to the arrival of unanticipated events that significantly affect the market value of
the project. These events include wars, changes in political systems, changes in
the tax structure, changes in world oil prices, and changes in foreign exchange
rates. This important unanticipated information arrives infrequently only at
discrete points in time. This component of the project value is modeled by a

22
“jump” process reflecting the significant impact of this unanticipated information
on the project value and its security pricies.
Given the above specifications, the changes in the value of the infrastructure can be
formally written as the following stochastic differential equation:
where
dV / V = (µ − λκ ) dt + σdW
+ dq
(1)
V = the value of the infrastructure project;
µ = the instantaneous expected return on the infrastructure value;
σ 2 = the instantaneous variance of the return on infrastructure value, conditional on no
arrival of “abnormal” information;
dW = a standard Wiener process;
dq = a continuous-time Poisson process, assumed to be independent of dW for
simplicity;
λ = the intensity of the Poisson process (the mean number of arrivals of “abnormal”
information per unit time); and
k = E(Y – 1), where (Y – 1) is the random percentage change in the infrastructure value
if the Poisson event occurs and E is the expectation operator over the random
variable, Y.
In Equation (1), the σdW
term describes the instantaneous part of the unanticipated change
in infrastructure value due to the “normal” value changes, and the dq term describes the part due
to the “abnormal” value changes. If λ = 0 , and hence dq = 0, then the infrastructure value change

23
dynamic would be identical with those of the underlying asset dynamics posited by Black and
Scholes (1973) and Merton (1973).
In a continuous-time model, the number of jumps during the infinitesimal time is either
one or zero with probability λdt
and 1 − λdt
, respectively. More precisely, the instantaneous
probability of a Poisson event in the time interval dt is λ dt
no Poisson event occurring in the time interval dt is
, and the instantaneous probability of
1 − λdt
. Thus, the resulting sample path for
the infrastructure value V(t) will be continuous most of the time with finite jumps of differing
signs and amplitudes occurring at discrete in time. Given V(0) = V, and
µ , λ,
k and σ are
constants, the random infrastructure value at time t can be expressed as
2
V ( t)
= V exp[( µ −σ
/ 2−
λκ ) t + σW
( t)]
Y(
n)
(2)
where W(t) is a Wiener random variable with zero mean and variance equal to t; Y(n) = 1 if n = 0;
∏ N j = 1
Y(n) = Y for n ≥ 1, where Yj are the jump amplitudes assumed for simplicity to be
j
independently and identically distributed, and n is Poisson distributed with parameter λ t .
Under the assumptions that the capital asset pricing model (CAPM) holds and that the
jump risk is diversifiable, Merton (1976) shows that the price of any contingent claim, F(V,t),
which is a function of the underlying asset value and time must satisfy the following general
valuation equation:
2
1/
2σ
2 V F
VV
( V,
t)
+ ( r −λκ
) VF
+ λE(
F(
VY,
t)
− F(
V,
t))
= 0
V
( V,
t)
+ F ( V,
t)
− rF(
V , t)
t
(3)

24
which is an integro-differential equation where the expectation is taken over the random value of
the jump amplitude Y, and r is the constant instantaneous riskless rate of interest. The unique
market value of any contingent claim on the value of the infrastructure project is determined by
the initial and boundary conditions.
Assume that the infrastructure has two classes of claims: project stocks and a single
homogeneous class of discount project bonds with bearish and/or bullish event-risk provisions
allowing the bond-holders to redeem the debt at its principal amount, B, at the lower boundary if
a put-event occurs, or to convert the debt into a fraction of the project stocks at the upper
boundary if a call-event occurs. Thus the project debt is a puttable/convertible discount bond
that has a promised payment of B. In the event that the infrastructure fails to meet the promised
payment of debt at maturity date, the ownership of the entire infrastructure is transferred
immediately to the bondholders. Therefore, the value of the project bonds at maturity is the
minimum of either the value of the infrastructure or the maximum of the promised principal or
the convertible value, given that neither the “put event” nor the “call event” has occurred during
the life of the debt.
In the presence of a bearish event-risk provision, it is assumed that, if the lower boundary
is reached by a designed triggering put-event has occurred during the life of the debt, the
bondholders are allowed immediately to redeem the debt at its face value or to take over the
entire infrastructure if the equity-holders cannot meet the redemption payment when a put-event
occurs. Denote K(t) as the minimum safety covenant at time t. If the value of the infrastructure
falls to equal or below the safety barrier K(t), the downgrading of the debt takes place
immediately and hence triggers the put-event. With put rights, the bondholders can demand the
redemption payments of par value of the debt immediately when a put-event occurs; so the value

