Portfolio Financings of Wind Farms
Developers of wind projects in the US have been moving to portfolio financings this year as a way to recoup development costs and refinance construction and mini-perm loans. Such portfolio financings present interesting risk-allocation issues.
A portfolio financing starts with a group of wind projects, each of which has its own power sales agreement and is either currently operating or very close to completion. The projects are bundled under a special-purpose holding company owned by the developer. The holding company issues bonds that are repaid by the revenue streams generated by the projects. If the bonds are not repaid, then the bondholders have the right to assume ownership of the projects.
The first financing of a portfolio of wind projects, a $380 million bond offering, closed earlier this year. The transaction closed approximately five months after the underwriters were selected. Other offerings are in the pipeline from developers who have a critical mass of wind projects either nearing completion or already in operation.
Developers use portfolio financings to receive, in a lump sum, the approximate present value of the revenues expected to be generated by the projects. The proceeds are used to recoup development costs or pay off project-level debt, thereby freeing up capital to develop still more projects. Often, portfolio bond proceeds come at a lower cost to the developer than if the developer borrowed money based on its corporate credit or against a single project. Bonds also typically have a longer tenor than bank loans and less restrictive covenant packages. Most importantly, portfolio bonds have a lower required debt service coverage ratio (as low as 1.3x) than project-specific debt (typically 1.5x to 1.75x), which significantly increases the amount that can be borrowed against a given revenue stream.
The structure of a portfolio financing is straightforward. Each of the projects included in the portfolio is owned by a limited liability company. The projects are self-contained, meaning that they have their own power sales, operations and maintenance and other contracts. It is helpful to the transaction if the projects are located in diverse geographic locations and the power sales agreements are with different purchasers.
The developer then creates a new limited liability holding company to own the individual project companies that own the wind farms. Each of the companies that owns a project guarantees the repayment of the bonds by the holding company, thereby ensuring that all of the projects’ revenues are available for debt service.
The holding company issues the bonds and enters into various agreements that govern the relationship between the holding company and the bondholders and provide security for repayment of the bonds. It also contracts with another entity, usually another subsidiary of the developer, to provide administrative services to the holding company as it has no employees to prepare financials or otherwise comply with its obligations under the financing documents.
The bonds will typically mature roughly at the same time as the scheduled termination of the longest-lasting power sales agreement in the portfolio. The repayment schedule for the bonds is structured to match the expected revenues and scheduled expirations of the various power sales agreements, so payments on the bonds, and distributions of remaining amounts to the developer, will be somewhat uneven.
The security package consists of pledges of the equity interests in the holding company and the project companies and in all of the assets that comprise the projects, including physical assets, contracts, permits and revenues from power sales. Documenting the security package can be time consuming given the number and diversity of assets being pledged and is a long lead-time item affecting the timetable for the financing.
Some of the proceeds of the bonds are used to fund reserve accounts. These typically consist of a debt service reserve account with a required balance equal to the next payment or two of principal and interest on the bonds and one or more other accounts that have negotiated balances and would be tapped if major maintenance is required on the projects (for example, in the event of a design defect in a specific model of turbine) or if operations and maintenance costs exceed budgeted amounts.
The bonds are repaid out of revenues from electricity sales and ongoing capital contributions that the developer must make to the holding company. The ongoing capital contributions are a percentage of “section 45 tax credits” to which the developer is entitled. The US government allows a tax credit of 1.8¢ a kWh for generating electricity from wind. The credits may be claimed for 10 years after each project is placed in service. The present value of the tax savings from the credits is worth about a third of the capital cost of a wind farm. The capital contributions must be made, or guaranteed, by a creditworthy entity and are made on a “hell-or-high-water” basis, meaning that the developer is not excused from making them even if, for example, the tax credit is repealed by Congress or the credits cannot be used because the developer lacks the tax base to use them fully. Depending on the price at which the projects sell electricity, the capital contributions tied to tax credits might represent 20% to 40% of the revenue stream supporting the bonds.
The terms of the bonds are negotiated with the underwriter. Wind consultants and independent engineers prepare pro-forma financials and, based largely upon potential deviations from base-case financials, reserve accounts and coverage ratios are negotiated. Ratings are obtained, bondholders are identified and the bonds are issued and sold.
Any power project brings with it many risks. Financing a project, or a portfolio of projects, is an exercise in risk allocation. One must first identify all the risks and then assign them to various participants in the transaction. John Maynard Keynes said that a banker is someone who lends you his umbrella and, at the first sign of rain, asks for it back. Banks avoid taking any risks for themselves; this is doubly true for institutional investors participating in bond offerings.
The developer wants to maximize the amount it receives from the sale of the bonds. The amount the developer receives is a function of the price and interest rate at which the bonds are offered, the bonds’ ratings, required coverage ratios and balances in reserve funds. All of these factors are influenced by risk allocation and, when the risks are allocated away from the bondholders, the factors swing in favor of more proceeds being paid to the developer.
Portfolios of wind projects have many of the same risks as more traditional power plants. For example, all power projects have risks associated with operations and maintenance procedures, the credit quality of offtakers, changing environmental and energy-related regulations and force majeure events. Wind portfolio financings address these risks in the same way as all other projects: proper operations and maintenance are assured by contracting with reputable companies and maintaining reserve funds; offtaker credit is reflected in the bond ratings and debt service coverage ratios; regulatory risks are analyzed and accepted, and insurance is purchased and contracts are written to protect against force majeure events.
