PFR Solar ABS Roundtable 2019/20
To download this report as a PDF, click here.
Note from the Editor
Originators of residential and other small-scale solar assets have strived for years to cultivate a reliable, efficient, liquid market for solar asset-backed securities, and their hard work is beginning to bear fruit.
From the beginning, residential solar loans, leases and power purchase agreements appeared to have all the attributes of an asset class ripe for securitization. What’s more, they have an obvious appeal in the era of ESG investment criteria.
In the early 2010s, issuers, lawyers and bankers put in the hours to structure and market the first solar ABS bonds to investors, and SolarCity launched its debut offering to much fanfare in 2013.
Since then, primary market volume for solar ABS has grown to roughly $6 billion a year, though it is perhaps best described as several markets rather than just one. Besides the more public 144a bond format, there are many private deals. Then there are the transactions that could be considered securitization-adjacent, such as warehouse facilities, forward-flow agreements and sales of portfolios of whole loans.
Having educated investors about solar resource risk and tax equity partnerships and lived through challenges to net-metering, market participants are ready to consider the next evolution of the product—its expansion into the realm of commercial and industrial-scale projects and—who knows?—maybe even utility-scale solar and transmission.
Another test that solar ABS has yet to face—a major recession—lurks at some unpredictable point in the future. Nevertheless, deal watchers are monitoring the market carefully for the erosion of investor protections that may seem unnecessary in the good times but that bondholders will wish were in place if things go sour.
Participants (clockwise from left):
Matt Eastwick, Chief Investment Officer, CleanCapital
Eric Neglia, Managing Director and Head of Consumer ABS, Kroll Bond Rating Agency
Stephen Henne, Director, Securitization, KPMG
Benji Cohen, Founder and CEO, T-REX
Richard Metcalf, Editor, PFR (moderator)
Spencer Hunsberger, Managing Director, Securitized Products Finance – Energy, Credit Suisse
PFR: Let’s get a general overview, to begin with, of how the residential solar ABS market has performed since it kicked off in 2013. We could start in terms of primary issuance.
Spencer Hunsberger, Credit Suisse: I had the team pull together some numbers in anticipation of this. What we were able to find was—and this is talking about public deals; there are private deals that are in addition to this as well, and the private deals could be in loan format, ABS format or otherwise—we see 30 deals [as of the fourth quarter of 2019], excluding a transaction that was part home improvement and part solar loans, and excluding the PACE side of the business. And that’s split roughly 50/50 between third-party-owned lease-and-PPA business and the solar loans side. Across the 30 deals, there’s $6.3 billion of debt that has been sold, not counting residual interests that were retained by sponsors or by third parties. And that was $3.5 billion of lease-and-PPA volume and $2.8 billion of loan volume. In addition, there is the private market, which is either done through 4(a)(2) private placements or loan syndications that might not have been picked up in these statistics.
PFR: Does anyone have any idea how big the private market is?
Hunsberger, Credit Suisse: Between loans and 4(a)(2) placements, reported closed deals would be between $1.5 billion and $2 billion of total issuance.
Ben Sunderland, Vivint Solar: That sounds about right. We’ve done syndicated deals, for instance when we closed our first securitization in June 2018, where we not only did a public deal, but we did a private deal as well, with the help of Credit Suisse. The private deal was actually almost equivalent in size to our public deal.
PFR: Was that a 4(a)(2) deal or a loan?
Hunsberger, Credit Suisse: That was a syndication of a warehouse facility. I think the point is that there is another part of the market that is both (1) helpful for sponsors when you think about flexibility and developing markets, and (2) helpful to vary the types of access that you have into the traditional market as well. Not every investor might be a 144a buyer. They might want a duration asset or a loan asset or something else. And this is all beyond the bank loan market.
PFR: So, what’s the rationale behind perhaps doing two tranches? You just have a wider pool of investors that can look at it?
