The US Distributed Solar Market
A group of solar rooftop industry executives had a wide-ranging discussion at the PV America convention in Philadelphia in February about the basic business models in use in the US market, customer default rates, investor returns, barriers to entry, emerging new financing strategies and other issues. The panelists are Ben Cook, vice president of structured finance at SolarCity, Kristian Hanelt, senior vice president of renewable capital markets at Clean Power Finance, Laura Stern, president of Nautilus Solar Energy, Sandy Roskes, vice president for sales at Astrum Solar, and Song Yi, chief financial officer of Standard Solar. The moderator is Keith Martin with Chadbourne in Washington.
MR. MARTIN: What is your basic business proposition for residential customers?
MR. ROSKES: We offer flexible options that meet customers’ needs with market-based pricing, but with a high-quality experience from the time of initial contact through to interconnection.
MR. MARTIN: I have heard Lyndon Rive, CEO of SolarCity, say that his basic proposition is a customer can have solar panels on his roof for free and draw electricity for monthly payments that are roughly 85% of what he is paying the local utility. Is your basic proposition the same?
MR. ROSKES: Absolutely. Most people want to go solar for free and save money immediately. We spend a lot of time trying to figure out what their true needs are and what best fits their goals. Fifteen percent off your bill for no money out of pocket is a good benchmark. Depending on what state you are in, and what the particulars are, you might even do better than that.
MR. COOK: We introduced the solar lease in 2008 and, since then, have been offering customers, at no money down, a save-money-month-one proposition. People are already paying a utility for electricity on a per-kilowatt-hour basis. Rather than try to sell them a lot of equipment and require a huge upfront payment, we allow them to go solar in the same way they are already used to buying electricity.
The proposition is the same for both residential and commercial customers. Customers want to save money. The most important green to many customers is the dollars that remain in their pockets.
MR. MARTIN: Laura Stern, you focus solely on the commercial and industrial market. Is your offer to customers the same?
MS. STERN: We have two different models. One model is a power contract with the customer paying for electricity at a discount to local retail rates. The other model is for markets with feed-in tariffs where we pay customers for the right to have our solar panels on their roofs, and we sell the electricity to the local utility to earn the feed-in tariff.
MR. YI: We have both commercial and residential customers. For residential customers, we offer four different products. We offer solar panels, energy efficiency, traditional generators and smart homes.
Energy efficiency provides probably the greatest value. For about $3,000, a customer can save close to 15% to 20% on his energy bill. The savings add up rapidly. It is basically new insulation, windows and changing out light bulbs to LED light bulbs. We give the customer a picture that shows where he is leaking energy and what to do.
MR. MARTIN: What percentage of your business is energy efficiency as opposed to pure solar?
MR. YI: Probably 20%. It is a low-ticket, high-volume business. A lot of the utilities like Baltimore Gas & Electric and Pepco provide energy audits for free. We receive about $200 to $300 from the utility when we audit. Where we make money is in retrofits.
MR. COOK: We offer the same service and are seeing growing demand among customers for it. As Song Yi said, once you go solar, all of a sudden you have a much better understanding of how much electricity you are using. After we install a solar system, or sometimes even where we have not installed the solar system, we will offer a home energy evaluation and build a full energy model of the house. We determine every aspect of how the home is using energy and offer solutions. There are a number of things that the typical customer can do to lower energy costs. This is still a very small part of our business, but it is growing quickly.
MR. MARTIN: Kristian Hanelt, you work with various solar companies. Do you see most of them coalescing around a single business model and, if so, what is it?
MR. HANELT: Most offer 85% of what the customer is paying the local utility, plus or minus 10%. Some companies are very successful at just selling solar at a very slight savings, but what they are selling is getting off the utility.
MR. MARTIN: For those of you who start at a percentage of the monthly utility bill, do you adjust the payments over time for inflation and, if so, what is the inflation adjustment? Or do you adjust the monthly customer payment by looking periodically at the utility bill and always trying to stay below it?
MR. HANELT: Our installers work with homeowners to find the right inflation adjustment that the homeowner wants. It could be zero or it could be up to a 3% annual escalator.