25
of debt at time t is equal to its face value or the value of the infrastructure if stockholders fail to
meet the redemption payments. Thus, the value of the debt at the lower boundary can be
expressed as F(V,t) = Min [V, B] for V(t) ≤ K(t).
It is clear that the safety barrier at maturity date
is the total promised payment; that is, K(T) = B.
In the presence of a bullish event-risk provision, it is assumed that, if a designated
triggering call-event occurs, the bondholders will convert the debt into a fraction of the
infrastructure’s project stock provided that the conversion value is greater than the face of the
debt. Since the debt and the common stock are the only two claims in the capital structure of the
project, only the common stock remains after the call-event has occurred. The value of the debt
isγ
V
= V
, since it is convertible for a fraction γ of the post-conversion equity. This means that F(V,t)
γ for V(t) ≥ B / γ , where γ is the dilution factor indicates the fraction of the common equity
that would be held by the bondholders if the entire debt were converted. If there are N shares of
common stock outstanding and the debt can be exchanged for n shares in aggregate, then the
dilution factor is γ ≡ n /( n + N).
Thus, the valuation of a risky discount bond with bearish and bullish event-risk
provisions can be written as
2 2
1/
2σ
V F
vv
( V,
t)
+ ( r −λk)
VF
v
( V,
t)
+ F ( V , t)
− rF(
V , t)
+ λE(
F(
VY,
t)
− rF(
V , t)
= 0
t
(4)
subject to
F(0,t) = 0
(4a)
F(V,t)
≤ V
(4b)
F ( V , t)
= Min[
V , B]
for V ≤ K(
t)
at t < T
(4c)
F ( V , t)
= γ V , for V ≥ B/
γ at t < T
(4d)

26
F( V , T)
= Max[
γV
, Min(
V , B)]
(4e)
Conditions (4a) and (4b) show the limited liability for both the project bondholders and the
project equity-holders. Condition (4c) indicates the lower boundary condition for the put-event
provision, and condition (4d) indicates the upper boundary condition for the call-event provision.
Condition (4e) says that the debt is worth the maximum of the conversion value or the maximum
of the value of the infrastructure or the promised principal at maturity, given that neither the
blower boundary from put-event nor the upper boundary from call-event has been reached during
the life of the debt.
It can be seen that the valuation model for a risky project bond with event-risk provisions
expressed in Equation (4) and conditions (4a)--(4e) is a generalization of the valuation of a risky
discount bond with safety covenants studied by Black and Cox (1976) using a diffusion process
and by Masson and Bhattacharya (1981) using a jump process, and by valuation of a non-callable
convertible discount bond studied by Ingersoll (1977) using a diffusion process.
With complex initial and boundary conditions, it is difficult to derive a closed-form
solution to the valuation problem in Equation (4) unless in very restricted cases. Kou and Wang
(2001) show that one can obtain explicit solutions to the problem of first passage times of a
jump/diffusion process if the jump process has jump sizes having a double exponential
distribution. However, a useful approach to solving the equation system (4) and (4a)—(4e) is the
Cox-Ross (1976) risk-neutral pricing technique, which finds the value of a contingent claim by
discounting the expected payoff to the contingent claim in a risk-neutral world. Therefore, a
numerical method such as the finite difference approximation to the partial differential equation

