by Stephen Lacey
February 16, 2017

Stephen Lacey: This is The Interchange -- a weekly conversation about the changing business of energy and cleantech from Greentech Media. That intro music you hear is actually beatboxing, and it's coming from Cory Honeyman, our Associate Director of GTM Research's solar practice. Hey, Cory.

Cory Honeyman: Hey, Stephen.

Stephen Lacey: I'm in Boston, seated next to my regular co-host, Shayle Kann. Shayle, how are you?

Shayle Kann: I'm great, Stephen.

Stephen Lacey: We didn't invite Cory on to entertain you with his beatboxing, although it is very entertaining. He's here to fill us in on the latest with U.S. community solar. He and his team have been tracking this space for a few years now and their numbers show that community solar is gaining traction. Last year, installations surged from 125 megawatts in 2015, to 343 megawatts. By 2019, the yearly market will be half a gigawatt. That means 20-25 percent of non-residential installations in the U.S. will be of the shared variety.

But take more than a cursory glance at community solar and you will quickly see complicated and lumpy the market is. The complexity of program design, project structure, and customer acquisition make community projects really hard. In today's conversation, we're going to unwind that complexity and look to the future of when community solar actually scales sustainably, if it scales sustainably.

I've got to say, a few years ago, we really started paying attention to this space, and at one of our conferences, I think it was Solar Summit, we had Paul Spencer from Clean Energy Collective on stage for a community solar panel. They talked about how difficult the project structures were. This was when Colorado was just getting off the ground; They were operating only in Colorado. I came away form that conversation thinking, "I do not envy any of the people trying to crack this market." It is, three years on, still as complicated as it seemed in 2013 or so.

Cory, tell us about why this market is so complicated and why it's so policy-dependent, because I think that will help set up the rest of the conversation.

Cory Honeyman: Sure. Yeah. I think three years ago, you wouldn't envy anyone trying to get into this space, and I think there are still a lot of questions about the long-term viability of community solar that make it still equally challenging for a lot of developers trying to scale up within the segment.

In the community solar market, in particular the states that are driving it to date, it is one that is still very much this policy construct: You have to have either legislation that creates community solar out of thin air, or you have utilities that are voluntarily rolling out community solar programs. Because you don't have this cross-cutting policy mechanism, like say net energy metering or, in most states, not having virtual net energy metering, community solar is that much more policy-dependent on every single aspect of the process, from project development to bill crediting to managing the interconnection queues that allow for community solar in the first place. It is really fragmented on a state-by-state basis, fragmented on a utility-by-utility basis, and so because of that, it has really become a market that is just taking off, but still one that is very much policy-constrained.

Shayle Kann: Let's step back for a second. Community solar at its core is pretty simple: You build one larger-sized solar farm or solar array and then you sell off little chunks of it in one form or another to individual ratepayers, be they homeowners or renters or even businesses. Right? You're just sort of taking one project and selling it off in little pieces. Why do you need any legislation or regulation to do that? Why can't you just build the project and then go sign a bunch of deals with customers?

Cory Honeyman: You could do that, but the question is: What are you actually selling to customers if you're trying to create a community solar project outside of legislation? Right? I think one of the key elements that has enabled community solar to date is the fact that you have legislation that basically ascribes a value to the kilowatt-hours generated by those projects at, or close to, full retail rate. You create an equivalent net-energy-metering policy mechanism for community solar that has enabled project economics to pencil out.

Shayle Kann: You can, theoretically. Right? Let's just start with this: You could theoretically build a 2-megawatt solar project and then go to a bunch of different customers and say, "OK. Here's the deal. I'm going to give you this share of that project and I will pay you out based on the revenue from that project. That project sells into the wholesale market and you're going to get a little cut of it." That generally is called "crowdfunding." Right? There have been companies that have tried crowdfunding over the years.

What is unique, as I understand it, about community solar, is that the credit that you get, if you're a customer and you've signed up for community solar, comes right on your bill. Somehow, this has to get tied in with your utility bill that you're already getting, which is why it gets all mired in this regulatory legislative or utility world, because you can't do it without giving a credit directly on the bill.

Is that right, or is there something else that's distinct about community solar versus crowdfunding?

