Free PDF Ebooks Files @AcrobatPlanet.Com

Modigliani and Miller (1958) show that corporate financing decisions do not affect firm value under certain conditions. Their “irrelevance” proposition is powerful because it highlights the factors that make financing decisions value relevant. One of the key assumptions underlying their irrelevance proposition is that that financing and investment decisions are separable and independent activities. When this assumption holds, various financing decisions such as the firm’s organizational, capital, board, and ownership structures do not affect investment decisions or subsequent cash flows.

The rise of project finance, defined as the creation of a legally-independent project company financed with nonrecourse debt, provides strong prima facia evidence that financing structures do, indeed, matter. Total project-financed investment has grown from less than $10 billion per year in the late 1980s to almost $220 billion in 2001 (Esty, 2002a). Within the United States, firms financed $68 billion of capital expenditures through project companies in 2001, approximately twice the amount raised in initial public offerings (IPOs) or invested by venture capital firms. While considerably smaller than the U.S. leasing, asset-backed security, and corporate debt markets—$240 billion, $354 billion, and $434 billion, respectively, in 2001—project finance is, nevertheless, one of the most important financing vehicles for investments in the natural resources and infrastructure sectors such as power plants, toll roads, mines, pipelines, and telecommunications systems. Despite an estimated 40% decline in the use of project finance during 2002, it is likely to grow in the coming years as cash-strapped governments in both developed and developing countries seek ways to finance desperately needed infrastructure investments with private sector participation and capital.

Given the fact that it takes longer and costs more to structure a legally independent project company than to finance a similar asset as part of a corporate balance sheet, it is not immediately clear why firms use project finance. For it to be rational, project finance must entail significant countervailing benefits to offset the incremental transaction costs and time. Yet these benefits are not well understood nor have they been accurately described in the academic or practitioner literatures. Nevitt and Fabozzi (2000, p. 5), for example, claim that, “Project financing can sometimes be used to improve the return on the capital invested in a project by leveraging the investment to a greater extent than would be possible in a straight commercial financing of the project.” While it is true that leverage increases expected equity returns, this motivation for using project finance fails to recognize that higher leverage also increases equity risk and expected distress costs. By itself, this explanation does not provide a compelling reason to use of project finance.

Drawing on existing finance theory, detailed case studies, and extensive field research, I describe the three primary motivations for using project finance. The motivations explain why financing assets separately with nonrecourse debt creates value, and why it can create more value than financing assets jointly with corporate debt, the most likely financing alternative. I argue that project finance solves two financing problems: 1) it reduces the cost of agency conflicts inside project companies; and 2) it reduces the opportunity cost of underinvestment due to leverage and incremental distress costs in sponsoring firms.

Before describing these motivations, I need to highlight an important assumption underlying my analysis and arguments. Project finance involves both an investment decision involving a capital asset and a financing decision. Looking first at the investment decision, McConnell and Muscarella (1985) show that firms experience positive and significant returns when they announce increases in capital expenditures. This finding differs from the announcements regarding acquisitions: Jensen and Ruback (1983) show that merger announcements generate non-positive returns for acquirers. Whereas investment decisions, particularly the decision to acquire, could reflect empire building by managers, it is more difficult to imagine reasons why a manager might overspend on financing an acquisition or an investment or what personal benefits he or she could derive from such a financing choice.

For this reason, I assume the decision to use project finance—a change away from the traditional way of financing investment opportunities—reflects an attempt by managers to reduce total financing costs. Alternatively, one can interpret the decision to use project finance as a neutral mutation with no effect on firm value or as a manifestation of value destroying agency conflicts implemented by managers guided by personal, not firm, objectives. The amount of money at stake makes it likely the financing decisions are careful and deliberate, not expedient. The need to raise external capital, typically from banks, makes it significantly more difficult to finance negative NPV projects. Convincing bankers, who have limited upside potential yet bear significant downside exposure, to provide the majority of the capital is an important constraint on the investment process. In fact, Bharadwaj and Shivdasani (2003) show that the returns to firms announcing cash tender offers increase proportionally with the amount financed by bank debt. Under the critical assumption that financing choices reflect rational attempts to increase firm value, I describe how project finance can reduce the financing costs associated with new capital investments.

The first motivation to use project finance, the agency cost motivation, recognizes that certain assets, namely large, tangible assets with high free cash flows, are susceptible to costly agency conflicts. The creation of a project company provides an opportunity to create a new, asset-specific governance system to address the conflicts between ownership and control. In many ways, the observed governance structures in project companies resemble leveraged-buyouts (LBOs) and achieve many of the same results described by Jensen (1989) and Kaplan (1989 and 1991). What makes project companies a particularly attractive setting in which to study free cash flow problems is the fact that they have few growth options. Most projects are, in fact, wasting assets (e.g., a gold mine) that optimally need to shrink over time.

Project structures can also reduce agency conflicts between owners and related parties. The transaction-specific nature of project assets creates a need to deter strategic behavior by suppliers of critical inputs or expropriation by host governments. The threat of opportunistic behavior or “hold-up” is especially severe in project companies where the deals typically involve negotiations between bi-lateral monopolists. Project companies utilize joint ownership and high leverage to discourage costly agency conflicts among participants. Today, these agency cost motivations remain the most important reasons why firms use project finance.