MR. ROSKES: We offer a similar range. We offer the same fixed payment for the full term of the contract or we can start with a lower initial payment and adjust for inflation.
Lease Versus PPA
MR. MARTIN: Laura Stern, you told me Nautilus has had systems in operation since 2007. Ben Cook, what is the oldest system SolarCity has in operation?
MR. COOK: We started installing systems in 2006 and offered our first “zero-down” solar lease in early 2008.
MR. ROSKES: We started installing in 2008. Our first finance product was in 2010.
MR. YI: We have been installing solar since 2006, but we started offering financing products like a lease or PPA to homeowners in 2011.
MR. HANELT: Clean Power Finance was founded in 2006, but did not start financing until 2011.
MR. MARTIN: Some companies offer to sell customers electricity under long-term power contracts or PPAs, some offer to lease solar systems to customers, and Laura Stern has a completely different model where she is leasing space on a roof and selling electricity to the local utility. Is the power contract or lease always 20 years? Why choose to sell customers electricity rather than lease them systems, or vice versa?
MR. HANELT: We offer leases, PPAs and loans. We work with installers. The idea is to give them all of the tools they need to make the sale. We offer leases of up to 25 years and PPAs of up to 20 years. Certain installers prefer one over the other. Homeowners sometimes choose a lease because maybe they lease a car and a PPA is a little foreign. The two products are really very similar; there are slight differences in who bears the risks and the rewards of production.
MR. YI: We offer both a PPA and a lease in substance, but we call them both PPAs. The difference is that the customer pays a per-kilowatt-hour charge for electricity under a PPA and the customer pays a set monthly fee, regardless of consumption, under a lease. We use third-party financing. For some reason, the financiers do not want the product to be called a lease, so we call both PPAs.
MS. STERN: All of our projects are under PPAs, but some are with the host customer whose property we are on and some are with the local utility. We do not pass off any risk of production, installation or operation to our customers. They just buy kilowatt hours.
MR. COOK: We offer both leases and PPAs. Some states do not allow PPAs and so we offer leases, but more and more, we do PPAs where they are allowed because the value proposition is simpler. If you are spending X cents per kilowatt hour and we can offer you something less than that, why wouldn’t we do it? We now are in more than 500 Home Depot stores across the country, and people come in to buy dirt, a ladder or a hose and they walk out having signed up for 20 years of electricity.
The states where we cannot offer PPAs are those that do not allow retail sales. The local utility has a monopoly on that. An example is Arizona.
MR. MARTIN: Sandy Roskes, you are head of national sales for Astrum Solar. Which do you find easier to sell: a PPA or a lease?
MR. ROSKES: We have used both and still have a mix, but I have a different point of view. I think leases are more attractive to customers for the simple reason that we can tell them what their monthly payments are going to be forever.
Customer Default Rates
MR. MARTIN: What are customer default rates?
MR. ROSKES: They are about as close to zero as you can get.
MR. COOK: The simplest way to think about default risk is that utility payments are the operating payments for a household. Operating costs are generally paid first regardless of assets. People pay for electricity. We have installed more than 30,000 systems to date and have had very few defaults.
Unlike financing a restaurant where, if the investment does not pan out, you have lost your money, with solar, especially residential, it is not a question of risk of loss but a question of cash flow interruption. Folks who have decided not to pay for electricity are generally not going to be in the home very long. What you really have is a question of how long until the foreclosure and until there is a new tenant in that house.
MR. MARTIN: There was a fear in California, Nevada and some other states where homeowners were under water on their mortgages that the homeowners might abandon their houses. How real did that fear prove?
MR. COOK: We were a great test case because we started offering solar leases in 2008 before the crash. We found that, despite all these fears about mortgages being under water, people continued to pay for electricity. We have had quite a few lease transfers. Some of the companies on this panel offer 25-year PPAs or leases, and the average person stays in his home for six years, so we should expect to see lots of transfers as houses change hands. We have a full-time department that does nothing but help transfer solar contracts from one customer to another, but defaults have been exceptionally low.