Cory Honeyman: I think that's right. With crowdfunding, it's not so much you're getting a benefit that is savings. I think it's usually more talked about as, "You're getting some return on your investment when you do a crowdfunding approach." Right?

Shayle Kann: Right. So it's not tied to your electricity bill.

Cory Honeyman: With community solar, the cross-cutting trait is the fact that you're getting some credit on your bill for subscribing and participating in that community solar project. That's true.

Shayle Kann: Right. Which makes it distinct in terms of then how the customers think about it. Right? If you're going to sign up with one of the crowdfunding companies, you're thinking about it as an investment. You're going to get some rate of return on that investment. Whereas, if you're signing up for community solar, you're thinking of it as savings on an electricity bill.

Cory Honeyman: Right. No, that's true. I think one of the important features about community solar, compared to, say, crowdfunding or a traditional green tariff program, is the fact that you do have this tangible financial relationship and also bill credit relationship with a community solar project. The structure of it is such that you have a real relationship between that project, whereas with crowdfunding or green tariff programs you're basically participating in a program where you don't have a tie to a particular project. I think that that brings up some other challenges, too.

Shayle Kann: Yeah. What you're talking about with a green tariff program, this has existed for a long time. I think my parents did something like this like 20 years ago in Wisconsin, which was simple: You were just buying RECs, and as a result, you could call your electricity "green" and it came at a premium on your bill, whereas obviously the big distinction with community solar is it's supposed to engender savings as opposed to paying more.

Now we're in this world of, you want to do it on the bill and it's going to be tied to a specific project, so you have to somehow connect the customer with the project. Now you're in this world of, the utility has to be involved one way or another. The way that you and team have sort of broken out the community solar market is into these two segments: There's a utility-led segment and then there's a third-party-led segment. Can you describe the difference between those two?

Cory Honeyman: Yeah. The defining difference between utility-led community solar and third-party-led community solar is who really manages the subscriber acquisition and overall relationship management process with those customers. In utility-led community solar, that is typically owned by the utility. The utility is doing the direct marketing and sales efforts, and also managing the customer retention process. Some of that could be outsourced to a development partner, but that is usually owned by the utility. Whereas in third-party-led markets, this is often times driven by legislation, but the cross-cutting trait is that you have a company, like say a Clean Energy Collective, that is owning the customer acquisition process.

Shayle Kann: From the utility perspective, presumably, they're always going to want the utility-led version of this. They want to own that customer relationship for as long as they can.

Cory Honeyman: Yeah. I think any third-party-led program to date has been created by legislation. No utility is voluntarily saying, "Hey, why don't you developers go out and own this new relationship with our customer base." I think that is something where they want to use community solar as a way to strengthen their relationship with their retail base, not to outsource that to another developer.

Stephen Lacey: Is there a lot of resistance in regulated states where these programs are being drafted in legislatures? You're effectively creating a backdoor hybrid deregulated market in which these other energy suppliers can come in and take your customers.

Shayle Kann: I think it's more complicated than that. Right? On one hand, you've got utilities, a lot of these utilities in the regulated states are facing some form of a threat in the form of rooftop solar, so they view community solar as a little bit closer to what they're familiar with. Right? They still get the lines and wires fees on these projects. So they're involved in a community solar project, even if it's a third-party-led project, more than they are if you or I put solar on our own rooftop. On the other hand, I think you're right. They don't love the idea of somebody else taking over a customer relationship and, like you said, "backdoor" competitive retailing them. I feel like utilities have a very complex relationship with community solar.

Cory Honeyman: Yeah. I would agree with that. I think this brings up the major theme that we covered in our report where we were tracking some of the major trends defining community solar right now, is that the love-hate relationship all comes down to how the community solar program has ultimately been designed. Right? There are a number of headaches and regulatory debates that have emerged in the community solar market that have ultimately pitted solar industry stakeholders and developers against utilities in the implementation of these programs.

Shayle Kann: Community solar has all these basic things going for it in theory. Right? When you want to describe why community solar is attractive, it's very easy to come up with a bunch of reasons why community solar could be, and arguably should be, one of, if not the biggest market segment for solar in the U.S. You have a much larger total addressable market of customers than you do for rooftop solar. You've got utility-scale project economics with retail customers, which means there should be some opportunity for arbitrage there. You've got the benefit of if a customer defaults or a customer moves from their home, you don't have to go remove a system. You can just have a waiting list and cycle in new customers.