Hunsberger, Credit Suisse: It could be a wider pool of investors, or it could be flexibility on timing and closing and then actually transferring that to the market in the future, versus having one or a concentrated number of holders for a period of time. There are a variety of reasons. But what we see a lot more of is, when you think about issuers that might be trying to balance their duration profile, the ABS markets tend to be more of a seven-to-ten year ARD [anticipated repayment date] type market, and when you go to an institutional loan format or 4(a)(2) format, that could be a very long tenor, 20 to 25 years. So it’s not just how you go to market and whether you’re doing a public deal or a private deal, but actually the structure of the investment itself.
Benji Cohen, T-REX: A big part of it, though, is also the sophistication of investors. Over the last six years, since Credit Suisse did the initial SolarCity issuance in November 2013, you’ve had institutional investors who have become more familiar with the asset class. There’s more data around the asset class, around the performance of these entire portfolios. And so some of them are setting up direct deals, whether it’s 4(a)(2)s or whether they’re syndicating or doing forward-flows. There are some would-be issuers who don’t show up in this at all. Sunlight Financial is one of the top originators of residential solar and they’ve never done an issuance that would show up in any of these statistics. But where’s the capital going? They’re not holding it long-term. They’re selling it on to institutional investors directly. For the reasons that Spencer said, these stats aren’t public, but I would say that in 2019 there’s at least 75% of volume that is being done directly, and not in the 144a context.
Eric Neglia, Kroll Bond Rating Agency: I can echo that. We see the majority of major participants accessing the 144a securitization market. Mosaic issued two, Dividend Solar issued one and Sunnova issued its inaugural loan securitization in 2019 as well. But, to your point, I think what we’ve seen is that they do have forward-flow agreements with several banks and credit unions who are buying the whole loans and either holding them on their balance sheet or looking at some sort of long-term private financing that never enters into the capital markets.
Stephen Henne, KPMG: We think that the execution that our clients are seeing on some of their 144a securitizations is improving, which makes it more attractive, but obviously there are still some headwinds. We’re bullish on securitization issuances going forward.
Sunderland, Vivint Solar: Vivint Solar got into the securitization game a little bit late. SolarCity did the first one on the PPA/lease side; Sunrun dabbled in it a little bit in 2015; and then we jumped into it in 2018. I think Sunnova was in 2017. And the way we looked at it was, the market just never had enough depth and so pricing wasn’t very competitive. But I think what you started to see change is the rise of solar loans. Solar loans created significant deal flow, and so then, all of a sudden, solar loans—which are a more traditional product for consumer finance ABS buyers—brought a lot of investors into the space. And then, as they got comfortable with the solar loan asset-backed securities, they realised: ‘Oh! They’re not that different than the PPA and lease assets.’ And so then the PPA and lease assets started performing better, and that’s what I think propelled the two Tesla transactions in 2017, plus the Sunnova transaction in 2017. That’s what gave us pause, and we said: ‘We should re-evaluate this market; there seems to be a lot of positive momentum there.’
To me, SolarCity helped kick things off, but we wouldn’t be where we are today without the help of all the volume that the solar loan guys helped create.
Neglia, KBRA: To that point, we are seeing strong demand from investors. One case in point is Mosaic’s most recent 2019-2 transaction. It was a $208 million deal, significantly oversubscribed, to the point where their senior classes, which KBRA rated AA- (sf) and A- (sf), were oversubscribed and spreads came in tighter than expectations. The junior notes, the Class C and D, priced at the low end of the guidance. That better pricing created additional excess spread for the transaction, which allowed us to notch up the ratings of Class B, C & D between initial announcement and closing.
PFR: Interesting. Moving onto the secondary market, we now have outstanding bonds dating back to 2013, unless those have been refinanced. How have they performed? And is that tied to the performance of the underlying assets?
Neglia, KBRA: I can speak to the performance of the solar loan collateral on the bond side. What we’ve seen is that losses have been coming in quicker than expected. We don’t think it’s a cause for concern, where KBRA is going to raise our base case assumptions, at this time. We think it’s just a frontloading of the curve. We do see an increase in losses around the ITC payment date, when some of these loans reamortize, and may result in a higher monthly payment if the borrower hasn’t paid down at least 30% of their original loan balance. Borrowers are deciding, for a variety of reasons, to default on their loan payment. So we’re monitoring performance, but there’s nothing that’s causing us to increase our lifetime loss expectations at this time. Instead we feel it’s more appropriate to shift our timing curves.