MS. STERN: We have customers that range from nonprofits to utilities. The only hiccough we have had was with an investment-grade utility.
MR. MARTIN: What minimum FICO scores do you insist on for residential customers?
MR. YI: Our customers must have FICO scores of 700 or higher.
MR. COOK: We require 680, but as the industry matures, solar will be offered to lower and lower FICO score customers. The fact is that people pay for electricity. Much like the mortgage industry where you do not have a cut-off where, if you are above a certain FICO score, everybody can get financing for a house and, below that, it is “I’m sorry, you just can’t qualify for a mortgage,” I think you will start to see more and more differentiation.
MR. MARTIN: Kristian Hanelt, what is your minimum FICO score?
MR. HANELT: We have five funds under management that all have unique credit requirements. The FICO score minimum of the lowest fund is roughly 660. The fund looks at other things in addition to FICO. We also have a super prime product that has an average FICO in the 780s. We try to place products along the credit curve. A fund with 700 FICO customers should get a worse price than a fund with 780 FICO customers.
MR. ROSKES: Our minimum has been around 700. The market is creeping down to 680, but I concur with everybody here. You are going to end up with different thresholds that bring different cost financing. You will end up with a sliding scale much like mortgages. The default rate will remain low because the customer’s alternative is to pay more for electricity from the local utility.
MR. MARTIN: If customer defaults are not a problem, then what is the most common customer problem?
MR. COOK: Education. Solar has spread on the back of a very basic value proposition — “save money month one, no money down” — which is why the third-party ownership business model has prospered. Where five years ago, a very small percentage of the systems used this model, now in many markets it is well over 80% or 90%.
MR. MARTIN: So the biggest problem is the effort it takes to acquire customers. What is the most common problem once you have customers?
MR. COOK: The most common problem once you have the customers is the logistics of keeping thousands and thousands of customers happy and making sure that everybody has a good customer experience. It is one thing to have 10 or 100 customers where you can think about each individually. When you are dealing with thousands of customers across thousands of jurisdictions for building permits and interconnection standards, it becomes very challenging to keep the customer wait times low and the administration reasonable. It is one of the things that SolarCity has done very well.
MR. HANELT: We started with a broadband-based approach to monitoring, but it was surprising how many households do not have a reliable internet connection, so we moved to cellular monitoring. Homeowners would promise to keep internet
service, but a significant amount of the portfolio may not be getting a signal at any given time.
MR. MARTIN: So they pay their electricity bills, but maybe not their internet provider bills.
MR. YI: We are more of an installer, and probably 60% of our installs are cash deals and 40% are financing through a third party. For financed systems, we offer O&M service after installation. If the system goes down, we go out to repair and put the system back in service. For both types of systems, once we install, we collect cash and, from our perspective, ours is a cash-and-walk-away deal. So we have basically a zero default rate. We collect every cent that we install.
MR. MARTIN: Some people have suggested that asking customers to sign a 20-year contract is like asking them to sign up for a lifetime of electricity. They suggest solar companies need to move toward a cell phone-type model where the customer can switch providers with a month’s notice. Do you think this is the direction in which the industry will eventually move?
MS. STERN: It does not work for commercial and industrial projects. We have a large number of rooftop systems, but we also have ground-mounted projects and parking canopies. These are very large, capital-intensive investments that really are not intended to be transported and moved. There is currently no secondary market for these projects. As of today, they are intended to have 30-year lives.
MR. MARTIN: People lease cars that cost as much as a solar system from General Electric Capital Corporation or the local bank, but it is for a short time period. Why should solar be different?
MS. STERN: The difference is that you can buy a previously leased car on a secondary market from a dealership and have a great warranty and be perfectly happy with that car. Someone else can drive it after your four years are up. You cannot move a parking canopy efficiently. You can recover the solar panels, but they are becoming an increasingly smaller part of the capital cost of the project.
MR. COOK: With the cost of PV modules declining, a greater and greater percentage of the cost of a solar asset becomes the labor associated with installation or de-installation. The real difference between the auto market and the solar market is solar does not have wheels. You cannot move it so easily to another place when the value of the equipment is really the ability to produce electricity at a particular location and when the labor content is so high.