27
suggested by Brennan and Schwartz (1978) can be used to solve the problem of pricing the risky
infrastructure bond with event-risk provisions stated in equation system (4) and (4a)—(4b).
A general characterization of the value of the debt with both the put-event and the callevent
provisions based on the risk-neutral pricing technique can be outlined as follows. From the
valuation equation stated in (4) and (4a)—(4e), we can see that there is a two-sided boundary
imposed upon the infrastructure value change that affects the payments to the bondholders. If
the value reaches the upper or the lower boundary prior to the maturity of the debt, the
bondholders will receive the conversion value or the redemption payments immediately. Thus,
solving the valuation problem of equation system (4) and (4a)—(4e) involves pricing
options with payoffs at random times. Therefore, the holders of the bond with both put-event
and call-event provisions may receive payments upon the occurrence of the put-event, the
occurrence of the call-event or at maturity, provided that neither the put-event nor the call-event
has occurred prior to the maturity. Let F m (V,t) be the portion of debt value due to the payments
to bondholders at maturity, F p (V,t) be the portion of debt value from the payments due to the
occurrence of a put-event, and F c (V,t) be the portion of debt value from the payments due to the
occurrence of a call-event, then the current value of the debt with event-risk provisions should be
the sum of these three parts.
Using the risk-neutral pricing technique, we can write out the three components of the
current value of the discount bond with both put-event and call-event provisions. For the portion
of current debt value due to payments at maturity, conditional on the infrastructure value not
reaching the upper and lower boundaries prior to the maturity of the debt, we have the following
expression:

28
− r T −t)
K ( T )
Fm[ V,
t;
K(
t),
B,
γ ] = e ∫ V ( T)
p(
V ( T),
T;
Vt ) dV (
(
T
0
( )
+ e
− r T − t
∫ ∞ Max[
γV
( T),
B]
p(
V ( T),
T;
Vt ) dV ( T),
(5)
K ( T )
)
where p(·)dV is the (defective) probability density function of the market value of the
infrastructure at maturity T, conditional on the current market value of the infrastructure being V t
and conditional on neither the put-event nor the call-event occurring prior to the maturity. This
probability density is defective since the probability of reaching maturity T is not one. The
probability of project’s default at maturity is the integral of p(.)dV over the interval [0,K(T)], and
the probability that the project is not in default at maturity is the integral of p(.)dV over the
interval [K(T),
∞
]. The first integral in Equation (5) gives the portion of debt value in the event
the project is in default at maturity, given that no put-event or call-event has occurred prior to the
maturity of the debt. The second integral in Equation (5) gives the portion of debt value arising
from the promised principal or conversion value at maturity, given that no put-event or call-event
has occurred previously and that the project is not in default at maturity.
For the portion of debt value due to the payments from the occurrence of a put-event, we
have the following expression,
where
Fp
V , t;
K(
t),
B,
γ ]
e
Min[
V ( s),
B]
g ( s;
V
T − r ( s −t)
[ = ∫t
1 t )
ds , (6)
( ) is the first passage time density function of the project value to the lower boundary,
g 1 •
that is the probability density function of a put-event occurring at time s, conditional on the
project’s current value of Vt and on neither a put-event nor call-event having previously occurred.
The integral in Equation (6) gives the value of the bond with event-risk provisions due to the

29
redemption payments from exercising a put-right prior to the maturity of the debt. If a put-event
occurs at time s, the bondholders receive the redemption payments equal to the promised
principal of the debt or the value of the project at that time V(s) if the stockholders fail to meet
the redemption payments.
The general properties of
F p
[•
] and their effects on the value and the risk of infrastructure
debt with even-risk provisions are clear. For instance, the probability of reaching the lower
boundary (i.e., V(t)
≤ K(t)
) is an increasing function of the variance rate of the project value, the
frequency of the arrivals of abnormal information about project-related events ( λ ) and the jump
size of change in the project value (Y). As a result, the effective maturity date of the bond is
sooner and the bondholders obtain a greater value from the debt.
The portion of debt value due to the payments from the call-event occurring can be
expressed as follows:
T − r
[ , ; ( ), , ]
( h −t)
Fc
V t K t B γ = ∫ e γV
( h)
g 2 ( h;
Vt
) dh ,
t
(7)
where
g
( ) is the first passage time density function of project value to the upper boundary of a
2 •
call-event occurring at time h, conditional on a current value of Vt for the project and conditional
on neither put-event nor call-event having previously occurred. The integral in Equation (7)
gives the portion of debt value arising from the possibility of receiving a conversion value of γ V
upon the occurrence of the call-event and converting the debt into common equity of the
infrastructure. The general properties of
F c [•]
and the effects on the value and risk of project
debt with event-risk provisions are similar to that of F p [•]