Cohen, T-REX: I completely agree with that. I think that that’s the majority of the answer around performance. What’s interesting for us is to look at different loan products, and also look at loan products versus PPAs and leases, and see how they perform. Loans have only been around for a few years, in bulk. How are those performing over time? What do the different originators of those loans do? What is the capital markets activity? How are institutional investors looking at exactly this, at your 18-month flip, when you have a 40% balloon that’s meant to mimic the latest that you can get the ITC in. How does that look? And we’ve seen big differences in private data, between different originators, on losses in a range of half a percent to 1.5%, which is pretty big. And it is determined in no small part by products and, to some extent, credit methodologies at origination.
Hunsberger, Credit Suisse: Yes, I would agree with that. There’s some helpfulness in separating out loans from third-party-owned markets, and then, within each of those markets, separating out issuers as well. One thing that has become more apparent, and might even be overdue, is some differentiation of specific issuances in the market, from a pricing perspective. Because the pricing on the primary side is being driven a lot by overall market trends—whether the credit markets are healthy, whether investor demand is there. A lot of them are binary points that Ben touched upon earlier. And now we’re seeing people saying: ‘We’ve got this data on this issue over this time period, and that’s proving out a certain thesis, and this data from this other issuer over this time period might have a slightly different thesis.’ It doesn’t mean that one is bankable and one is not, but it does mean two different sets of assumptions.
It’s something that the team at Kroll is certainly trying to take into account as well, and we’ve heard feedback from Eric Neglia and his colleagues about how that’s going to flow through.
The other thing that’s interesting, if we move on to how trades have performed in the secondary market, is that when we looked at public data on closed loan transactions, what we found is that everyone has slightly different expectations—base case—of what loan defaults would do and what that curve is going to look like. Overall, losses might have come in faster on some asset classes, but, overall, the initial projections have been accurate across the industry as a whole. But one of the things that gets missed is the correlation between prepayments and cumulative defaults. If your prepayment speed comes down, because fewer people are prepaying loans around the ITC date, even at the same constant default rate, your cumulative defaults go up. You have a longer duration investment. I think that part sometimes gets missed.
Although we’re seeing the model of cumulative loss go up, the constant default rate may be staying the same. And the duration of the bonds are extending, because prepayments are lower than expected. They’re two separate points, but they get conflated in the cumulative default projections.
Henne, KPMG: One of the reasons we’re bullish is we’re seeing our clients securitizing more and more. Securitization’s very much a snowball rolling downhill—it’s gets easier the more you do it. Investors are getting an appetite, they’re getting comfortable with your assets, what you’re putting out there. But also, if you can bring years of data that says: ‘My delinquency’s less than 0.1%’—as reported on certain Sunrun transactions—‘and when we transfer an asset from one borrower to another, our recovery is over 100%’—I think that gives comfort to the buyers.
PFR: For the ABS market, the solar deals have been on the smaller side, in terms of the size of the individual deals. Does that impact liquidity? Has there been much trading in these bonds, relative to other asset classes?
Hunsberger, Credit Suisse: The deals average $200 million to 300 million, when you look across all of the industry together, which I actually don’t think is small. It’s certainly small relative to mortgages, but versus other esoteric asset classes, it’s a pretty good size and we do see efficiency at that size.
To the second point, it’s still not an actively traded asset class in the secondary market. There is a market. Bonds do trade. We were able to find TRACE data on four or five bonds that have just traded recently. And those are telling a lot of the story that we were just talking about.
The overall market looks healthy. At the secondary level, trading is usually inside the primary levels, and the primary levels are 205 to 210 basis points over, when you look at the most recent issuances on the lease-and-PPA side, and slightly inside that on the loan side. We see the secondary market doing what you should see in mature asset classes, which is secondary trading inside of the primary. But there is differentiation: Sponsor differentiation, specific issuance differentiation. If you have a bond go on watch and come off, it may trade behind its peers.