MR. HANELT: I think that some of the community solar initiatives that people have been trying to get underway in California and other states are an attempt to provide a transferable service where you can do that. You basically are buying solar electricity from a centralized location under a shorter-term contract.
MR. ROSKES: The key is to have agreements that contemplate any potential outcome. There is really no outcome where the customer will not need electricity, but the customer may move, pass away or sell the house. The key is to provide answers to all of those potential outcomes. A customer who is unwilling to commit to 20 years of service may prefer to purchase the system. It may make financial sense to do so. Our mix is much more weighted toward leases than purchases today because a lease is such a great financial option, but it is not the only option. Transferring a leased system to a new homeowner has to be easy. You have to build the transfer agreement into the original contract. Moving a system is probably the least beneficial outcome for all; however, it is an option that we offer. If you spell out all the potential options in the customer agreement, then the fear of entering into a 20-year agreement is diminished.
MR. MARTIN: Some have suggested that the cost to acquire the customer is about 50% of the cost of an installed solar system. The equipment is about 25%, and the cost of capital is about 25%. Do those numbers sound right?
MR. YI: For Standard Solar, the customer acquisition cost is much lower, easily below 20%.
MR. HANELT: We see a lot of different business models, but from what I can piece together, good originators will originate for $2,500 per solar system per customer, while best-in-class folks are claiming they can get into the $1,000 range or even below that. There are a lot of different models for originating customers — be it call centers or door-to-door sales or anything else — and it is a big part of the cost and often greater than the cost of the modules, depending on the installer.
MR. MARTIN: I think when people say 50% cost to acquire the customer, they are talking about the allocated cost of the web site, the call center and the salesmen out calling on potential customers. Let me switch topics to the operating phase. You have a high capital cost to install, but once you get into the operating phase, isn’t it like an alarm business where you collect a monthly fee without having to do much for it? How expensive is it actually to monitor and repair?
MR. COOK: Yes. The fact that you have no fuel costs and no moving parts means that in the grand scheme of things, the operating costs of a solar project should be very low. Typically the kinds of things that you need to worry about are monitoring and then the need to fix things when a system malfunctions. You also need to make sure that you are providing reporting to the customer or to investor partners on a periodic basis.
MR. MARTIN: What are costs as a percentage of revenue once you get into the operating phase?
MR. ROSKES: The percentage is very low. Solar panels are a steady state product; they do not break. Communication with the customer is key. The customer gets monitoring. He gets alerts.
The beauty of a 20-year agreement is you are on the roof for 20 years. It is terrific that you have these opportunities on a monthly basis to communicate with your customers, potentially to provide them other services or to solicit referrals. The key is about turning that into an opportunity.
MR. MARTIN: You have high installation costs. Once you get into the operating phase, you have very low costs and, presumably, that is a very good part of the business to be in. Has anybody in this market turned a profit yet?
MR. YI: We have been turning profit for the past two years. We have a lot of direct sales, but the O&M business is also taking off. We have a lot of systems in, and we are also accumulating a lot of O&M contracts from different developers who do not have that capability.
MR. COOK: I cannot comment on profitability since we are now a public company. Speaking more generally on behalf of the industry, I think it is important to distinguish between profitability and positive cash flow. For the third party ownership model, where there is substantial ongoing involvement, you do not get sale treatment upon installation of the system. You amortize the gain from installation over the term of the customer agreement. It is important when looking at companies with this business model to look beyond GAAP profitability and focus on cash flow. If you are growing quickly and gain on installation is spread over the term of the customer agreement and all of the operating costs associated with customer acquisition and everything else are expensed, then it is going to be a very difficult GAAP presentation. Looking at cash flow is a much more meaningful metric.
MR. MARTIN: What returns are you promising investors who invest in your companies?
MR. HANELT: Our funds have the ability to set their own return thresholds, but we have seen the returns drop quite substantially in residential since the year and a half we have been in the business. They have definitely come down from two digits to one digit.
MR. MARTIN: These are returns for the developer or the third-party investors?