30
3.1.4 Other Economic Implications of Project Bonds with Event-Risk Provisions
Issuing project bonds that include event-risk provisions may have some other important
economic implications for an infrastructure project. First, in the presence of bearish event-risk
provisions, managers and stockholders of project companies are less likely to engage in project
restructuring activities that might enhance equity value at the expense of bondholders. Therefore,
with event-risk provisions in project bond indentures, agency problems between bondholders and
stockholders of project companies and their associated agency costs can be reduced. As a result,
the decrease in deadweight costs should improve the overall efficiency of infrastructure projects.
Second, the inclusion of event-risk provisions in project bond contracts should be a
deterrent to politicians’ attempts to make undesirable policy changes in the host countries. Thus,
a more stable economic growth can be expected.
Third, project bonds with event-risk provisions in the contracts will make them more
attractive to both domestic and foreign private bond-investors because they represent a package
of project bonds and the attached put/call options. Thus, the explicit costs of debt financing for
an infrastructure project will be lower with the inclusion of event-risk provisions in its debt.
3.2 Financing with Project Stocks
Any infrastructure project can be financed by the mechanism of initial public offerings of
project securities—stocks and bonds. With the assistance of some internationally reputable
investment bankers, the economic value of an infrastructure project could be estimated with
reasonable accuracy, and certainly as easily as estimating the value to any potential foreign direct
investor interested in the project. The possibility of wider participation in the project would also
reduce the financial risks borne by a wider base of investors; hence the costs of raising the

31
required capital for the project would be lower. Thus, a world-wide IPO of project stocks would
raise funds for the project from global private investors immediately. The invisible hand of the
global capital markets would work better in financing the infrastructure project than the visible
hands of domestic politicians.
Due to the high degree of informational asymmetry of an infrastructure project, foreign
private investors might naturally require some guarantee from the project company or other
guarantors as to the future returns on their project investments. In the following, I will briefly
describe two methods of issuing project stocks that could reduce the costs associated with
informational asymmetries. The first is to perform an IPO with puttable project stock, and the
second is to perform an IPO of project stock with contingent-value-rights (CVRs) attached.
3.2.1 Issuing Puttable Project Stock
One of the financial innovations of investment bankers in the late 1980s was common
stock that could be “put” back to the issuer. A “puttable stock” can be used to reduce the
underpricing problem in some initial public offerings (IPOs) as well as to resolve other problems
arising from informational asymmetry. Puttable project stock is simply a package consisting of
project stock and a put option on the project stock. 33
In simple terms, it is a “money-back”
guarantee that can be provided by the project company or other international guarantors of the
infrastructure project. Purchasers of the puttable project stock obtain the right (the put option) to
redeem the shares at some prespecified price (the minimum guaranteed or the strike price). Thus,
the downside risk due to a decline in the share price is eliminated with puttable stock. Such an
arrangement would appeal to potential foreign investors who might otherwise be reluctant to
invest without the “money-back” guarantee. The innovation of puttable project stocks provides
33 See, for example, Chen and Kensinger (1988) for more discussion and analysis of the downside protection and
upside potential of puttable stocks.