Cohen, T-REX: We’ve seen some trades in the secondary market for some of the oversubscribed bonds, especially for first-time issuers. There isn’t always as much information getting out, but some of it is the drama of a new issuer coming to market in a relatively new asset class. We’ve seen a few funds who have taken those bonds and flipped them very quickly, which is good—that there’s secondary market appetite—and not as good—that these guys are not holding them. That speaks to lack of maturity. I don’t think it’s a huge issue that’s rife throughout this market, but it’s an interesting one to note, which speaks to availability of information and information flow.
PFR: We’re comparing this market to other esoteric ABS asset classes. Is that where solar ABS is going to stay, do you think, or will it break out of the esoteric category?
Hunsberger, Credit Suisse: I could be accused of being biased here, but I think it is breaking out and it has broken out. When you include the private stuff, this is a pretty healthy, growing market. I don’t know whether there’s some line in the sand when you have an asset class that’s its own asset class as opposed to an esoteric asset class, but if you’re getting to the point where you’re averaging nearly $2 billion of issuance a year, if you got that up 50%, you’ve got $3 billion of reported data a year—that’s an asset class. People set up teams to do just this asset class outside of traditional esoteric investments. And the other thing that’s interesting, when we talk about the comparison between a traditional public ABS investor versus a private investor—you’ve got infrastructure, pension, other long-dated type investment money that’s coming into this asset class that, perhaps, was slower to come in. While the initial momentum came from crossover accounts on the consumer side, once you start seeing buying that could be compared to other infrastructure asset classes in the D.G. [distributed generation] space, that could be compared to other project-financed, long-duration investments, then there’s all the more reason to think that interest in these assets is broadening from a specific bucket of ABS into a broader range of buckets including ESG-focused, long-dated consumer-focused, project-focused.
PFR: From the sponsor’s perspective, how does that affect how you think about how to finance these assets?
Sunderland, Vivint Solar: When we entered 2018, we did not have any plans to do securitization, but as we started to assess some of the deals that had closed at the end of 2017, we started to evaluate that option more seriously. And after we closed our first securitization, in June of 2018, we overhauled our perspective and capital markets financing plan.
As a result of the execution in the market and where we think this market’s going to go in the long term, we recently closed a private temporary warehouse where we’re placing our assets until we’ve accumulated enough volume to go to the securitization market. We’ve structured it in a way to make it very easy to just take those assets and drop them into a securitization.
In the past, we were more focused on the bank market, where they’re not doing the same type of diligence. A good example of this is that in the securitization market they have a custodian that’s going to review literally every single customer contract. That’s not something that happens on bank loans, per se. So we’ve completely overhauled our internal processes to make that very easy. We plan to continue to go to this market. We think it’s where the low-cost, high advance rates are, and we’ll continue to evaluate that. Things could change. Market dynamics change, but right now it’s one of our core focuses.
PFR: So the demands of the bond market and the ABS market actually affect the way your business is set up and the way you do business, to some extent?
Sunderland, Vivint Solar: Yes. Internally, we sometimes wish that it was easier to do certain things, but there is an amount of rigor that’s actually helpful to the business and the industry at large. Having every contract reviewed by an independent third party—that actually makes a lot of sense, when you’re talking about bundling up large sets of assets. And it forces us to be more accountable internally, as we evaluate our internal processes. The standards that are created, not only in the ABS market, but with some of our other financial partners, go towards making this asset class better and making it more financeable in the long run.
Some of the smaller players, who don’t have the same access to the capital markets, aren’t held to that same level of scrutiny. Because we’re held to a higher level of scrutiny, that’s really helped us build out our compliance team, build our supply chain, build out our customer management processes.
PFR: I wonder how many ordinary members of the public realize what a beneficial effect securitization can have on the business practices of a company. It’s not the story that they’re most familiar with.
Hunsberger, Credit Suisse: I really appreciate the sentiment.
Everyone around this table picks and chooses the parties that we work with and try to find like-minded people that want to take a long-term approach to the market. That approach goes a long way. It’s not about a specific issuance in a month or a year, it’s about how we build tools to build long-term companies.
