The Next Generation of Solar Project Finance
A group of solar industry veterans talked at an Infocast distributed solar conference in San Diego in November about the need for solar rooftop companies to move over time to a new business model, growth rates in the US residential and commercial rooftop markets, possible pivot points that could cause dramatic shifts in the market, how the costs of installed solar systems break down among equipment, labor, customer acquisition and capital, and current returns for developers and financial players.
The panelists are Channing Chen, vice president of project finance and general manager of the small systems group at SunEdison, Ed Feo, managing director of USRG Renewable Finance, Ori Franco, senior director for finance at Sunrun, Inc., David Loomis, West Coast sales manager for ReneSola, and Chase Weir, CEO of Distributed Sun. The panel was moderated by Keith Martin from the Chadbourne Washington office.
MR. MARTIN: The business model that has gained the most traction in the US solar rooftop market is the offer to customers to have solar panels on their roofs for no upfront payment. The solar company retains ownership of the panels. The customer signs an agreement to buy electricity or lease the panels for 18 to 20 years for monthly payments that are something like 85% of what the customer is paying currently for electricity from the local utility. What other successful business models have you seen?
MR. LOOMIS: The models that have been successful and will continue to be successful are low- or no-down-payment products where the customer can see that he or she is saving money from the start.
MR. WEIR: Most of the innovations are in the models with third-party ownership. It is probably not widely known yet, but BMW is in the process of implementing a long-term strategy to provide the automobile as a service. It proposes to move away from the model where people own their cars. The solar rooftop market has been moving in the same direction.
MR. FRANCO: We may lease systems in form to our customers, but in substance we are providing a service. Our customers are busy people who do not want to decide which installation company to use or what technology to use. They worry about the risks and burdens of ownership. Many customers have the money to purchase solar systems themselves, but they choose instead to go with Sunrun or one of our competitors because they want a service rather than the equipment.
MR. MARTIN: The benefit of this model is that customers do not have to pay up front, but some customers choose to prepay the full rent they would have to pay over an 18- or 20-year agreement. Why?
MR. FRANCO: Service. As a professional owner of the equipment, we can do a better job than the homeowner in monitoring, maintaining and ensuring that the equipment performs.
MR. MARTIN: How much of a discount does the homeowner get for paying up front?
MR. FRANCO: We have seen a discount of 5% to 10% compared to the cash purchase price.
MR. MARTIN: One unsuccessful business model was the PACE programs. Please explain what those are and whether they have a future.
MR. FEO: In PACE programs, a municipality advances a local homeowner or business the money to install a solar system or energy efficiency improvements. The property owner agrees to a property tax lien as the means for repaying the funds advanced.
The issue with residential PACE is that the Federal Housing Finance Authority said it would not allow a volitional tax lien to have a higher priority claim than the mortgage on the house to which the financed equipment is attached. The commercial PACE programs do not have this problem because the mortgages on commercial properties are not federally insured.
One of the companies I am involved with is a group called Clean Fund. It just did the first commercial PACE deal in San Francisco on the ProLogis building on Pier 1. The deal has first-party ownership of the asset, with the financing coming through a senior secured tax lien. The tax lien payments are monetized in the capital markets, which leads to a very high level of leverage at a relatively low rate.
MR. MARTIN: This is a PACE program aimed at commercial properties? Is the loan repaid through property tax bills?
MR. FEO: Yes. This is commercial PACE. The agency that runs the program issues municipal bonds at low interest rates and advances the proceeds to the property owner for the purpose of paying for the energy efficiency or the solar system. The property owner agrees to an additional property tax, which becomes the source of repayment.
A PACE program does not preclude third-party ownership, so you could take the proceeds received under a PACE funding and use them to prepay a lease or power contract and combine low-cost debt with the benefits of third-party ownership of the asset itself.
MR. MARTIN: Is it worth it to combine such debt with third-party ownership? Are the systems too small to justify such a complicated financing structure?
