Understanding Risk and Preparing the RFP

Transcription

1 Understanding Risk and Preparing the RFP Table of Contents Excerpts from Risk Allocation Primer... A Sample term sheet... B Description of disasters... C

2 Excerpts from Risk Allocation Primer Please note that you can access the complete document via the BRC website at marketplace.businessrenewables.org. A 1

3 Organization of the Primer A 2

4 Price Risk This page covers two questions: What types of price risk are inherent to a power purchase agreement? How could price risk be allocated between the corporate buyer and the project developer? What are the types of price risk? As described in the Deal Structure Primer, a virtual PPA typically involves two different cash flows related to the electricity generated by the wind or solar plant: The corporate buyer pays a fixed price per megawatt-hour of electricity generated by the plant. The fixed price is stated in the PPA itself and can be subject to an annual escalation or other adjustment, but it is fixed in the sense that it is stipulated by the PPA. The corporate buyer receives a floating price per megawatt-hour of electricity generated by the plant. This floating price is determined by the real time wholesale electricity market in which the plant operates. The price risk of a PPA is a reflection of this latter cash flow: the price received by the corporate buyer will vary in each settlement interval and may move higher or lower over the life of the PPA. If the market price floats to a level higher than the fixed PPA price, then the PPA is in the money for that interval by the magnitude of the difference. Conversely, if the market price falls below the fixed PPA price, then the PPA is out of the money for that interval by the magnitude of the difference. Which is more likely? The answer depends on several factors and will vary geographically. In most parts of the U.S., wholesale electricity market prices are determined largely by natural gas prices because natural gas-fired generators are usually the marginal generators. U.S. natural gas prices are at very low levels due to the shale boom of the last decade, and they are particularly low in areas with substantial A 3

5 natural gas production relative to transportation capacity (e.g., Pennsylvania). Natural gas prices are also highly volatile. In the last decade, natural gas prices have seen a slide of nearly 52% in five months (8/31/05 to 1/31/06), a spike of 144% in ten months (7/31/07 to 5/31/08), and a separate fall of nearly 78% over the subsequent 14 months (5/31/08 to 7/31/09). Any upward movement in natural gas prices over the lifetime of a PPA therefore could push power prices higher. Conversely, if natural gas prices fall even further, if renewables reach significantly higher levels of grid penetration, or if demand for electricity falls, there would be downward pressure on wholesale power prices. The volatility itself is an issue for corporate buyers to consider carefully. Given the price swings noted above, corporate buyers must be prepared for the possibility that the profit & loss (P&L) on the PPA could vary significantly from month to month potentially far more than the variability in retail electricity bills to which the corporate buyer may be accustomed. Under a physical PPA, the corporate buyer may still face price risk if it sells the electricity into the market at a floating price. However, if the practice is to arrange for a contract path in essence, financially mimic the delivery of electrons from the wind or solar plant to the buyer s facilities then the corporate buyer s exposure to market prices may be mitigated. How can price risk be allocated? The table below describes three potential allocations of price risk. [See following page.} A 4

6 Risk allocation 100% to buyer Shared Risk 100% to developer Description Off-taker has right to revenues from sales into wholesale market, whatever the price Wholesale market prices received by off-taker are subject to a floor or collar, or they are pegged to a published market price Developer has no fixed revenue stream from offtaker, meaning it in effect remains a merchant generator Implications This allocation will introduce the least amount of finance risk for the transaction (all else being equal) This allocation may make it more difficult for the developer to obtain financing Developer s primary motivation to strike a deal with a corporate buyer is removed Frequency 67% 33% 0%

7 Basis Risk This page covers two questions: What types of price risk are inherent to a power purchase agreement? How could price risk be allocated between the corporate buyer and the project developer? What are the types of price risk? As described in the Deal Structure Primer, a virtual PPA typically involves two different cash flows related to the electricity generated by the wind or solar plant: The corporate buyer pays a fixed price per megawatt-hour of electricity generated by the plant. The fixed price is stated in the PPA itself and can be subject to an annual escalation or other adjustment, but it is fixed in the sense that it is stipulated by the PPA. The corporate buyer receives a floating price per megawatt-hour of electricity generated by the plant. This floating price is determined by the real time wholesale electricity market in which the plant operates. The price risk of a PPA is a reflection of this latter cash flow: the price received by the corporate buyer will vary in each settlement interval and may move higher or lower over the life of the PPA. If the market price floats to a level higher than the fixed PPA price, then the PPA is in the money for that interval by the magnitude of the difference. Conversely, if the market price falls below the fixed PPA price, then the PPA is out of the money for that interval by the magnitude of the difference. Which is more likely? The answer depends on several factors and will vary geographically. In most parts of the U.S., wholesale electricity market prices are determined largely by natural gas prices because natural gas-fired generators are usually the marginal generators. U.S. natural gas prices are at very low levels due to the shale boom of the last decade, and they are particularly low in areas with substantial A 6