All these things make community solar sound so great. Why is it that the market has been so slow to develop? Is it all just regulatory delays?

Cory Honeyman: Yeah. I think you can bucket it into two sets of risks. I think regulatory delays are more relevant for that third-party-led set of markets. A lot of the major headaches that happened in Minnesota are things that are relatable to other markets that are driven by third-party-led programs, but then a whole other set of risks and challenges in that utility-led segment.

On the utility-led side, all of the things that you just said are really attractive in theory, but they haven't been proven out yet. So try telling any of those value propositions to a financier looking to provide capital for a portfolio of community solar projects, and they say, "Hey, that's great in theory, but have you actually proved out that concept yet?"

Shayle Kann: Right. I just mentioned one of the benefits which people talk about as a reason why community solar should be eminently financeable. Let's just say you're going to build a community solar project. You need a hundred customers for that project. You sign up 120, you put 20 on the wait list, and then the second that one of those first hundred defaults, you just swap in another one. Very exciting. Means that it should be more financeable, arguably, than rooftop solar should be. On the other hand, you're signing, what, a 20-year contract with these customers. There's no guarantee that your remaining 20 are still going to be there and ready to sign up at the time when you need to swap them in. Could be 10 years from now and they've forgotten about it entirely.

It sounds great, in theory, to finance. Actually difficult in reality. Have you found that the community solar that is getting built, is it getting financed by the traditional solar project finance providers, or is there a different class?

Cory Honeyman: It is being financed by a subset of the traditional class of institutional investors and banks that have supported rooftop solar and utility-scale solar. This actually brings up one of the defining questions for community solar moving forward, is that is community solar a project and product offering that is either very similar still to rooftop solar and utility-scale solar, or does it really become its own segment altogether, in terms of the types of projects and the types of product offerings that are there?

In a lot of third-party-led markets, the reason why you have been able to attract institutional investors and banks is because the product offerings, the policy mechanisms that are being leveraged are very much similar still to rooftop solar. You have 20-25-year leases or PPAs that are being pitched to subscribers, much like rooftop solar. On that side, the reason why a lot of community solar projects still look and have a feel very similar to rooftop solar is because they haven't been able to prove out some of the differentiators of community solar as a business model. If you've only installed 5 megawatts, how do you prove that you can actually manage a pipeline of customers and manage customer retention and turnover? It is a bit of a wait-and-see game in the same way that we saw rooftop solar five or so years ago. Right? I don't think any investor would have been clamoring for an asset-backed security in rooftop solar back then, before you could actually prove out customer default rate risk or other fundamental risks that needed to be proven out several years back, for rooftop solar.

Stephen Lacey: This brings us to the customer retention piece and the customer experience. What do we know so far about what customers think of the experience, what kind of acquisition strategies are working, and what customers are demanding, in terms of flexibility for a contract length?

Cory Honeyman: Yeah. With community solar, there are different set of expectations for traditional, like C&I and investment-grade municipal customers, versus residential, much in the same way you think about different expectations in the traditional residential rooftop market versus customer-sited C&I. Right?

With residential subscribers, more often than not, they're looking for savings that are still similar to what they could get with a rooftop solar loan or lease. One of the things about community solar is because it's such a new segment, a lot of the customers that are being targeted, in particular, third-party-led markets are the same ones that SolarCity is targeting or Vivint is targeting, because financiers, whether it's community solar or rooftop solar, are looking for customers with a minimum FICO score of 680 and having all the other traits that they look for in rooftop solar, because everything about the product offering for community solar is still too much like rooftop solar, at least in a lot of these major markets.

Shayle Kann: Yeah. It's funny. It's like, there are all these things, these benefits community solar should theoretically have, one of which being you shouldn't need as high a FICO score, you shouldn't need to be a homeowner, etc. Another being, in theory, shorter contract lengths. You should be able to offer customers a three-year contract instead of 20, and then cycle other customers in. All these things should be benefits of community solar. We're just not at the point in the market yet when anybody can finance that stuff.

Stephen Lacey: Right. But we have seen people try to develop programs that pool that risk. You bring and you say, "Okay. A quarter of customers have to be low-income customers." You can kind of pool together C&I customers, regular residential customers, and low-income customers. Right? We've seen attempts at pooling that together, the question is can you get the money to support it?