MR. FEO: The structures are essentially levered tax equity deals. The cost of capital is lower if you can put tax equity and debt together. Since these structures are just getting out of the gate, I think that it is fair to say the transaction costs, on a deal-by-deal basis, are relatively high, but will get lower.
MR. MARTIN: What about the community utility-scale solar installations where people who live in apartments or condos buy individual panels in a utility-scale solar facility that sells its electricity to the local utility. The panel owners get a credit for the electricity sold that they can use against their utility bills. What is the future of that model?
MR. FEO: It is an awesome model.
MR. FRANCO: I agree. I am personally and professionally excited about deploying solar, and that is certainly one way to do it. There are still kinks to work out. Administratively, you need to worry about virtual net metering and ensuring that the local utility can charge the customer the appropriate amount, so there is a regulatory change that needs to occur, but it is a great way to deploy solar.
MR. CHEN: SunEdison is looking at that model, and we are excited about it, but we probably have not moved as far as others actually to deploy it.
MR. MARTIN: Have any municipalities enacted the ordinances necessary to allow the credit mechanism?
MR. WEIR: It is happening in states across the country: Arizona, California, Colorado, Connecticut, Delaware, Maine and Maryland. It is too bad that California’s SB 843 did not pass, but maybe it will get some new life in 2013. At the end of the day, it is about equity. Ratepayers and taxpayers are already financing solar. It is also about meeting the targets in renewable portfolio standards. When community solar takes hold in four or five states is when solar will become a significant part of our generation mix.
MR. FEO: The model has the potential to permit solar to be provided to people who may not otherwise be qualified to buy a system or to sign a long-term lease or power purchase agreement. It allows for different tenors. Right now, the available solar financing is a pretty inflexible tool, with very long tenor contracts. A long-term power contract or lease is great for customers who qualify and who do not expect to move. I think the future of the business is going to be around bringing other customers in and getting closer to a model where you are offering customers something which is more of a service. It can have flexibility in terms of not just rate, but also contract term.
MR. MARTIN: What about other business models?
MR. LOOMIS: I like to think that the next evolution in solar will be when a department store installs 1.5 megawatts on the roof even though it only needs one megawatt of capacity for its own use. The other 500 kilowatts could serve the community or be used to provide discounts to employees on their utility bills.
Moving to a Service Model
MR. MARTIN: Ed Feo, you said the future of this business is really finding a way to have more flexibility. It is not locking people into 20-year contracts, but giving them options. I have always thought this business had more in common with the cable television business. Is your vision one of moving truly to a cable company? No one is locked into a cable contract for more than a month at a time.
MR. FEO: My view is that that the industry should adopt a theme of flexibility, where different terms are on offer, and the customer obligation is less about an equipment acquisition and financing and more about the terms of the service. For that to happen will require moving away from a model where the only place you can offer solar is to a customer with a roof. Community solar is one way to break the current model.
It is more of a regulatory challenge initially. Ultimately, flexible terms are what the solar industry must offer. Am I interested in a long-term contract from one of the solar finance companies? Well, I can go to a utility today and it costs me very little to hook up, and if I want to move, I tell the utility that I am leaving and I get back my deposit minus some nominal amount. What the utility offers is an incredibly flexible product. Today the solar industry is offering to sell equipment for cash or long-term financing that is relatively expensive. The average homeowner moves every seven years. That tells you something about that market. Non-homeowners cycle every 18 months. In order to capture more of the retail market, we have to offer something other than a long-term contract for a piece of equipment.
MR. FRANCO: That is absolutely right. Many customers do move every seven to 10 years. You need to ensure that the reassignment happens smoothly. You make sure that the new homeowner has an incentive to take over the contract by making that process easy. We have been through quite a few reassignments and have had very positive experiences with those reassignments because the new homeowner sees it as a no brainer. Why wouldn’t he or she sign up to save 15% versus the local utility bill and not have to worry about maintaining the system?
MR. MARTIN: How do you incentivize a new buyer to take over the house if the contract has had a 4% escalator running since inception?