8 natural gas production relative to transportation capacity (e.g., Pennsylvania). Natural gas prices are also highly volatile. In the last decade, natural gas prices have seen a slide of nearly 52% in five months (8/31/05 to 1/31/06), a spike of 144% in ten months (7/31/07 to 5/31/08), and a separate fall of nearly 78% over the subsequent 14 months (5/31/08 to 7/31/09). Any upward movement in natural gas prices over the lifetime of a PPA therefore could push power prices higher. Conversely, if natural gas prices fall even further, if renewables reach significantly higher levels of grid penetration, or if demand for electricity falls, there would be downward pressure on wholesale power prices. The volatility itself is an issue for corporate buyers to consider carefully. Given the price swings noted above, corporate buyers must be prepared for the possibility that the profit & loss (P&L) on the PPA could vary significantly from month to month potentially far more than the variability in retail electricity bills to which the corporate buyer may be accustomed. Under a physical PPA, the corporate buyer may still face price risk if it sells the electricity into the market at a floating price. However, if the practice is to arrange for a contract path in essence, financially mimic the delivery of electrons from the wind or solar plant to the buyer s facilities then the corporate buyer s exposure to market prices may be mitigated. How can price risk be allocated? The table below describes three potential allocations of price risk. [See following page.] A 7

9 Risk allocation 100% to buyer Shared Risk 100% to developer Description Off-taker has right to revenues from sales into wholesale market, whatever the price Wholesale market prices received by off-taker are subject to a floor or collar, or they are pegged to a published market price Developer has no fixed revenue stream from offtaker, meaning it in effect remains a merchant generator Implications This allocation will introduce the least amount of finance risk for the transaction (all else being equal) This allocation may make it more difficult for the developer to obtain financing Developer s primary motivation to strike a deal with a corporate buyer is removed Frequency 67% 33% 0%

10 Congestion Risk This page covers two questions: What is congestion risk? How could congestion risk be allocated between the corporate buyer and the project developer? What is congestion risk? Congestion risk (and curtailment risk, which is related) arises due to the physical limitations of the grid itself and those limitations impacts on supply and demand (and thus price) for electricity at different grid locations. It is not unique to wind or solar projects: any generating technology, whether renewable or conventional, faces some level of congestion risk if it supplies electricity to the grid. To understand congestion risk, consider a simple grid consisting of two locations with one transmission line between them: In this scenario, all of the plants at Location A will generate an aggregate 500 MW, which the transmission line will carry to the customers at Location B. Now imagine two things: Location B adds more customers, who now demand a total of 600 MW, and you want to build a new power plant that can reliably generate 100 MW to serve the new customers. The question is where you should locate the project. It should be obvious that if you locate your plant at Location A, you will have a problem, because the transmission line is already at capacity and thus will not be able to carry any more power to Location B. The addition of your plant at Location A will mean that Location A has a total of 600 MW of potential supply, but due to congestion on the transmission line A 9

11 the maximum demand that the generators can serve is only 500 MW. With greater supply relative to demand at Location A, the value of supply (and thus the price received by generators) at Location A falls. This is congestion risk: the risk that technical limits in certain points on the grid (in our example, the transmission line) will depress the market price of electricity where your wind or solar plant will generate, therefore resulting in less revenue from the PPA than you might otherwise have expected. In extreme cases, market prices can be depressed so much that they fall below zero, meaning generators must pay in order to generate. In some cases, these technical limits can cause the grid operator to force some generators to shut down even a wind plant that can generate at essentially no marginal cost could be forced to stop generating for some period of time while the wind is blowing. This is known as curtailment and is the subject of the next page in this primer. The important risk allocation issue related to congestion risk is whether the corporate buyer is obligated to purchase electricity from the project when the wholesale market price of electricity falls below zero or some other specified level. How could congestion risk be allocated? The following table summarizes potential allocations of congestion risk. The frequencies indicated are based on information collected by the BRC from parties to transactions in 2015 or (YTD) [See following page.] A 10

12 Risk allocation 100% to buyer 100% to developer Description Rationale Buyer must procure electricity from the project, no matter the applicable wholesale market price Seller cannot control congestion Buyer must procure electricity from the project, but only if the wholesale market price is above some minimum level The risk is not known to or controlled by either party, and it is pareto optimal to share it in a way that facilitates financing but limit s buyer s risk. Implications Buyer s P&L on the PPA could be significantly negative over some time periods if the plant experiences low (or even negative) prices Provides some protection to buyer if the plant experiences low prices Frequency 25% 75% Note: Of the 75% of deals that shared the risk, 25% used the negative PTC value ( $23), 17% used the negative PTC value grossed up for tax, 25% used zero, and 8% used some other value/mechanism.