In contrast to the agency cost motivation, which relates to the asset being financed, the two underinvestment motivations relate to the firms making the capital investments—these firms are known as “sponsoring firms” or “sponsors.” Although underinvestment in positive net present value (NPV) projects can occur for many reasons, I focus on the effects of leverage and incremental distress costs as two important reasons, and show how project finance mitigates both effects. Project finance solves leverage induced underinvestment by allocating project returns to new capital providers in a way that cannot be replicated using corporate debt. This debt overhang motivation is similar to the motivation described by Stulz and Johnson (1985) for using secured debt, but it is even more effective because it eliminates all recourse to the sponsor’s balance sheet and it eliminates the possibility that new capital will subsidize pre-existing claims with higher seniority or reduce the value of junior claims (Myers, 1977). While it is true that the origin of the debt overhang problem is also an agency conflict, I distinguish the debt overhang motivation from the agency cost motivation because the conflict occurs at the sponsor rather than the project level.

The third motivation, risk management, recognizes that investing in risky assets can generate incremental distress costs for sponsoring firms. When these indirect or collateral distress costs are sufficiently large, at least in expectation, they can exceed the asset’s net present value (NPV), thereby turning a positive NPV project into a negative investment (the total NPV is negative). By isolating the asset in a standalone project company, project finance reduces the possibility of risk contamination, the phenomenon whereby a failing asset drags an otherwise healthy sponsoring firm into distress. It also reduces the possibility that a risky asset will impose indirect distress costs on a sponsoring firm even short of actual default. For example, Lamont (1997) shows how a shock to oil prices affected investment decisions in nonoil subsidiaries. Similarly, a large loss on a corporate-financed asset could affect investment decisions in the sponsor’s other divisions, or investment decisions by related firms involved in joint production with the sponsoring firm. Even when project debt is fairly priced, the expectation of costly externalities of this kind can discourage investment.

Although I cast the risk management motivation in terms of a rational manager rejecting a positive NPV asset because of the incremental distress costs it imposes on the sponsoring firm, I can, alternatively, describe this motivation in terms of a risk averse manager seeking to protect his or her poorly diversified human capital. While the end result is the same—failure to invest in positive NPV projects—the underlying causes are different. In the former case, it is about a rational, risk-neutral manager avoiding a negative NPV investment due to the incremental distress costs while in the latter case, it is about an agency conflict between ownership (risk neutral shareholders) and control (risk averse managers).

Compared to the first two motivations, what is different about the risk management motivation is the interaction between the sponsoring firm and the asset: the agency cost motivation addresses the asset while the debt overhang motivation addresses the sponsoring firm. The risk management motivation has not been described previously in the finance literature, yet it is consistent with an emerging literature on the magnitude of investment distortions (Parrino, Poteshman, and Weisbach, 2002). It is also consistent with Stulz’s (1999) observation that the same market imperfections we readily incorporate into capital structure and risk management theories, such as distress costs, are often ignored in capital budgeting analysis. Project finance differs from traditional risk management strategies because it involves a change in organizational form rather than the use of financial instruments or derivatives. Risk management via organizational form is more appropriate for situations where the financial instruments do not exist or are expensive to purchase, or where the possibility of a total loss exists (i.e., the distribution of possible losses exhibits large negative skewness).

The risk management motivation is intriguing because it generates counter intuitive implications. For example, the benefits of diversification are a cornerstone of portfolio theory. But in the context of financing and investment decisions, financing assets jointly (corporate finance) permits risk contamination and is a necessary condition for debt overhang. Whereas diversification can be costly, specialization—financing assets separately through project companies—limits the amount of collateral damage a failing investment can impose on a sponsoring firm, and prevents sub-optimal investment strategies due to debt overhang.

In addition to describing each of the motivations, I illustrate them with specific examples and, where possible, with data from larger-sample research. Although I use the case studies to develop and illustrate the ideas in this paper, their consistency does not constitute a statistical test of validity. Instead, support for the ideas comes from a much larger set of detailed case studies (Esty, 2003a) as well as interviews with more than 150 practitioners including project finance bankers, advisors, lawyers, and corporate executives. Clinical research of this nature has proven useful in other settings such as Donaldson’s (1961) analysis of corporate debt policy and Sahlman’s (1990) analysis of venture capital organizations. It has also proven useful as a way to study contractual agreements [see Smith and Warner (1979) on bond covenants] and organizational structures [see Mian and Smith (1992) on accounts receivable management.]

The paper contains five sections. Section 2 defines project finance and describes the key structural features that characterize project companies. In addition to providing basic institutional details that are necessary to understand the arguments made later in the paper, this section provides new data on the structural attributes of project companies and contrasts them with what we observe in public corporations. I describe the agency cost and underinvestment motivations for using project finance in Sections 3 and 4, respectively, and conclude in Section 5. Throughout this paper, I refer to various Harvard Business School case studies by their shorthand names and an assigned case number [e.g., Mozal, case #C-7, instead of “Financing the Mozal Project”). The Appendix provides a list of these case studies and a brief description of each one.

Download
PDF Ebook The Economic Motivations for Using Project Finance