MR. FRANCO: We do not have a 4% escalator in any of our contracts. Ours is at most 2.9% and in markets like the east coast, the escalator is closer to 1.5%. We have two products in order to mitigate the price risk to the homeowner. One is that we are willing to sell you power at a rate close to your current utility rate that will remain the same fixed rate for 20 years without any escalator. The other product is we are willing to sell you power at a lower rate initially that escalates at less than 3% a year, but with a cap that ensures your rate will always be below the local utility rate.
MR. MARTIN: What are the annual growth rates for the US market leaders in this sector?
MR. WEIR: Greentech Media said this year it is 71%. Q1 year-over-year growth was 100%, but it will probably end up for 2012 as a whole at 71%. That is growth in capacity.
MR. FEO: That includes utility-scale solar. Looking just at residential and commercial and industrial or C&I, it is a different story. C&I is down significantly, and residential is up modestly at 5%.
MR. WEIR: In seven of the last 10 quarters, C&I has been the largest segment. It was the largest segment in Q1 and Q2, but not in Q3. There have been a lot of utility-scale projects in the pipeline. Earlier this year, commercial was around three times the size of residential, but it has slowed down a bit this quarter. We do not see that as a sustained slowing.
MR. MARTIN: Does 44 million roofs sound like the potential market in the US?
MR. CHEN: I think that is about right. However, once you look at the technical constraints — whether the house is owner occupied and other factors — the viable market is closer to 30% to 35% of those 44 million roofs.
MR. FRANCO: In the current states in which we operate, there are about 29 million single-family homes. Roughly 15 million are owner occupied. From there, some percentage will not be viable solar candidates due to structure, shade and other causes.
There are about 300,000 residential solar facilities installed in the United States. We are seeing an acceleration in growth in the residential market. It took us four years to get to 10,000 customers and then 11 months to get to 20,000 customers.
MR. MARTIN: So there is a lot of room still for growth in the residential market.
MR. WEIR: The National Renewable Energy Laboratory estimated that commercial rooftops have 100 gigawatts of potential. We have analyzed a lot of nationwide portfolios, companies that own buildings across the country, and we see the potential for solar with the current technology in less than 10% of sites.
Potential Pivot Points
MR. MARTIN: Investors watch for potential pivot points in markets. These are things like changes in law, shifts in technology, inflation spikes, and changes in culture and weather that could cause the market to turn up or down. What are the potential pivot points in the next three to five years?
MR. CHEN: Technology will play a big role. There is a lot of effort to drag down the cost of polysilicon and increase the cell efficiency at a cost point that will make solar more attractive to homeowners to purchase systems.
MR. MARTIN: What will that do to the third-party business model?
MR. CHEN: I think for the next couple years, the third-party model will still drive the growth, but it remains to be seen what happens in 2016 when the 30% investment tax credit expires. I think it will be a very different world. Maybe that is when the more local or regional banks get involved with their solar loan products or securitization takes hold.
MR. MARTIN: What are other potential pivot points?
MR. WEIR: One pivot is when sustained commercial prices fall below $2 a watt. Installed cost is a big pivot point. Another pivot point will be when you see solar competing with conventional generation at 18¢ a kilowatt hour. The major pivot point is the technology. There are new technologies emerging that will radically expand how many roofs are accessible. We are a big proponent of the third-party ownership model, but we now view our business as services that lead to transactions that then lead to ownership. If we own the system, that is fine. If our partner, client or customer owns it, that is also fine. It is about delivering value to the marketplace. For us, the focus is the commercial and industrial sector.
We are beginning to see that the key to getting traction at scale is learning how to deal with the different channels within commercial solar. How you deal with REITs is radically different than the way you deal with a food distribution company. We are pursuing national account strategies, and the only way to do that is to be flexible. If you require that they take your form of power purchase agreement, then there is a lot of business you are not going to do. Flexibility is key.
MR. LOOMIS: All the third-party financing is driven by the current tax incentives: investment tax credits and MACRS depreciation. That is why we have customer agreements that run seven to 10 or 20 years. As the cost of systems falls and the tax credits expire, you will see 2- and 3-year contracts like cell phone contracts. That is the future. That is when you will get mass adoption.
