At Infocast's Community Choice Energy Summit, esteemed speakers discussed financing projects with CCA contracts. The panelists were David McNeil, finance manager at Marin Clean Energy, Siobhan Doherty, director of power resources at Peninsula Clean Energy, Elizabeth Waters, a managing director and deal team leader at MUFG, Magali Cohen, a director with the power and infrastructure group at Investec, Sondra Martinez, a senior director on the originations team at NORD/LB, and Vince Plaxico, director of project finance with Recurrent Energy. The moderator was Deanne Barrow with Norton Rose Fulbright in Washington. Below is an edited transcription of the panel.
Deanne Barrow: Let’s start by getting a sense of the market. Vince Plaxico, apart from the Mustang deal, has Recurrent signed power purchase agreements with any other community choice aggregators?
Vince Plaxico: Yes. We have a new PPA with Peninsula Clean Energy for a project that is expected to go into service in 2019. We see CCAs as an important market for our growth in California. Recurrent has a competitive advantage because we are the only solar developer that has fully financed a project where the entire output was committed to a CCA. It takes a lot of time to work through the issues in these deals. On Mustang, we were able to achieve that with a lot of information sharing and the helpful attitudes of our lenders and our partners, Marin Clean Energy and Sonoma Clean Power.
Barrow: The Mustang deal was financed with a mixture of back-levered debt and tax equity. What’s the plan for your new project with Peninsula Clean Energy in terms of financing?
Plaxico: The delivery is in 2019, so we have more development work to do on that project before we start focusing on the financing, but I expect it will be back-levered debt and tax equity.
Barrow: Sondra Martinez and Magali Cohen, has NORD/LB or Investec done any other project financings of CCA projects, other than Mustang?
Elizabeth Waters: To my knowledge, Mustang is the only transaction that has term debt on it. I know there is at least one other in the market that is currently getting done, but there are a couple other clients that have deals coming with whom we are in active discussions about financing that are pure CCA offtakes.
Magali Cohen: We have closed the Mustang transaction so far, and we are actively talking to several counterparties and looking at doing additional deals.
Sondra Martinez: There are several transactions where CCAs are part of a portfolio with other, non-CCA projects, and I think that is what Beth was alluding to. Those look different in the way we bankers think about them. In terms of a pure financing of a CCA, to my knowledge, there is just Mustang.
Barrow: Beth Waters, can you speak to the portfolio structure?
Waters: Sure. This is very common, not just for CCA projects, but also whenever a developer has an assortment of assets with different credit profiles. In the early days of solar, the individual projects were not big enough for banks of our size, so the developers presented portfolios of projects. The particular client I have in mind had a number of solar projects, and one of them happened to be with a CCA as the offtaker. When there is only one CCA in a portfolio, we run a sensitivity analysis. The CCA represented only 10% of the projected cash flow. You ask what happens if none of the CCA revenue is there. Can the developer still repay the debt. In this case, the answer was yes. It was toe-dipping exercise, meaning a way to get used to CCAs. A standalone large project with all the output going to a CCA requires a different analysis.
Barrow: So portfolios are good toe-dipping exercises. Baby steps.
Waters: It is like what is happening currently with energy storage. You are only seeing energy storage in current project financings as part of other projects rather than as standalone assets.
Barrow: David McNeil and Siobhan Doherty, give us a sense of the market from your perspective. How many megawatts do you have under contract, for what kinds of resources, and also give us a sense of the tenors of the PPAs that you are signing.
Siobhan Doherty: Sure. Peninsula Clean Energy has been operating for about a year now and, in the past year, we have signed nine PPAs for a total of 550 megawatts, and there is a big range in the tenors of the contracts. We have PPAs with a term of one year and all the way up to 25 years. We have signed solar, wind and small hydro PPAs, with sizes that range from two megawatts to 200 megawatts. We are working to build up a diverse portfolio.
David McNeil: Marin Clean Energy has about $2 billion worth of energy under contract right now. Of that, about $1.8 billion is renewable, long-term PPAs. Those range in terms from 10 to about 25 years. Large wind and solar PPA terms vary from 12 to 20 years. In 2016, Marin Clean Energy entered into long-term, renewable PPAs with contract values totaling more than $800 million.
Barrow: Siobhan Doherty, you gave quite a range. Do you have a sweet spot in terms of size, length and also resource? There are a lot of developers in the room.
Doherty: We do not have a sweet spot. We want to build a diverse portfolio. We have a goal of 100% renewables by 2025, and so we are looking at ways to fulfill our load shape with different sources of renewable energy. Our board has asked us to look at a variety of term lengths, and a variety of resources, to help us reach our goal. We are going through an exercise right now of looking at where there are holes in our portfolio. We will be putting out a request for offers early next year in an effort to fill in holes.