MR. HANELT: The returns for the investors. Our model is different because we are not a sponsor. We do not take long-term ownership of the asset, so we do not have the same kind of issues that the other companies may have. GAAP explains our business just fine. Our investors own 100% of the assets and use 100% of the cash and tax benefits, so our return is an unlevered after-tax weighted average cost of capital for the whole financing product that we sell.
MR. MARTIN: This is a market that has had low barriers to entry. A lot of roofing companies act as installers. Are we at a stage in the industry life cycle where there are still a lot of new entrants or are we starting to see some consolidation?
MR. COOK: There are low barriers to entry for companies that do relatively small volumes. If you are a small roofer or a small installer of some other type of product, then maybe it is not so hard to expand into solar. If you want to build the infrastructure that Sandy Roskes described, then it is a very high barrier to entry. The barriers will become even greater as customers become more demanding about post-installation service.
MR. ROSKES: The market is bifurcating. There are very low barriers to entry at the low end of the market. There is nothing to prevent a roofer or an electrician from installing solar. There are no national roofers or electricians. However, there are emerging national solar brands. What we are seeing is that there is a consolidation in the middle with people exiting the market or consolidating. You either get big or stay very small — that is the bottom line — because of the advantages of scale in things like customer acquisition, customer care, information and communication. These are areas where the benefits of scale are enormous. You will continue to see a hollowing out of the middle. There will be plenty of roofers and electricians installing one or two systems a month, and a very small number of large players that are regional or national in scope.
MR. MARTIN: Homeowners who purchase systems qualify for a 30% residential tax credit. Where the solar company retains ownership, it qualifies for a 30% investment tax credit and five-year depreciation that is equivalent to an additional 26% tax credit in terms of additional tax savings over time. Is tax equity the main way of financing — trying to get value for these tax subsidies that the solar company has too little tax capacity to use directly?
MR. COOK: Tax equity is essential unless you are part of a parent company that has unlimited tax capacity. Basic math requires that you monetize that tax equity externally. However, tax equity is not the only part of the equation. One of the essential ways to lower the cost of capital and be able to enter more and more solar markets will be the ability to monetize the cash flows from the system independently of the tax equity.
MR. MARTIN: Do you foresee people combining securitizations and tax equity?
MR. COOK: I would frame it more broadly as debt. Securitization is a form of debt and one that is naturally suited for large bundles of diversified customer credits, but more generally, debt is something that looks to a fixed series of payments over a long period of time in exchange for up-front capital. Debt and tax equity can coexist. As solar becomes a more mature asset class, we are seeing more and more lenders who are comfortable with the risks associated with solar projects.
MR. MARTIN: Laura Stern, you told me before this session started that you are now looking for strategic investors. What is the challenge with tax equity?
MS. STERN: Tax equity is required. It is essential to have an owner who has tax appetite and who is going to monetize the tax benefits efficiently. Tax equity is challenging in the commercial rooftop market because you don’t have the same deal flow as in the residential sector. The challenge is bundling enough projects and having investors come in before the projects are put in service and to execute efficiently with reasonable transaction costs.
MR. MARTIN: How large a portfolio do you think somebody needs in order to be able to raise tax equity, and what do you think current rates are for tax equity? How much does it cost?
MR. HANELT: We have done residential funds as small as $25 million. We have gotten pretty good at making them easy to close. As far as rates go, we have seen as low as 7% after-tax unlevered and as high as something that is not even really meaningful, but in the upper teens.
MR. MARTIN: Where do you think rates are right now for a partnership flip transaction?
MR. HANELT: They are in the single-digit range for a solid developer working with a large tax equity investor who does a lot of deals. The expiration of the section 1603 program will raise the bar on what it takes to raise capital. If you can get over the bar, then tax equity rates are pretty good. If you cannot get over the bar, then you may not be able to find tax equity on any terms.
MS. STERN: One should also distinguish between the yield that a tax equity investor is getting versus the cost to the developer to implement it. There can be two tax equity investors in the same project; one gets a single digit return and one gets a double digit return because they have different residual value assumptions, and they have different tax rates. They are booking it differently. People tend to focus on what the yield to the investor is, but the way to look at it is, what does this investment cost me? What value am I getting out of it?
