Stay current on your favorite topics
Project bond financing is undoubtedly complex. The process of setting a bond’s structure, terms and covenants may involve many groups and specialties, including investors, government officials, regulators, construction firms, banks, capital markets specialists and, of course, lawyers. Such complicated negotiations may give issuers and investors pause. But much of the time, their concerns revolve around one of four “myths” which, with the right expertise, can be overcome.
Myth 1: Completion risk is a deal killer
What if the money is spent but the project is never completed, and thus never generates cash flow to service the debt? This is called completion risk.
Historically, bond investors have been less likely than commercial bank lenders to assume completion risk. This reluctance on the part of bondholders can be attributed to two factors. First, there’s the risk (shared by all creditors) that an incomplete project will not be able to generate the cash flow necessary to repay its debt.
Second, there’s a question of whether bondholders have the ability to effectively evaluate completion risk or have the technical or legal staff to monitor the project.
Such concerns are understandable, but not insurmountable. Bondholders accept completion risk under several different circumstances. Some have tolerated completion risk in certain sectors where that risk is customary, for example, transportation infrastructure deals, power plants or casino financings. In other instances, an attractive interest rate—particularly in a hot bond market—has been enough to persuade bondholders to shoulder completion risk. Some investors have bought project bonds when issuers have offered traditional mitigants and credit enhancements, such as cost-overrun undertakings, parent completion guarantees, sponsor equity support agreements and subordinated debt tranches. Lastly, investors have taken on completion risk when project documentation has mitigated construction risk, either through turnkey construction contracts or enhanced liquidated damage provisions.
of regional GDP: the estimated cost of annual infrastructure needs in Africa and South Asia.
of global GDP: the estimated global cost of annual infrastructure investment and maintenance needs.
HOW ONE PROJECT BOND MANAGED COMPLETION RISK
In 2010, the US$1.5 billion Odebrecht Norbe VIII/IX project bond offering was going to finance the construction of two ultra-deepwater drillships for oil exploration off the coast of Brazil. The parent company, Odebrecht Oil & Gas, agreed to provide support in the form of additional equity or subordinated debt (capped at a percentage of total project costs) to each project company in the event of cost overruns before completion of the drillships.
Myth 2: Getting consents and waivers is impossible
There is a difference in the relationship between a project company and its creditors when a project bond is involved than when a loan is used. Because bonds are transferable, for instance, a bond issuer may not know who ultimately owns its bonds at any given time. As one might imagine, this can make it more difficult to obtain consents and waivers from bondholders than it would be from a long-term relationship lender. Adding to this complication, bondholders may not be as active as commercial lenders in monitoring the project, may need more time to get up to speed on a consent or waiver issue and often charge a fee for consents and waivers.
However, most bonds, including project bonds, are designed to accommodate this different relationship. The covenants for a project bond are usually designed to allow the project to operate with more flexibility expressly because potential bondholders may be more numerous and diffuse, potentially have less project finance expertise, and may not have an active relationship with the project. For example, the covenants are often designed to include objective, rather than subjective, tests to reduce the need for consents and waivers for routine operational matters.
In addition to covenant flexibility, many project bonds also have some type of proxy mechanisms to reduce the need for bondholder consent. These proxy mechanisms include enlisting other creditors to make certain determinations or hiring technical consultants or other third-party experts to make technical determinations. Some project bonds have been structured so that bondholders will be deemed to have consented to an action if a rating agency reaffirms the project’s bond rating, although this approach has fallen out of favor since the financial crisis. Last, some projects have common covenants with bank facilities and oftentimes the bondholders are effectively disenfranchised because they are the minority creditor.
FROM MINORITY TO MAJORITY CREDITOR
In project bonds where there are common covenants, whoever owns the majority amount of debt may change over time. Bondholders are often the minority creditors early in the life of a project. However, there is often an inflection point after the passage of time where bondholders become the majority creditors because the project bonds may have a longer tenor than the bank loans and may amortize more slowly (or not at all). Some intercreditor structures ignore this change and other deals have detailed provisions to address this dynamic.
Myth 3: Project bonds need an investment-grade rating
It is true that certain investors have restrictions on investing in noninvestment- grade bonds, which means that it can be more difficult to market high-yield project bonds. Additionally, the higher interest rate of a non-investment-grade bond may not be appealing to project sponsors or competitive with other sources of financing. Despite these liquidity and pricing challenges, quite a number of high-yield project bonds have been successfully issued, especially through the private placement market.