McNeil: The same thing is pretty much true for Marin Clean Energy. We have a lot of solar in our portfolio today. We are focused on filling in high-demand hours, so we will be looking at wind projects and possibly storage in our next open season.
Martinez: From a financing perspective, when we were looking to finance the Mustang solar project, what they are describing was really important to the banks because they should be managing their portfolios like a utility thinks about managing its portfolio. It is a good thing that the CCAs are looking for a variety of short- and longer-term contracts and for diversification in terms of assets. That is something that we spend a lot of time getting information on and understanding how they operate. The creditworthiness of the CCA is our risk as a lender. If a CCA is poor at managing its portfolio, then it could lead eventually to operating losses.
Plaxico: Taking that a step further, from the developer’s perspective, we are bidding for PPAs with the CCAs. We look at all their publicly-available information, their business plans, what does it look like in five years, what type of reserves will they have built up in that time, and what risk-mitigation strategies do they have in place? We do all that at the front end before we even approach any of the CCAs, as that will put us ultimately in a better position to deliver what we promised.
Barrow: Is it hard as a developer to monetize a shorter offtake contract and, for that reason, are you looking solely for long offtakes?
Plaxico: We have done short-term offtakes with Marin and in some other places. We are flexible, but in general, you need a longer-term PPA in order to support the initial capital cost of the project.
Waters: The bankers had to deal with an influx of corporate PPAs before we started to see PPAs with CCAs. I have felt as a banker that there has been a communications gap. I understand why it is important for CCAs to have a mix of contracts in their portfolios, but a developer will not be able to finance a new project without a long-term contract.
Martinez: Or the developer will have a large bullet payment to make on the loan at the end of the short-term contract and a merchant tail that may not be possible to refinance.
Pricing and other considerations
Barrow: Let’s get into pricing. What were the margins like in the Mustang transaction? Was it more like a typical utility deal or a merchant deal?
Plaxico: Mustang was a first-of-its-kind deal. Depending on the sponsor, project characteristics and other factors, the spread for future CCA deals should be around 50 to 75 basis points above a plain-vanilla deal.
Cohen: It was the first deal with CCA-only offtakers, which required a significant spread premium at the time. The market dynamics have changed since then. There is a lot of liquidity in the bank market right now. It is definitely a sponsor’s market. That will work in Recurrent’s favor in the next deal.
Martinez: We saw the same thing in the merchant gas space. It is really important to get a number of lenders across the line on the first transaction. From NORD/LB’s perspective, we do not consider every community choice aggregator as equivalent. We got very comfortable with Marin Clean Energy and Sonoma by doing a deep-dive analysis, much like how Vince Plaxico mentioned he does as a developer, so I am not sure pricing would be the same for every community choice aggregator. It is a sponsor’s market, but at the end of the day, it will be about execution, and Vince probably knows this better than anybody that these deals can be difficult for banks to execute. Some banks will remain completely out of the market because they will not look at a project that has an offtaker without a third-party credit rating, so you already have a limited universe of banks that can do this type of deal.
Plaxico: The key is information. We are all working together to lower the prices that we are able to offer CCAs. A part of the electricity price is the cost of debt. The more information the lenders have that they can bring to their credit committee, the more likely they are to get approval. The way that Marin, Sonoma and Peninsula share information online makes it easier for us to track their balance sheets, what the last board meeting was about, and so on. Transparency is really important. We spend a lot more time than you think looking through those documents.
Martinez: We probably started talking about the financing and sharing information in February, and then really started to nail down terms in April, and the deal closed in July. We had our credit people involved in the discussions, in order to understand the structure, as early as February to make sure people are getting comfortable and ensure that we can execute on the transaction.
Cohen: Information sharing is critical since, as Sondra mentioned, CCAs have different characteristics, so it will be critical for the next deals to be able to analyze the specific credit profile of the offtaker.
McNeil: My suggestion for other CCAs is to build your credit package. At Marin Clean Energy, we put together a data room that contained all sorts of pertinent information — the joint powers agreement, all the government documents, all the financial statements — and then constructed a risk profile that can be read and valued by the credit committees and investors who are involved in the process. The easier you can make their jobs, and the clearer the story you can tell about your CCA, the better the odds you will be able to get your deals financed.
Martinez: It was really helpful that both Marin and Sonoma not only were easy to get on the phone with us, but they also answered written questions. There was a nice working relationship.