MR. MARTIN: What have you been paying for tax equity?
MS. STERN: It is not entirely visible. But a developer should focus on other things. What are your options? What does it cost to put the financing facility into place? Would you rather do it with one investor or another? Are you going to be able to raise tax equity at 1.3 times the investment tax credit value, 1.25x, 1.35x? These are the considerations you should be looking at as opposed to what is the yield to the investor.
MR. MARTIN: There are three main structures for tax equity: master sale-leasebacks, master partnership flips and master inverted leases. SolarCity does a lot of these. Which structure is preferred and why?
MR. COOK: We use all three structures, but we see mostly partnership flips and inverted leases. These structures make more sense for a company that hopes to own the asset long term. In an inverted lease, the asset comes back to the developer without the need to pay anything for it. In a partnership flip, the developer gets 95% of the asset back and has an option to repurchase the remaining 5%. In a sale-leaseback, if the developer wants to keep the asset at the end of the lease, he must pay full value for it.
MR. MARTIN: There has been a lot of talk about securitizations. The main impediments are a lack of default data — there are only a few years of data on customer default rates — the need for standardized contracts across the industry and the need for known service providers. On standardized contracts, if you look at the other securitization markets like student loans and home mortgages, they use form contracts. You cannot change a term. What progress is being made on securitizing customer receivables in the rooftop solar market?
MR. HANELT: There is a group called Solar Access to Public Capital that National Renewable Energy Laboratory is sponsoring in which SolarCity and Clean Power Finance and perhaps other solar companies are participating along with the rating agencies and other industry constituents. The group is focused at the moment on creating a standardized contract. NREL is also leading a charge to get all the major players to share anonymous and aggregated data around performance and default rates that should help rating agencies get comfortable. Progress is being made, but it is going to take a while before you have a truly liquid and fluid securitization market.
MR. MARTIN: How long? When do you think we will see the first securitization?
MR. HANELT: I think you will see the first one this year. If the tax credit is truly stepping down to 10% in 2017, then we need to have this mechanism in place by then because debt is going to be paramount.
MR. MARTIN: Another potentially new form of financing that people have been talking about is real estate investment trusts or REITs. Have any of you been pushing in this direction? Do you see much promise?
MR. COOK: The purpose of these strategies — securitizations, REITs and other structures — is to monetize future cash flows. The same work that is being done to facilitate securitizations will have to be done for any type of debt. The most important thing we as an industry need to do now is to make sure that tax equity and debt can coexist. Then you can shop across different types of capital and find the most efficient one. But if you cannot separate the tax benefits from the cash, then all these structures are theoretical.
MR. MARTIN: Many people thought, going into this year, that the cost of capital would increase for renewable energy companies because the two cheapest sources of capital are disappearing: Treasury cash grants that cover 30% of the capital costs and Department of Energy loan guarantees. In fact, has the cost of capital been going up? Has it been stable? Is it coming down?
MS. STERN: We see it as actually going down modestly, both as a function of industry maturation as well as investors seeking yield.
MR. MARTIN: So the financial markets are deciding over time that the solar rooftop sector is not as risky as thought earlier. Ben Cook, you described your strategy to try to drive down the cost of capital by trying to combine securitizations, solar REITs and other forms of debt with tax equity. What about the others of you on this panel? Are there other strategies you are pursuing to push down the cost of capital?
MR. COOK: There are many new players coming into the market on both the tax equity and debt sides. This is part of a broader maturation of the industry. There are various pools of capital across the debt markets taking a hard look at rooftop solar for the first time as the scale of available projects increases, and it is all leading to a lower cost of capital.
MR. MARTIN: When you combine project-level debt with tax equity, the debt is ahead of the tax equity in terms of priority of repayment. However, the tax equity will want some forbearance by the lenders if there is a default. The tax equity investor claims investment tax credits. They will be recaptured if the lenders foreclose on the assets within the first five years. Do you think forbearance will require the lenders to agr