One of the challenges for project bonds in emerging markets is the rating ceiling effectively created by a country’s sovereign rating. This challenge is not unique to project bonds, but also exists for corporate bonds. In theory, if the economics of the project are not exposed to the same risks as the sovereign, it should be easier to decouple the two ratings. But, in practice, this is often difficult. As a result, a number of the traditional methods for improving the rating of a borderline project, such as additional credit support, may not work as well in an emerging-markets deal. Conversely, project bonds that have direct or indirect sovereign support from investment-grade sovereigns often benefit from a “notching up” of ratings. This notching up often occurs with Middle Eastern project bonds issued by projects located in the Gulf States.
Interestingly, we have found that the covenants for investment grade project bonds are sometimes actually tighter than those for high yield project bonds. This can occur if the investment-grade project bond shares covenants with a bank loan, because bank covenants tend to be stricter. Most high-yield project bonds follow the looser, incurrence based covenant model of corporate high-yield bonds. The practical effect is that a project with high-yield covenants may have more latitude to sell assets, make dividends (restricted payments) and have its minority shareholders transfer equity interests. As the project bond market matures, we expect more convergence among the covenants and fewer anomalies.
Myth 4: ECAs and bondholders have trouble coexisting
When the ability of commercial banks to provide funding for projects became limited after the recent global financial crisis, export-credit agencies (ECAs) helped provide capital for a number of projects. ECA financings have typically provided favorable interest rates in exchange for relatively conservative covenant packages. Consequently, a number of projects that are looking to refinance their existing bank loan debt with project bonds often want to leave their cheap ECA financing in place. This scenario presents a challenge because a project may end up with a capital structure composed of ECA lenders who have lent for policy-specific reasons bondholders. In an enforcement scenario, ECAs may want to act according to their policy objectives while mark-to-market bondholders may look to maximize their recovery. Because of these different potential objectives, some project bond participants believe that ECAs do not want to co-lend into financings with bondholders.
It would be a gross overstatement to say that ECAs are anti-capital markets, especially given that a number of ECAs act as guarantors for covered bonds. However, there is a broad spectrum of opinions by the ECAs about project bonds. While some ECAs maintain control, or “golden vetoes,” over common covenants and intercreditor matters, many are willing to allow bonds into a capital structure and treat bondholders much as they do commercial lenders. Many ECAs over the last few years have come to recognize that project bonds can constructively provide an alternative source of capital to commercial lenders and help bridge funding gaps.
trillion Additional sum needed to upgrade and expand ports and airports from 2010 to 2030.
trillion OECD estimate of necessary infrastructure spending from 2010 to 2030.
HOW ONE PROJECT ACCOMMODATED BOTH ECA FINANCING AND PROJECT BONDS
In 2013, the Shuweihat 2 Independent Water and Power Project in Abu Dhabi was looking to partially refinance the loans it had incurred at the height of the credit crunch. An US$825 million project bond was issued and put in place alongside the project’s commercial loans and an ECA loan from the Japan Bank for International Cooperation (JBIC). This was one of the first project bonds to exist alongside a JBIC loan. To allow the bond into the capital structure, JBIC retained a golden veto, meaning that it controlled intercreditor decisions for waivers and enforcement for as long as its loan remained outstanding.
Not a Myth: Project bond financing is growing
The number of developments financed through project bonds—and the dollar amounts involved—continue to gain traction. In 2014, 67 projects used project bonds to help finance existing projects, issuing more than US$34 billion, according to Infrastructure Journal and Project Finance. In addition, InfraDeals reports that nearly US$14.8 billion of project bonds were issued to help finance 38 greenfield projects worldwide in 2014. European officials have launched the Project Bond Credit Enhancement structure. Under it, the European Investment Bank (EIB) provides participating developers with subordinated debt, which should enhance the credit quality of senior project bonds and make the bonds more palatable to institutional investors.
To be sure, the complexities and risks of project bonds can give pause even to the most experienced of dealmakers and investors. But the common perception that some project bond issues are insurmountable is a myth, and as the number of successful project bond deals increase over time, this myth will continue to fade away.
possible maturity of
PROJECT BOND PRIMER
Project bonds finance a variety of investments in transportation, communications, energy and other infrastructure. The bond’s proceeds are used to provide limited recourse, or nonrecourse, financing to a single-purpose vehicle that owns, develops and builds a project. Buyers include insurance companies, pension funds and other long-term investors.
Project bonds are appealing because they can match long term liabilities to long-term cash flows from projects (maturities may extend 20 years or more). The bonds themselves often offer stable returns at higher rates than similarly structured sovereign debt. They usually have low volatility and little correlation with other asset classes. They have flexible financing structures and can be adapted to include multiple assets and projects.
Besides the general capital market risks associated with bonds, including interest rate risk and market execution risk, project bonds also have standard project finance risks, including construction risk and political risk, and may have unique risks such as negative carry.
Click here to download PDF.
This publication is provided for your convenience and does not constitute legal advice. This publication is protected by copyright.
© 2015 White & Case LLP