13 Curtailment Risk This page covers two questions: What is curtailment risk? How could curtailment risk be allocated between the corporate buyer and the project developer? What is curtailment risk? Like congestion risk, curtailment risk arises due to the need to keep electricity supply and demand in balance and due to the physical limitations of the grid itself. Consider our simplified example (from the congestion risk page) of a grid consisting of two locations connected by a single transmission line. Now imagine that all of the generation at Location A is wind powered, and that the amount of demand at Location B is only 400 MW. On a windy day, the wind plants would be able to generate their maximum capacity of 500 MW. The problem is that the demand is only 400 MW, and the grid operator must keep electricity supply and demand in balance at all times (this is a rule of physics that may be avoided only by adding energy storage to our grid). The grid operator cannot add demand at either location, and all generators have offered the same price so a market price signal is ineffective; therefore, the grid operator s only option is to force all the wind generators at Location A to generate 25% less. This forced reduction in generation is known as curtailment. It can arise because of imbalances in supply and demand, as described in the above example, or due to transmission constraints (similar to congestion). As the previous example shows, curtailment risk is a function of the location of the wind plant, the amount of generation, the amount of A 12

14 electricity demand, and the grid s physical limitations. The impact is that the plant will in practice generate less electricity (and thus fewer RECs) than one would expect based on the amount of wind resource alone. In a typical PPA the buyer pays a unit price for each megawatt-hour of electricity generated, so curtailment results in less revenue from sales into the wholesale market but also less expense incurred via payment of the PPA price. In these contracts, curtailment does not represent a financial risk to the buyer, but it could result in the buyer receiving fewer RECs than it was expecting. However, some PPAs include take or pay provisions by which the buyer pays the unit price for electricity and pays for available capacity, which refers to energy that the project would have delivered if it had not been curtailed (i.e., during periods with sufficient wind resource to generate electricity). The take or pay provision may specify that the buyer is required to pay regardless of the reason for the curtailment, or it may exempt curtailment imposed by the grid operator or non-generation because of a force majeure event. Curtailment is a more serious risk for the project developer. If the plant generates less electricity (and thus less revenue from the PPA) than expected, the project developer will have less cash to provide its expected return on equity or, in more extreme situations, to cover debt service. For this reason, developers financiers would prefer that the PPA contain a take or pay clause, so that the plant would continue generating revenue even if curtailed. Alternatively, developers financiers will rely on third party experts to assess the probability of curtailment and may reduce the debt allowed on the project if curtailment is assessed as a risk, thereby increasing the cost of the project. The important risk allocation issues related to curtailment risk are: What would be the harm to the buyer if the project were to produce fewer RECs than expected? How much curtailment could the plant experience before the project developer s special purpose vehicle would risk a default on its financing? A 13

15 How could curtailment risk be allocated? The following table summarizes potential allocations of curtailment risk. The frequencies indicated are based on information collected by the BRC from parties to transactions in 2015 or (YTD) [See following page.] A 14

16 Risk allocation 100% to buyer Shared risk 100% to developer Description Upon curtailment buyer has to pay seller for deemed generation plus a PTC grossup if applicable Pick XXX hours per year of curtailment, below which the seller bears the risk and above which the buyer bears it or the parties share, etc. Upon curtailment the seller gets no payment under the PPA (and may owe the buyer replacement RECs equal to deemed generation) Rationale Seller cannot control curtailment, and seller s financing parties will not bear the risk of curtailment, so buyer must pay even during curtailment periods The risk is not known to or controlled by either party, and it is pareto optimal to share it in a way that facilitates financing but limit s buyer s risk in most situations. Buyer bargained for a certain amount of generation (and/or RECs) in each year during the term; buyer is not best positioned to assess curtailment risk so should not bear it Implications As curtailment increases over the agreement term, buyer may be paying and getting no benefit Both buyer and seller will assume the worst case in modeling, but the case is at least quantifiable and bounded Seller s financiers will have to study and assess curtailment risk for the project and may charge for this, which will be reflected in the price offered by seller Frequency 11% 11% 78%