Cory Honeyman: Right. To your original question about what questions are actually getting out of community solar deals right now. Right? Just to be clear, for residential, it is typically anywhere between 5 and 15 percent year one savings. What works more often, not surprisingly, would be still being able to offer year one savings north of 10 percent, having that double-digit-savings value proposition.

For C&I customers, depending on how big of a portfolio of projects those anchor tenants can subscribe to, being able to subscribe to a portfolio of community solar assets allows for somewhat slimmer savings. I think for C&I customers, offering as low as a fixed 1 percent discount on their bill has worked in a market like, say, Minnesota, and then up to, I would say, still that 10-15 percent year one savings in some of those other major markets that we're seeing.

Shayle Kann: You've alluded to C&I customers a couple of times. I think it's worth just drawing a finer point on that, because a lot of the time I think the messaging around community solar makes it seem like it's just a residential thing. The reality is that a lot of these projects, in order to get financing in part, have what you said, an "anchor tenant," which would be one or two large customers that take up a big chunk of the production from a single project and are credit worthy, so that you could go finance off of them, and then you fill in the rest with the residential customers or the small commercial-

Cory Honeyman: Yeah. I think it's an important point to note, that the reason why community solar providers have had to target such cookie-cutter customer segments, like high FICO score residential subscribers and investment-grade C&I or public customer entities, is primarily because of how nascent community solar segment is. Right? For a lot of investors, you still have to explain what is community solar, even if they have invested in rooftop solar or utility-scale. So there's a question of how much has to be deployed for investors to get comfortable with understanding what community solar is, to branch out into more underserved customer segments, and what other things need to be proved out? Is it really just scaling the learning curve, or are there other key questions that are going to really unlock demand and capital for community solar?

Shayle Kann: I've had sort of a hard time figuring out how community solar fits in to these visions of the utility of the future and the grid of the future. I know New York is probably the place to talk about this, because New York both has the REV initiative, which is like the perfect emblem of that conversation, and a community solar legislation, which I think is wrapped into REV now. Right. The whole thing about REV and other initiatives is they're all about ultimately enabling consumer choice, which seems like it would support community solar, but also just like putting the right assets in the right places and pricing them accordingly.

That's where I feel like community solar seems like a weird fit to me. If there's a place on the grid that it would be really valuable to have a two-megawatt solar project, REV should solve that problem via locational pricing for that project. But then you take that project and you sell it to a customer who's not in that location, doesn't that sort of break this whole system that we're designing?

Cory Honeyman: Yeah. Maybe taking a step back from what you're saying here: New York is another prime example of why community solar is so hard to scale right now, because for every single state there are unique police and regulatory risks that you have to navigate. I think one of the things that I had heard from a developer a while back is that one of the key components in building out a good software platform for community solar is actually having a regulatory management tool where you have clear understanding of different regulatory elements that you have to account for in your bill-crediting management and subscriber retention and things along those lines. New York is another good example of this.

I think that, to your question about where does community solar fit in this next set of questions and evolutions of the utility business model, it can still fit in that. Right? The value of community solar is still based off of its location on the grid, so the benefits that you then assign to customers are more about the location of the project, not about the location of that customer. Right? You can create rules where a subscriber has to be in the same load zone, or other more location-based relationships to that project.

But I think community solar is, for lack of a better phrase, an awkward segment that you have to fit into this conversation about utilities 2.0. It is not a distributed energy resource asset in the conventional way we think about it, in terms of it being customer-sited.

Shayle Kann: Can we talk a little bit about what markets are coming up next? The early days of community solar were largely, what, Colorado and then --

Cory Honeyman: Just them.

Shayle Kann: Just basically Colorado. And then now Massachusetts and Minnesota coming up. There's legislation that has passed that should create a burgeoning market in New York and Maryland. Is there anywhere else that's important enough to mention in that conversation?

Stephen Lacey: California.

Cory Honeyman: Technically, yes. Although, that's a market that has its own set of complicated policy debates right now. But yeah, California is the sixth major market that we profiled in the report.