The capital markets have been around for a long time, there’s certainly criticism, and some of that criticism is due. I’m not speaking about our capital markets space, but capital markets as a broader industry. There are things they do very well. There’s transparency and pricing transparency, and how things are going to be perceived by a broad pool of the investing public, and they pass useful data on to issuers as well.
PFR: Matt Eastwick, what’s your view?
Eastwick, CleanCapital: The core mission of my company, CleanCapital, is to attract additional institutional investors into the asset class. While we always look at traditional project finance lenders in the bank market or institutional fund market, all else being equal, the structured product investor in the securitization market is an interesting one for us, because they probably haven’t seen a lot of our assets in the marketplace. So we’ve designed our financing strategies to ultimately hit the securitization market.
In fact, we did a financing earlier this year with Spencer’s team at Credit Suisse, exactly with that intention in mind. We are aggregating a lot of C&I solar assets and once we get to a critical mass within a specific portfolio, then we’ll be able to evaluate these alternatives.
The discipline of working with a bank like Credit Suisse has been very helpful for our team to make sure that we’re thinking about everything the right way as we prepare for securitization.
PFR: And since CleanCapital’s assets are C&I—commercial and industrial—as opposed to residential, that would be a novelty for the ABS market, wouldn’t it?
Hunsberger, Credit Suisse: There have been approaches to the market over the past few years, there are some underway currently, and some that didn’t ultimately result in a transaction. I think there will be a flurry of announcements in the next six to 12 months in that space, when you think about transactions that have been reported in our industry media as being underway, and when you think about the volume of capital that has been attracted, from the sponsor side, to C&I, over the past 12 to 18 months in particular, where that was an asset class that had been particularly starved of investment relative to the pure utility-scale or pure residential D.G. for a long period of time.
PFR: When you're looking at commercial and industrial assets rather than residential assets, what impact does that have on the deal?
Eastwick, CleanCapital: Well, it's a different set of characteristics, and they are lumpier assets. Our typical project size is certainly many, many times larger than a residential solar asset, so the goal of trying to get the right level of diversification into the portfolio is more challenging, and that's been something that we've had to focus on. It’s probably one of the largest hurdles restricting the amount of C&I securitization volume.
PFR: Does the fact that the assets are C&I have any effect on the choice of which nook of the market to tap—whether it's 144a or 4(a)(2) or a loan?
Hunsberger, Credit Suisse: That's a good question. I think that you'll see a higher percentage of private deals in the first year to year-and-a-half of development here than you see in resi D.G. That's just because you're going to be working through methodology, generally, and you're going to be working through sponsor-specific considerations around their portfolio.
There isn't a direct map from utility or resi into this asset class. It's lumpier, from an exposure-to-single-credits perspective, and you have to think about rating those credits if they're not publicly rated or comparatively rated already. You have to think about a distributed maintenance plan, because you don't have one asset in one place with dedicated staff. It's a higher current, typically more complex system than in a residential home.
Getting back to Ben's earlier comment about why the lease-and-PPA, third-party-owned side of the residential market was able to grow quickly, it’s because we had corollary areas to borrow from. There are corollaries on the C&I side, but we're still figuring all those out right now. A lot of this is going to need to be figured out in the next six to 12 months, at which point in time we will see a healthy public market, a healthy 144a market.
PFR: There have been approaches to the market, so does that mean deal marketing sent to investors that did not result in a transaction?
Hunsberger, Credit Suisse: Yes, there was a deal a few years ago, that really would have been the first public C&I portfolio. If you look at reported media, there are also transactions in the market right now that have a C&I exposure. And this is in addition to transactions where C&I has been a small subset of the overall pool. If you look back, some of the earlier Tesla transactions might have had less than 5% of assets that might qualify as traditional C&I blended in, but when we're talking about new, standalone deals, it's 100% C&I.
PFR: Is that unusual, for an emerging esoteric ABS asset class to have hiccups like that?