32
attractive investment opportunities for private domestic and foreign investors. Indeed, although
a homemade “synthetic protective put” strategy could be created by using a dynamic asset
allocation scheme, high transactions costs will prevent private domestic and foreign investors
from using such a strategy. A simple example can be used to illustrate how puttable stock can
solve the IPO underpricing problem. Suppose that one billion shares of the new project stock are
issued for an economic value of $10 billion worth of the equity portion of the infrastructure
project. If the project company or a guarantor is willing to redeem the shares at $10 per share,
say within the next three years after the issuance date, then the value of the putttable project
stock would be worth more than $10 per share because of the value of its implicit put option.
3.2.2 Issuing Project Stock with CVRs
Investors’ commonly portrayed investment psychology of “fear for downside risk and
greed for upside gain” can be exploited with another strategy. In particular, the project stocks of
an infrastructure offered to domestic and foreign investors can be combined with Contingent
Value Rights (CVRs), which permit investors to reap the benefits of possible upside appreciation
while maintaining downside protection. 34 Thus, a project stock plus CVRs can be even more
attractive than the puttable-stock to risk-averse foreign investors seeking to invest in the shares
of a new infrastructure project in an environment with large information asymmetries. In
essence, the CVR guarantees a base price for the project stock while establishing a target price
which the stock is expected to achieve prior to the CVRs expiration. In the language of financial
engineering, a CVR is really what is known as a “bearish put spread.” In other words, holders of
34 See, for example Chen and Kensinger (1992), and Chen, Chen and Laiss (1993) for the discussion and analysis of
using CVRs to reduce the agency costs of asymmetric information in corporate mergers and acquisitions. The
CVRs were used by Dow Chemical Co. to acquire Marion Laboratories in 1989, and by the French government in
its Rhône Poulenc S.A.’s acquisition of Rorer Group Inc. in 1990. In 1993, the Pension Benefit Guaranty Corp.
issued CVRs to the creditors of LTV Corp. in order to get them to approve a settlement of LTV’s lengthy

33
a CVR own a portfolio of two options on the same underlying project stock, consisting of a long
position in a put option at a higher strike price (target price) and a short position in a put option
at a lower strike price (base price). It should be noted that the payoffs of CVRs relative to base
and target prices are usually calculated based on the average prices of the underlying stock for a
number of days. In other words, valuation of the CVRs involves pricing of the average-price (or
Asian) options. 35
In the following, the valuation of a simple CVR will be discussed. The payoff for a CVR
at expiration is simply the difference between the payoff for a long position in a put option at the
target price (TP) and the payoff for a short position in a put option at the base price (BP).
Thus, we can express the payoff to a CVR as follows:
CVR = Put (TP) – Put (BP)
= Max [X TP - S T , 0] – Max [X BP – S T , 0] (8)
where,
X TP = the exercise price of a put option at the target price;
X BP = the exercise price of a put option at the base price;
S T = the price of project stock at the expiration.
Figure 1 depicts the payoffs at expiration for holding a CVR (Panel A) and holding both a
CVR and project stock (Panel B). Let us assume that the base price is $30 and the target price is
$50. As Figure 2 shows, the payoff for a CVR at expiration is simply the difference between the
bankruptcy. In 1994, Viacom used CVRs in its acquisition of Paramount Communications Inc. A recent usage of
CVRs was in General Mills’ successful acquisition of Pillsbury from Diageo in October 2001.
35 See Chen, Chen and Laiss (1993) for the discussion of proposed methods of pricing Asian options and the
references therein. They have analysized the prices of Marion Merrell Dow’s CVRs based upon an application of a
closed-form analytical approximation for the valuation of arithmetic options.

34
payoff for a long position in a put at the target price (TP) = $50 and the payoff for a short
position in a put at the base price (BP) = $30.
=========================
Insert Figure 1 here
=========================
An example will illustrate how the stock-plus-CVR combination preserves the downside
protection and at the same time enjoys the upside potential. Assume again that the equity portion
of an infrastructure project is worth $10 billion and that 1 billion shares of project stocks are
offered in the local and global capital markets at a price of $10 per share. If the base price is set
at $30 per share and the target price is set at $50 per share, then at the CVR expiration date (e.g.,
3 years after the date of issue) the market value of the package will be as follows in Table 1.
====================
Insert Table 1 here
====================
3.3 Financing with Redeemable and Convertible Project Preferred Stocks
Project preferred stock has features similar to both project stocks and project bonds of the
project company. It resembles a project bond in that 1) it promises to pay to its holder fixed
dividends each year, and 2) it does not have the voting power regarding the management of the
project company. Project preferred stock is an equity investment in the sense that failure to pay
the dividend does not precipitate bankruptcy of the project company.
In addition to issuing the ordinary project preferred stocks, the project company can also
issue redeemable project preferred stocks that are callable by the project company within some
predetermined period of time. By issuing the redeemable project preferred stocks, the project