McNeil: The better your credit pack, the less time you will have to spend on the phone answering questions.
Waters: As was already mentioned, there is a lot of liquidity in the market, which will make lenders more aggressive on pricing. The more lenders there are chasing deals, the more aggressive each bank will have to be to get a piece of any one transaction.
Plaxico: Great! [Laughter]
Barrow: Siobhan Doherty, tell us more about information sharing. One thing that makes CCAs different from investor-owned utilities is that they are locally-elected government bodies subject to the Brown Act, which is a 1953 statute that guarantees the public’s right to attend and participate in meetings of locally-elected bodies. Can you share a little about Peninsula’s stance on information sharing?
Doherty: We share a lot of information. I came from a developer background and that was one of the things that has taken some getting used to. In the developer world, you hold your cards very close to your vest. In the CCA world, almost everything is public. We have monthly board meetings that are open to the public. We publish our agenda a couple days in advance. We publish our slides. We publish our contracts. We can redact certain commercially-sensitive terms. For every CCA, you can go onto its website and get a ton of information. Similar to what David said, we have created a section of our website with all of our financial documents. We have our quarterly financials going back to the beginning of our launch last year, as well as our joint powers agreement and formative documents.
Barrow: It takes time for any new business model to be accepted by the financial community. There is a learning curve where lenders struggle to get comfortable with risks. Let’s talk next about some of the unique challenges in CCA financings — we touched on some already — and also get some lessons from Mustang. Sondra Martinez, are there any key takeaways or pieces of advice you want to share?
Martinez: Sure. Much like Beth Waters mentioned, NORD/LB financed several short-term PPAs with CCAs as part of a package with a long-term utility offtake. We dipped our toes in initially to understand what CCAs are, but it was easy to take the risk because, as Beth mentioned, even if the CCA contracts fell away, we still felt comfortable that our debt would be repaid. When it came to Mustang, that was completely different. There was no external credit rating. It was really important to our credit guys that we needed to be able to value on a regular basis the creditworthiness of the entity. One reason we were able to do the transaction is we had NORD/LB do its own internal rating of the offtakers. We needed the financials to do that. It was important for us to put into our credit documentation that we would receive quarterly unaudited and annual audited financial statements so that we can monitor whether the financial health of the CCA is deteriorating. That was a driver of structural mitigants, such as a blocking of cash, cash sweeps and things like that. As a lender, you worry about long-term creditworthiness. You can diligence all you want, but if you start to see the credit deteriorating, you want to be repaid faster or you want the sponsor to step up and right size the loan or you want to do something to make it painful enough to the sponsor and its return that it will help solve the problem on the financing side. We compared the PPA price to potential merchant curves, things of that sort, but you also have to believe that the regulators are going to stay out of the way and let CCAs develop.
Plaxico: We constructed Mustang using a different type of financing, and then we refinanced that later. What Sondra did not mention in her timeline of February to July 2016 is that Recurrent had already been talking to more than 20 banks before her timeline started. Most gave a thumbs down due to lack of a credit rating.
Barrow: Magali Cohen, any thoughts?
Cohen: A critical element for us was to make sure that the value proposition to customers would be sustainable and to ensure that the CCAs had an adequate power procurement and operating reserve strategy. We spent a lot of time analyzing the counterparty risk. Investec is used to financing complex transactions that may include unrated offtakers, but we need quarterly financial statements and historical operating data. We read the information that the CCAs provided on their websites to be able to understand their business strategies and the regulatory framework they operate under. We evaluated the CCAs’ financial and operating performance to establish an internal rating.
Martinez: It is a long-term partnership, right? We are looking at long-term amortization profiles for the debt.
McNeil: Marin Clean Energy has a reserve policy that is helpful. We accumulate surpluses over time in order to build balance sheet strength. It sends an important message to lenders in terms of strategic direction and the capital structure of the business. I encourage CCAs that do not have reserve policies to develop one. We take an approach of contributing a percentage — in our case, 4% — of our annual revenues to reserves, so that’s the amount of our annual surplus. It is predictable. You can see the growth over time. You can take a different strategy. You can take a strategy of accumulating more reserves faster in the earlier part of your development. I think that would also be a reasonable approach. But having a policy document that is approved by the board, so that there is organizational buy-in to the strategy, is an important part of the credit story.
Barrow: Siobhan Doherty, does Peninsula Clean Energy also have reserves?
Doherty: We do. We set aside 5% of our revenues currently. We are going through a process to refine that and add some more details to it, but we have a similar reserve policy to Marin Clean Energy and are looking to build those reserves over time.