However, wind projects also carry unique risks that are not present in other power projects. These risks include abnormal weather patterns, still-evolving turbine and blade technologies and potential changes to the tax code, renewable portfolio standards and other government programs. Also, with respect to portfolios in which some of the projects are not yet fully operational, a creditworthy entity must be ready to replace the lost revenues from power sales and payments associated with tax credits if the projects are not completed on schedule.
Wind is what runs wind farms just as fossil fuels are what power more traditional power plants. The developer of a wind farm takes the wind risk just as the developer of a more traditional power plant takes the risk that he will not be able to obtain enough fuel.
The wind industry has a challenge to make the investor community comfortable that wind risks are as manageable as risks with fossil fuels. Wind developers do this mainly by offering very conservative economic terms that ensure the bonds will be repaid even if the projects are afflicted with extraordinarily poor wind conditions.
A wind consultant projects long-term wind speeds, wind speed frequency distribution and seasonal variations. The data is then matched to the power sales agreements and incorporated into a model of the long-term energy output for each project and the portfolio as a whole. A financial base-case is developed based on conditions that the consultant opines are as likely as not to be experienced (the “P50” scenario). A worst-case scenario, which might have only a 5% chance of occurring (the “P95” scenario), is also modeled. The base case becomes the model that determines the aggregate amount of bonds that will be sold and the worst-case scenario guides the required balances for reserve funds and the conditions that must be met for distributions to be made out of the bond structure.
For example, the bonds might be structured so that they will still be paid if the projects operate at only 90% of projected levels, there is a 50% increase in operations and maintenance costs, or various low-wind scenarios occur in conjunction with, for example, increased major maintenance expenses. These considerations might require a relatively high average projected debt service coverage ratio — for example, 1.85x over the 20-year life of the bonds — compared to bonds supported by more traditional technologies. Coverage ratios can be expected to decrease over time as institutional investors become more experienced with such projects.
Another risk that bondholders will require the developer to address is the risk of pervasive defective technology (as opposed to discrete individual breakdowns) in the turbines and blades that transform wind into electricity. Wind turbines have gone through two generations of technology in the last 10 years. Just as with any new technology, there may be some wariness about whether it will suffer from teething problems.
Technology risk is mitigated in various ways by the developer: by grouping many technologies into one portfolio so that a design defect does not affect the economics of the entire portfolio, by obtaining an independent engineer’s assessment of the appropriateness of the various technologies to the corresponding sites and the adequacy of the developer’s arrangements for operations and maintenance of the projects, and by pushing as much of the risk as possible on a creditworthy vendor for the turbines.
A wind portfolio financing will be much more attractive to bondholders if the projects are spread across different geographic areas and there is a mix of equipment from different manufacturers. For example, the first successful windpower portfolio financing consisted of seven facilities in six states (in four wind “regions”) using five models of turbines from four manufacturers. The diversity of technology was commented upon favorably by the rating agencies and contributed to the investment-grade ratings for the bonds, which, in turn, expanded the pool of potential bondholders. Simple demand and supply suggest the more institutions willing to lend, the lower the cost of the money.
Developers also mitigate technology risk with contractual arrangements and reserve accounts. Wind turbines have been through at least two generations in the past decade; a number of models have been recalled or reengineered. This unsettled history results in investors demanding substantial major maintenance reserve accounts and provisions in the bond documents requiring redemption of at least a portion of the bonds upon a recall or other indication that faulty technology will prevent a project from contributing its share of the portfolio’s revenues.
Somewhat counterintuitively given their reputation as environmentally benign sources of electricity, wind projects also face the risk of regulatory change, especially in the environmental area. For example, several currently operating wind projects faced opposition to renewals of their permits due to concerns over the number of bird strikes at the facilities. Investors who are more accustomed to worrying about changing clean air regulations now must become comfortable with the practices and procedures of the Fish and Wildlife Service. In other words, the same regulatory risks are present in windpower as with more traditional power plants, just with different regulators.
Federal tax credits provide roughly a third of the capital cost of a wind project. Lenders are willing to take credit risks, but not tax risk. This means that a creditworthy user of the tax credits will have to agree to make ongoing capital contributions to the borrower so that the tax credits are part of the revenue stream that can be used to repay the bonds. These capital contributions must be made on a “hell-or-high-water” basis, meaning that if the wind blows, such contributions must be made. They must be made even if Congress changes the tax law to withdraw the credits, the projects do not qualify for credits because they failed to be placed in service by the deadline to qualify for credits, or the developer lacks the means to use the credits. They do not have to be made to the extent the wind fails. Wind risk is addressed in other ways. Tax credits are viewed in the same manner as revenues received from offtakers who buy the electricity. The credit rating of the developer is added to the mix with the credit ratings of the various offtakers. The bonds themselves will likely be rated at the lowest rating assigned to the developer and the offtakers.
Construction risk is not as great in wind projects as in more traditional power projects because the typical wind farm takes only six months to construct. However, one wrinkle in wind projects is that incomplete projects not only fail to generate power sales revenues, they fail to trigger the availability of the associated tax credits. If any of the projects in the portfolio is not yet in operation at the time the bonds are sold, the developer must agree to replace both the lost power revenues and the lost income based on tax credits if the project is not completed on schedule. For example, in one portfolio financing this year, the bonds were sold a few months prior to scheduled completion of two of the seven projects. The developer entered into a construction completion agreement with the holding company that issued the bonds. The agreement provided that the developer would make payments to the holding company that would replace the lost revenues and tax credit payments if the projects were not completed on schedule. In a worst-case scenario, meaning the projects failed key tests by certain dates, the developer was required to make lump-sum payments to the holding company that would then be used to redeem a portion of the bonds.