35
company has effectively acquired call options on the preferred stocks from the investors and paid
the call premiums in terms of promising higher preferred dividends than that on the ordinary
project preferred stocks. This callable project preferred stock will be the same as what known as
PERCS, the Preferred Equity Redemption Cumulative Stocks, employed by many major “blue
chip” companies such as General Motors, Sears, Texas Instruments, Citicorp and others in the
recent past. In their empirical study, Chen et al (1994) show that PERCS provides advantages in
transaction costs and taxes to its investors.
Furthermore, the project company can also issue convertible project preferred stocks that
can be converted into project stocks at options of their holders. The convertible project preferred
stocks can be attractive to certain types of preferred stock investors in the domestic and global
capital markets.
4. Conclusion
In this paper, I have described some of the major problems associated with the current
practice of using the BOT and PPP approaches to finance infrastructure projects around the
world. The problems emerged from the fact that the creation of project companies under the
current BOT and PPP approaches is determined and influenced by the politicians of the host
country and hence is subject to a great deal of political risks. In addition, the structural
characteristics of the project companies under the current practice with concentrated equity
ownership and illiquid syndicated bank loans have created severe agency problems and agency
costs. I have argued that such problems can be avoided--or at least significantly reduced--if the
project financing is guided by the invisible hand of the capital markets through new designs of
project securities. In the new approach, the project companies of any new large-scale
infrastructures will be formed by using the proceeds from the IPOs of project securities – project

36
stocks, project bonds, and project preferred stocks – on a global scale. With financial
innovations in security designs, the project securities could be traded in capital markets around
the world. Therefore, I believe that the principal theories that we have learned from the
traditional “corporate finance” can be applied to address the financial problems of infrastructure
project finance.
Of course, access to well-functioning and globally linked capital markets is a prerequisite
for raising capital in the fashion I have proposed. In other words, for any nation to develop and
finance infrastructure projects most effectively, its capital markets will have to exhibit the key
characteristics of well-functioning capital markets, including freedom from insider trading and
security manipulation, advanced bodies of corporate and securities law, effective contract
enforcement, well-developed accounting standards, and transparent disclosure of accounting and
financial information. 36 The development of capital markets, legal systems, and accounting
systems are themselves important infrastructure projects vital to the economic growth of a
country. 37 If economies in the world can achieve such development, then we can hope that
financial innovations of the sort I have discussed earlier will be even more useful in facilitating
the financing of infrastructure development in any nation.
36 La Porta et al (1998) show the importance of legal systems as a determinant of financial development and
economic growth.
37 Merton (1990) presents an excellent and thorough discussion of this point.

37
References
Bicksler, J. and Chen, A., “Pricing Corporate Debt with Event-risk Provisions,” International
Review of Financial Analysis, Vol.1, 1992, 51-63.
Black, F. and Cox, J., “Valuing Corporate Securities: Some Effects of Bond Indenture
Provision,” Journal of Finance, Vol. 31, 1976, 351-367.
Black, F. and Scholes, M., “The Pricing of Options and Corporate Liabilities,” Journal of
Political Economy, Vol. 81, 1973, 637-659.
Bradford, M., “The British Model of Private Finance Initiative and Public-Private Partnership
Ten Years Later: Toward International Extension in the Defense Sector?” The Journal of
Structured and Project Finance, Fall 2001, 61-71.
Brennan, M. and Schwartz, E., “Finite Difference Methods and Jump Processes Arising in
the Pricing of Contingent Claims: A Synthesis,” Journal of Financial and Quantitative
Analysis, Vol. 13, 1978, 461-474.
Chen, A., “A New Perspective on Infrastructure Financing in Asia,” Pacific-Basin Finance
Journal, Vol. 10, 2002, 157-165.
Chen, A., Chen K. and Laiss, B., “Pricing Contingent Value Rights: Theory and Practice,”
Journal of Financial Engineering, Vol. 2, 1993, 155-173.
Chen, A. and Kensinger, J., “Puttable Stock: A New Innovation in Equity Financing,”
Financial Management, Vol.17, 1988, 27-37.
Chen, A. and Kensinger, J., “Uncommon Equity,” Journal of Applied Corporate Finance,
Vol. 1, 1992, 36-43.
Cox, J. and Ross, S., “The Valuation of Options for Alternative Stochastic Processes,”
Journal of Financial Economics, Vol. 9, 1976,145-166.
Esty, B., “Why Study large Projects? An Introduction to Research on Project Finance,”
European Financial Management, Vol. 10, 2004a, 213-224.
Esty, B., “When Do Foreign Banks Finance Domestic Projects? New Evidence on the
Importance of Legal and Financial Systems,” Working Paper, Harvard Business School,
September 22, 2004b.
Esty, B. and Megginson, W., “Creditor Rights, Enforcement, and Debt Ownership Structure:
Evidence from the Global Syndicated Loan Market,” Journal of Financial and Quantitative
Analysis, Vol. 38, 2003, 37-59.