Barrow: I understand you also use a lockbox account. Could you explain how it works?
Doherty: We set up a lockbox when we launched as a way to help partners that were unfamiliar with the CCA and were concerned about the credit structure. Revenues from the sale of electricity to our customers are directly deposited into a separate trust account. PPA providers that are part of the lockbox are paid directly out of that account each month. It is not until we pay all of those invoices that any of that money gets swept over to PCE to pay for operating expenses. It is a way for a developer or other counterparty to feel more secure that it will be paid ahead of us putting aside money into our reserves or spending it other ways.
Barrow: Beth Waters, as a new entrant dipping its toes into this market, what concerns you?
Waters: CCAs remind me of when we started financing coops eight or nine years ago. You look at the balance sheet and go, “Oh, my God, there’s nothing here” because they are not-for-profit and they are not meant to be profitable. CCAs are also not-for-profit, so it took a lot of time, but it helped that we were already lending to coops on the other side of the bank as that was a way to educate our credit people. I remember spending a month or two writing a strategy paper to help our credit people understand what CCAs are, how they function, why we should not be concerned. Most banks in our business do not finance a coop unless it is rated. There are very few that will finance coops when they are unrated. We came up with an in-house tool that we use to provide a shadow rating. That makes credit comfortable, and so we are moving ahead. I remember eight or nine years ago when the first corporate PPA appeared with someone like Google, our bank was like, “What? Who are they? Wait a minute. We know who they are, but they’ve only been around 10 years. Why are we lending on their credit? They are not a utility. They are not a load-serving entity.” The sanctity of the contract is key to this business. PPA prices a couple years ago were $100 a megawatt hour, and now they are $25. But that does not mean you can walk away from that prior contract. Markets go up and down. The entire system relies on certainty that offtakers must honor contracts over the long term.
McNeil: There are important distinctions between non-profit and for-profit organizations. But it is important to understand that CCAs like Marin Clean Energy are actively accumulating surpluses every year. Our revenues exceed our expenses. We retain those surpluses in the organization in the form of reserves. We do so in order to project financial strength and to demonstrate creditworthiness to our counterparties. It is a fundamental part of our strategy. In that respect, we are similar, at this stage of our growth, to a for-profit organization. The reserves are always retained in the entity. Our JPA members have no claims on any of the reserves in the organization. We do not pay dividends. We do not distribute funds to our members.
Martinez: That is an important point. There have been periods of time when CCAs were charging rates that were higher than the incumbent utility, and yet customers stayed with them. The reserve and not having to funnel that money back to your member city, but instead being able to use it in case there are blips in the market — for example, Sonoma has been affected recently by wildfires — is important. The reserve makes a CCA better able to weather those types of events.
Doherty: When I looked at our Q3 financials, we had about $40 million in cash. We continue to set aside money to ensure we have a strong balance sheet. We will be pursuing a credit rating over the next couple years. The reserves will be important to demonstrate liquidity to the credit rating agencies.
McNeil: We have about $70 million in assets and $35 million in cash. We enter into long-term PPAs, but those are not considered assets on the balance sheet. The way we incentivize renewables — with tax credits — in the United States is an impediment currently to CCA ownership of generating assets. However, the day will come when those tax subsidies go away and, at that point, we will be looking at acquiring assets and financing those on balance sheet, but that is probably five or six years away at a minimum.
Martinez: The assets are the customers, right? From a lender perspective, we thought a lot about the customer base and the stickiness of those customers.
McNeil: That is true. Customers are obviously the source of revenue.
Barrow: If customers are your biggest assets, let’s talk about the fact that customers can opt out at any time. They pay a nominal fee — I think it is $25 for a commercial customer and $5 for a residential customer — to exit. Let’s start with opt-out rates. What is the current opt-out rate for Marin Clean Energy?
McNeil: That’s a great question. We don’t as an industry do a great job of explaining opting out, so let me try. You have a total population that you can serve, and then you are only providing service to a certain percentage of that population. In our case, it is about 90% of eligible customers. So our opt-out rate by definition, is 10%. Most of those opt outs take place during the enrollment period. Minimizing opt outs during the enrollment period is incredibly important to Marin Clean Energy from a mission and revenue perspective. However, the opt-out rate during an enrollment period does not really matter from a risk perspective because we are not procuring for that load over the long term. The risk that CCAs have is that you have a whole bunch of customers, you procure for those customers, and then they opt out. So we have an opt-out rate of 10%. I think Peninsula has an opt-out rate of about 2%. You can have an opt out rate of 15%. From a risk management perspective, what matters is that you retain customers once the enrollment is complete. In our experience going back seven years, customers remain with the program.