17 Credit Risk from the Buyer s Perspective This page covers two questions: What is the credit risk from the perspective of the corporate buyer? How could this risk be allocated between the corporate buyer and the project developer? What is the credit risk? The corporate buyer must consider the creditworthiness of its counterparty to the PPA and evaluate the risk that the latter could become insolvent or otherwise default on its obligations under the contract. In addition, some wind or solar plants have multiple corporate buyers; in such cases, each buyer may consider it important to evaluate the creditworthiness of the others and the potential impact of a default by one of the other buyers. In many cases, the project developer will form a special purpose vehicle (SPV) a separate legal entity in the project developer s existing corporate structure that will sign the PPA with the corporate buyer and obtain financing for the development of the project. The SPV often will have minimal assets or operations beyond those required for the specific project. In this situation, the corporate buyer may be concerned about the risk that the SPV could default on its obligations and therefore may request that the project developer provide credit support (e.g., a letter of credit or guarantee) for the SPV. This credit support could relate to the period during which the project is under construction, the period after the plant has begun operating, or both. Separately, corporate buyers that contract for less than all of the output of a plant should consider the risk that another buyer of the plant s output (if there is one) could default on its obligations under its own PPA with the SPV. An important legal issue in such situations is whether the corporate buyer s potential liability is joint and several with A 16

18 the other buyer, meaning that the corporate buyer could be held responsible to cover a shortfall caused by non-payment by the other buyer. Joint and several liability is, of course, preferable to the project developer, who otherwise may pursue only the non-performing buyer, without recourse to the corporate buyer that is current in its obligations. Similarly, in cases where a portion of the plant s output is not subject to a PPA and instead is left merchant, this introduces market risk to the seller and therefore credit risk to the buyer. How could credit risk be allocated? The following table summarizes potential allocations of credit risk. The frequencies indicated are based on information collected by the BRC from parties to transactions in 2015 or (YTD) [See following page.] A 17

19 Method 100% to buyer Shared risk 100% to developer Description Buyer s counterparty is a single purpose LLC; if the LLC defaults buyer has no recourse to any security instrument of parent Seller provides a parent guaranty or letter of credit but there is some cap on the exposure under these instruments Seller provides an unlimited parent guaranty or a fully replenishable letter of credit to backstop the project LLC s obligations in the agreement Rationale Hard to justify Seller s parent will backstop the LLC s obligations through a PG or LOC, but these need to have a reasonable cap on liability and such a cap allows seller to provide buyer with more reasonable pricing Seller should not escape liability simply because it is a shell entity Implications In a default buyer will not get the benefit of its bargain; seller is judgment proof A reasonable balance of risk and price, but risk to buyer in an outside case Buyer is fully protected but the cost may be prohibitive for some developers Frequency 17% 83% 0% Note: In the 83% of cases where the seller s parent provides credit support, all provided it during construction, but only 67% also provided it after the plant has begun operating. Of that 67% that had credit support both during construction and after operation began, 42% provided a higher level of credit support after operation began compared to during construction, while 25% provided the same level of credit support during both periods.

20 Proposed Term Sheet Wind project PPA counterparty Year 1 PPA price Escalator PPA term PPA structure Windy Fields wind farm, located in western Texas (ERCOT) Windy Fields LLC, a special purpose vehicle formed Big Developer Corp. for the development of the Wind Fields wind farm $20 per MWh that either (i) is actually generated by Wind Fields or (ii) would have been generated if not fo curtailment by ERCOT Two percent 20 years Contract for differences Expected COD December 2018 Settlement Bus bar B 1

21 Disasters Any of the following scenarios could happen in a real PPA. If you structure a deal around the proposed term sheet, what would happen in each scenario? Scenario A: Market Price Disaster Natural gas prices fall to an historical low of $1/mmBTU. Lower natural gas prices cause wholesale electricity market prices to fall even lower than they are now and remain there for the life of your PPA. Scenario B: Basis Disaster Energy prices for your facilities continue to rise due to (i) increased need for air conditioning during daytime hours and (ii) your utility provider increasing daytime retail electricity rates due to its rising costs. However, at the same time, your wind farm increasingly generates during overnight hours, when wholesale electricity prices are lower, and generates less than expected during the daytime. Scenario C: Curtailment / Congestion Disaster In addition to your own wind farm, many other wind farms are built nearby, but investments in transmission infrastructure do not keep up. Wind so much wind generation in the area, wholesale market prices at that location fall and at times there is so much wind generation that ERCOT is forced to order your farm to stop generating. Scenario D: Default Disaster Windy Fields, LLC encounters problems developing the wind farm. Its costs rise so much that Windy Fields, LLC defaults on its financing arrangements and goes bankrupt. C 1