MR. CHEN: We have a lot of internal efforts to drive down the cost of polysilicon and increase cell efficiency. My dream is to have a SunPower-like module at Chinese-level pricing.
Another potential pivot point is when we see solar offered as an add-on product to an existing business infrastructure. Vivint leveraged its home security business to get into solar. I think we will see more of these traditional consumer business companies enter the US market with home automation-type products and solutions. Xfinity is another example. The next product could be a solar-type partnership. Solar is energy, which is a low engagement category. You can only talk about the kilowatt so much at a cocktail party before people start talking about other subjects. It complements existing businesses. I see a lot of potential for these traditional consumer-based companies, the Vivints, HVAC and cable companies of the world, to come in and have the same people who sell HVAC, security systems or cable TV service also offer solar.
MR. MARTIN: What does that do to the SunEdisons of the world? There are very low barriers to entry in this market. What does it do to the incumbents?
MR. CHEN: I believe we at SunEdison have the ability to continue to drive down installation costs and, at the end of the day, this is really just a cost game. We are hoping to drive down the cost of capital because we have done close to $4 billion in institutional financing over the last eight years with many of the big banks. We have proven underwriting capability on a solar asset. We have a technology angle as well. If you can drive down the cost of capital, you can drive down the cost of products, and I think you will be okay.
We will survive based on cost and our ability to partner. We have the ability to partner with smaller developers and new entrants from more traditional consumer-based businesses.
MR. MARTIN: What is the relevance of state renewable portfolio standards and net metering programs to the current business models?
MR. FEO: Net metering is highly relevant. RPS programs are largely irrelevant.
MR. WEIR: You could not do solar without net metering. We are using the grid as a battery. There is another potential pivot point, and that is the coming wider use of energy storage and the growing unpredictability of the electricity supply. Many of you may not have been affected by Hurricane Sandy, but I can assure you there are millions of people today who are thinking very differently about the reliability of the local utility. There are people who were without power for several weeks after the hurricane. We saw hospitals that thought they had backup power, but were shut down for days. Those are key pivots. When you can store the solar energy on site, you do not need to use the grid as a backup. Until wider adoption of storage, you have a misalignment between the generation of the solar facility and the consumption at the site.
MR. MARTIN: What percentage of the cost of an installed system is currently capital, labor and equipment?
MR. FRANCO: In commercial solar, equipment is probably half. On the residential side, I think equipment is 20% or 25%. Labor is very little. Most of the cost is in customer acquisition and permitting or overhead on the residential side.
MR. MARTIN: What is the overhead related to customer acquisition: marketing to potential customers and getting them to sign contracts?
MR FRANCO: Yes. It includes signing the contract, originating the customer, dealing with canceled customers and designing the system. There are a lot of technological efficiencies that can be gained in system design. All of that is overhead, and this is an area where solar companies are working hard to reduce costs over time.
MR. MARTIN: What percentage of the cost is capital?
MR. WEIR: If I am installing at $4 a watt today and paying the current cost for capital, if I was able to pay 0% instead for capital, I could do install at $3 a watt. By that translation, the current cost of capital is $1 a watt for residential. Commercial is about 25% more.
MR. LOOMIS: It used to be that the equipment was 50% to 60% of the cost of the system. Now it is down about 20% to 25%.
MR. MARTIN: Channing Chen, if the equipment cost is 25%, capital is 25% and customer acquisition is 50%, what does this do to your supposition that SunEdison has staying power because it can chip away at the cost of capital?
MR. CHEN: Cost of capital is just one part of the equation. In the residential solar sector, we are also looking to customer acquisition as a place to reduce costs. If you can create competition among financing providers, you can drive the capital costs down. You cannot enter into a single negotiation with a financing party and expect to get the best pricing. The key is managing a process around competition for financing.
MR. MARTIN: Many people think that with the two cheapest sources of capital being pulled away — the Treasury cash grants and government loan guarantees — the cost of capital is bound to go up. Ed Feo, do you see it that way?