38
Fabozzi, F., and Nevitt, P., Project Financing, 7 th ed. (London: Euromoney, 2000)
Flyvbjerg, B., Bruzelius, N., and Rothengatter, W., Megaprojects and Risk: an Anatomy of
Ambition (Port Chester NY: Cambridge University Press, 2003).
Ingersoll, J., “A Contingent Claims Valuation of Convertible Securities,” Journal of
Financial Economics, Vol. 4, 1977, 269-322.
Kane, E., “Good Intentions and Unintended Evil: The Case against Selective Credit
Allocation,” Journal of Money, Credit and Banking, Vol. 9, 1977, 55-69.
Kou, S. G., and Wang, H., “First Passage Times of a Jump Diffusion Process,” Working
Paper, Columbia University, April 24, 2001.
La Porta, R. and Lopez-de-Silanes, F., “The Benefits of Privatization: Evidence form
Mexico,” Quarterly Journal of Economics, Vol. 114, 1999, 1193-1143.
La Porta, R., Lopez-de-Silanes, F., Shleifer, A. and Vishny, R., “Law and Finance,” Journal
of Political Economy, Vol. 106, 1998, 1113-1155.
Masson, S., and Bhattacharya, A., “Risky Debt, Jump Processes, and Safety Covenants,”
Journal of Financial Economics, Vol. 9, 1981, 281-307.
Merton, R., “A Rational Theory of Option Pricing,” Bell Journal of Economics and
Management Science, Vol. 4, 1976, 141-183.
Merton, R., “Option Pricing when Underlying Stock Returns are Discontinuous,” Journal of
Financial Economics, Vol. 3, 1976, 125-144.
Merton, R., “The Financial System and Economic Performance,” Journal of Financial
Services Research, 1990, 263-300.
Miller, D. and Puthenpurackal, J., “Security Fungibility and the Cost of Capital: Evidence
from Global Bonds,” Journal of Financial and Quantitative Analysis, 2004, forthcoming.
Miller, M. H., “Financial Innovation: the Last Twenty Years and the Next,” Journal of
Financial and Quantitative Analysis, Vol. 21, 1986, 459-471.
Miller, R., Lessard, D., The Strategic Management of Large Engineering Projects
(Cambridge MA: MIT Press, 2000).
Modigliani, F., and Miller, M., “The Cost of Capital, Corporation Finance and the Theory of
Investment,” American Economic Review, Vol. 53, June 1958, 261-97.
Pope, H., “Caspian Pipeline Financing Solidifies U.S. Policy,” Wall Street Journal, February
3, 2004, A13.

39
Figure 1
The Payoffs of CVR and (Stock + CVR) at Expiration
$
Panel A: Payoff of CVR
BP = $30
TP = $50
20
0 30 50 Stock Price
$
Panel B: Payoff of (Stock + CVR)
50
20
0 30 50
Stock Price

40
Table 1
The Value of (Stock + CVRs) at Expiration Date
Stock Price Short Put Long Put Stock + CVR
($) @$30 @$50 ($)
0 -30 +50 +20
5 -25 +45 +25
30 0 +20 +50
40 0 +10 +50
50 0 0 +50
80 0 0 +80
100 0 0 +100