Barrow: Do you see opt-outs once you get past the enrollment period?
McNeil: We are going through the eighth enrollment in the company’s history this coming April. We have been enrolling communities every 12 to 18 months since inception in 2010. The opt-out rate has fallen over time. We were as high as 20% in some of the early communities and, over time, the rate has fallen to 10%.
Barrow: Siobhan Doherty, from your perspective, PCE has been in the game for less time than Marin Clean Energy, but what has PCE seen for opt-out rates?
Doherty: Our opt-out rate is about 2%. We have had two enrollment periods. Our first one was in October last year and then in April of this year. We saw higher opt outs during those periods, and we are very conscious about managing those. We look at our opt outs on a weekly basis. We look to see who has opted out both in terms of the type of customer — whether it is a large industrial customer or a residential customer — and where the customer is located, and we do a lot to manage that. If we see a particular number of opt outs from one city, we can do outreach in that city and drill down into what is causing opt outs in that particular city. If they are caused by social networking sites posting information that may be inaccurate, we can try to counter the misinformation.
Barrow: David McNeil, do you have anything to add to what causes opt outs?
McNeil: The biggest driver of opt-out rates has been negative advertising about CCAs and Marin Clean Energy. We experienced a lot of that in the early stages of our existence. A code of conduct was eventually enacted for the investor-owned utilities, so they cannot use ratepayer money to market against CCAs. Since then, opt-out rates have declined. Customers are to some extent rate sensitive. We were about an average of 4% more expensive on a total-bill basis from January 1, 2016 through the end of August 2016. Residential customers were 5% higher on average. We saw virtually no change in customer count over that period. However, if there were a bigger gap — say we are 10% to 15% more expensive — the truth is we do not know what would happen, but we also do not see that scenario on the horizon.
Barrow: I also understand there is a bit of opt up that can be a mitigant to opt out. Can you unpack that one for us?
Doherty: I can start. At PCE, we offer our customers two different products. Our default product is about 50% renewable energy, 80% greenhouse gas-free, and we have priced that one 5% below PG&E rates. Then we offer a second product that is 100% renewable and customers can choose to opt up to that product, and it is slightly more expensive. It is 1¢ per kilowatt hour higher than our default product. We have seen a bit under 2% of our customers choose to opt up to that product. We have not done aggressive marketing. We have been operating for only a year, so we have been focusing more of our marketing on brand recognition as opposed to getting people to opt up. We had a lot of success with our cities. Twelve of our cities plus the county of San Mateo have opted up their municipal accounts to the 100% product.
McNeil: Marin Clean Energy has a deep green product for which we charge a slightly higher rate. It is a 100% renewable product. The margins on that are thinner than the 50% default product, so I am not sure it would offset loss of revenue from opt outs. I don’t see it as a terribly material factor in our overall financial picture.
Barrow: I have one more question for the panel. We talked about opt out on the customer level. What about opt out at the JPA level? It is not really opt out, but what happens when a municipality withdraws from the JPA? Lenders, is that possibility of concern to you? Did you look into that risk in Mustang?
Martinez: It was a concern. I think it is one that is relatively easily mitigated. We looked into the cities. Many of those cities have their own goals for renewable energy. We are familiar with JPAs and SCPPA. There is a long history of how joint powers authority regulation works, so I think we were less worried about cities opting out since there are significant barriers for them to do so that are much higher than when an individual resident says, “I have changed my mind.” There would be a financial impact to them exiting. At the end of the day, we were less worried that they would opt out.
Cohen: We did a similar analysis to understand the financial penalties and termination payments associated with an entire community opting out. We view the individual customer attrition risk as more of a threat than the entire city opting out.
Waters: If you are embracing CCAs, you must be embracing that the members of one of these groups are not looking to leave. You can do a sensitivity analysis to see what is breakeven and see how many you would have to lose before it starts becoming a problem for the financing. There are a number of things you can do to get your arms around the risk and get comfortable.
Barrow: I think there’s also a provision in most JPA agreements that says that before a municipality can leave, the JPA must authorize the exit and the municipality will remain liable for any power costs that were procured on its behalf before it left. That is also a mitigant.
Waters: Meaning it makes a hole in their portion.
McNeil: You nailed it. That’s exactly what it implies. There is a provision in the JPA agreement that requires the board of the JPA to approve the departure of the member community, and the exiting community has to make the JPA whole for any losses that arise as a result of the departure of their customers from the service. Doherty: PCE’s works the same way.
Get news, insights, podcasts, videos, webinars, and events delivered to your inbox every Tuesday