MR. FEO: I think that is what will happen in the near term. That said, I have been impressed that the pricing has not been crazy. Costs of capital are not increasing as much as I thought they would. That may be a function of investors getting their own return expectations in line with where investment opportunities are in the rest of the market.
MR. MARTIN: Some of you mentioned securitization as a future source of financing for solar rooftop companies. There has been talk about real estate investment trusts or REITs. Are these ideas, particularly securitization, for use after the tax equity market is gone or are they financing options that solar companies will begin to tap more quickly?
MR. CHEN: I hope they become financing options more quickly. I know SunEdison and SolarCity are actively looking at potential securitizations. The barrier may be the market getting comfortable with sparse historical data. We do not have more than six or seven years of data. We have talked with the investment banks to get a better feel for how the markets are looking at this asset class. I think we will see the first securitization close within the next two to three years.
MR. MARTIN: Ori Franco, will we see Sunrun do a securitization in 2013?
MR. FRANCO: I do not want to make a prediction about timing, but at least for the residential sector, the securitization market is a very deep pool of capital to finance consumer credit. Securitization investors or lenders understand consumer credit a lot better than tax investors do.
MR. MARTIN: Are tax equity and securitization mutually exclusive?
MR. FRANCO: No.
MR. WEIR: They do not get along very well, though. Having the investment tax credit in the capital stack makes securitization more challenging. I think we will see securitizations within the next few years, but before we get there, we will need to have an industry-wide rule set and standardized contracts. Residential is virtually securitized right now. It is repeatable and standardized. The homeowner is not negotiating a deal. That is not what happens in the commercial and industrial sectors.
Developer and Financial Yields
MR. MARTIN: What are current yields for developers?
MR. WEIR: They vary. We were talking about the cable television model. In residential, direct marketing is an easy notion; having a phone bank is a really good strategy. In commercial, you are seeing a move away from origination for a number of players into platforms where you have members and partners that originate for you. Companies like that have a much higher margin. A multi-megawatt commercial project can take 12 to 24 months to develop. That is a big acquisition cost.
MR. MARTIN: What are current returns for developers? I’ve heard at other conferences that they are in the 7% to 8% range.
MR. WEIR: If you are relying on Treasury cash grants or investment tax credits, they are hovering around 7% to 8%. If you are higher than that, you are above average.
MR. FRANCO: In the residential sector, we compete with retail rates that are generally much higher than industrial and commercial rates. Returns are healthy enough to drive our business.
MR. LOOMIS: I have seen margins anywhere from 8% to 25%. This is hard on the independent installers. They have been in the business for a while, and they are used to a 30% margin, and now they are being challenged by larger, more sophisticated businesses coming in with greater marketing prowess. Peterson Dean is a prime example of a roofing company that sold solar as an add-on, but that is now being challenged. The good news is there is still plenty of business for everybody.
MR. MARTIN: What are current tax equity yields? How much does tax equity cost currently for rooftop solar?
MR. FRANCO: We have seen deals from 8% to the low teens, unleveraged, on an after-tax basis. It depends on the structure and how fast you want to close.
MR. MARTIN: What is the cost of debt for rooftop solar?
MR. FEO: There are not many players interested in small systems. The range is between 6% and 12%, depending on the lender, the credit and the seniority of the debt.
MR. MARTIN: Where do you think securitization will come in? Will securitized debt be less expensive than tax equity and straight debt and, if so, by how much?
MR. FEO: Definitely. A typical REIT return is 4.5% to 5%, so that tells you something about where the base line is. On pricing for securitized debt, if you have the good fortune to get to investment grade, you are at 4.5%.
MR. MARTIN: We heard today that 25% of the cost of an installed system is the cost of capital. Are there any other innovations you see over the next three to five years besides securitization that will bring down the cost of capital?
MR. WEIR: Crowd funding is coming. It is still very early. There are a few companies in a quiet period. The US Securities and Exchange Commission is about to make a ruling. That definitely brings down the cost of capital.
MR. MARTIN: Solar Mosaic is an example of a company planning to use crowd funding to raise equity. It plans to raise money in $25 increments over the internet.