It's worth noting that the five markets that you initially mentioned are basically the only state markets that are going to drive a majority of annual installation volumes for community solar for the next several years. We look at those five states, not to repeat, Massachusetts, Minnesota, Colorado, New York, and Maryland as driving over 90 percent of our third-party-led community solar outlook. There's a few other states to watch in that segment, that towards the end of this decade, or closer to 2019 and 2020, that we expect to pick up steam. Specifically, Oregon's market, Hawaii, and then also Illinois. Those three states are the only other ones that we're looking at that represent specifically triple-digit-megawatt programs or uncapped development opportunities.

Shayle Kann: And those are all legislated.

Cory Honeyman: Exactly. Yes.

Stephen Lacey: That actually transitions us well to the bigger question, and that is: Are these programs ever going to stand on their own, without legislative fiat? Is it just always going to depend on the government carving out a certain segment and saying, "This is how you approach customers. This is how the program's gonna work," or are we gonna get third-party-led projects that are just developed outside of any legislation or utility partnership.

Shayle Kann: The utility-led ones, you don't necessarily need legislation or even a lot of regulation to enable it. This is a third-party developer or third-party service provider, software provider, partnering with a utility, and then the utility says, "I want to roll out this program," gets approval from their regulator. Right? You could theoretically imagine that happening en masse.

Cory Honeyman: Yeah. I would argue that there are three buckets to think about, two of which exist today and one of which is only talked about, basically, in theory at this point. The first one, which we spent most of the time talking about thus far, is markets that are third-party led and created by some legislation. Those are ones where it involves a on-bill credit that is determined specifically for community solar projects per rules created by a legislature. The second one is how we think about how most utility-led community solar projects are created, where a utility voluntarily pushes for a program and then, on its own, decides what the value should be for projects that they procure and then market directly to customers.

The third one is, I think it's somewhere in between, it's still a project that is initiated and marketed by a developer, but doesn't hinge on a legislation or a utility-administered program to assign value, for community solar. This is where you see this bridge between wholesale markets, competitive retail suppliers, and community solar as we traditionally thought about it to date.

What some folks are thinking about, and the next wave of community solar business models, is repurposing what we traditionally think about competitive retail supply contracts that are either a hedge or a slight discount to a customer's generation component on their bill, and marketing a community solar project to be that hedge or discount to the generation component on their bill.

Stephen Lacey: Still sounds confusing if I'm a customer.

Cory Honeyman: Yeah. To me, too.

Shayle Kann: That is one issue with community solar: It's good in that if I'm a homeowner, I'm a little scared of somebody coming and drilling into my roof and putting a solar up there that maybe I don't know if I can trust or not, and community solar doesn't have that. On the other hand, you have to explain it in a way that doesn't sound like vaporware.

Stephen Lacey: Let's revisit the forecast again: By 2019, you're talking about a half-gigawatt market, yearly, which is pretty substantial. It represents about a fifth of non-residential installations. We're going to see pretty sizeable market.

Shayle Kann: Yeah. A sizable market, but not a whole lot of growth. 2017, we're predicting will be 400 megawatts of community solar. So from 400 megawatts this year, to 500 megawatts two years from now. Relative to what we generally expect from solar market growth, that's actually pretty incremental.

Cory Honeyman: Right. The next couple of years are when community solar hits a new level of maturity. Right? Last year, the market grew fourfold over a small base: 2015 was only a little over 50 megawatts; 2016, we're at 218 megawatts. 2017 is the last year where you see the market growing off of a much more manageable base, to scale quickly.

Over the ensuing years, there are a number of reasons at the state level that lead to incremental growth, but you can really just point to two things: one being the regulatory-driven boom and bust cycles of a lot of the major state markets we've talked about, and then second, this incremental maturation in the types of voluntary community solar programs. This evolution from smaller, sub-one-megawatt individual projects comprising pilot community solar programs, to both a lot of rural electric co-ops and, increasingly, invest-your-own utilities, rolling out more mature multi-project megawatt-scale programs.

Stephen Lacey: Cory Honeyman is Associate Director of GTM Research and Solar Practice. He joined us to talk about his new report that he put together with he and his team: U.S. Community Solar Outlook for 2017. Thanks for joining us, and I guess you can take it away and drop the beat.

Cory Honeyman: Sure thing.

Stephen Lacey: And I'm Stephen Lacey, joined by Shayle Kann. This is The Interchange from Greentech Media. We'll catch you next week.