Hunsberger, Credit Suisse: I don't think it's unusual. You could have the best-structured transaction, the best sponsor, the best assets, but if you launch it at a time when the capital markets are in a relatively risk-off mode, like we saw in the last month-and-a-half of last year, you might say: ‘There are better ways to finance this.’ When you think about ABS or capital markets generally as a tool, it’s a tool. It's not the only tool. I'm not going to come here and say it's the only thing you can do. Prudent sponsors might want to think about a mix between bank markets, institutional markets, and forward-flow arrangements, because those markets are bullish on risk at certain points in time and might be more conservative on risk at other points in time. Part of the sponsor debate is then how to think about blending those. It's not to say that because the transaction was offered to the market and pulled means that there's something wrong with the transaction.
Henne, KPMG: We have seen some of these deals come our way as well and looked at similar portfolios, and the concerns have been diversification of assets, lumpy assets, rated versus unrated offtakers, and, ultimately, several of the clients have decided they felt more comfortable leaving these assets on their books, at least temporarily, until they found another financing solution. They understood some of the risks, but the market hadn’t gotten to the place where they were comfortable with some of the things they’d have to look at.
Cohen, T-REX: I do think that there are examples where there was inadequate capital. SunEdison is a great example. SunEdison doesn't exist anymore. They were trying to take securitizations to market. They never made it to market. That wasn't the only reason. But capital then stayed on their books, and they had too much capital on their books, for which they didn't have liquidity, and so that was that.
That was years ago, and at this stage in the evolution of this market, which has been quite rapid, a lot more is coming. We're now much more in a place where C&I securitizations can launch. But there are necessary things that happen along the way.
PFR: Over the past six or seven years, investors have had time to get their heads around some of the things that make solar ABS bonds unique and different to other asset classes. Variable solar production, the regulatory regimes that affect solar installations in different states, and of course tax equity structures. How has the make-up of the investor universe grown? Who are the investors? Is it investors that just buy ABS bonds, or is it a wider or a narrower range?
Hunsberger, Credit Suisse: We need to differentiate the markets here. Let's put third-party on one side and loans on the other side. A lot of questions around production and tax equity are not going to be as relevant for the loan side, but there's probably more similarities than differences between the investor bases in those two markets. So, who's the traditional public 144a buyer? It's someone who is managing insurance or other institutional capital. The average team that we would work with is coming at this through a long-dated, consumer, unsecured asset class lens, versus a project finance lens. But then we do see infrastructure investors, pension plans being managed by an infrastructure investors, and traditional project finance institutional investors who have come and continue to come into this space as well.
Neglia, KBRA: One of the reasons the investor universe is growing is because a lot of the risks in the early days are being addressed. Some of the major risks that you have with a small start-up company are, do they have the operational controls and infrastructure to properly originate the loan, manage their installer relationships and price risk accordingly? As they mature and develop those processes, they are growing into mature, established companies. Another reason is the accumulation of historical performance data. One challenge in the analysis is how to develop loss curves from this data, since it doesn't neatly fit into the box of short-duration consumer loans. We’ve been able to rely on proxy loss forecasts thus far. As you start gathering actual data, you're able to create the front part of the curve. We’re also starting to see prepayment data come in, which helps improve the accuracy of our forecasts. Addressing some of those issues gets investors more comfortable, which helps expand the investor base.
Eastwick, CleanCapital: It’s always been a great time to be a lender to solar. The performance has been great for those who’ve got the experience and have done the work to understand some of the complexities and the structure. What Benji and his firm, T-REX, are doing with the fintech component of helping people understand and analyze is, I think, going to continue to accelerate the participation, so that everyone gets to experience how great it is to be a lender in this space.
PFR: If they approach it initially through a consumer finance lens, how hard is it then to incorporate everything else? If you're used to seeing consumer finance deals, you're not used to looking at solar production, for instance. How much of a challenge is it for an investor become confident enough to participate?
Hunsberger, Credit Suisse: It's good question. I was just talking about the average investor. I’m not saying everyone’s coming from a consumer lens, and, certainly, people came into it from an energy and project finance perspective.
The first deals took a long time, and they had very simple tax equity structures, where there really wasn't any cash diversion component. It's probably a single fund. You might have a cash grant transaction.
Where you started from was, what's the solar resource risk and how is that converted first into power and then into money. The first transactions allowed people to grapple with that, and they realized that the data was showing them that that is one of the most certain aspects of a diverse pool of assets. Versus a utility-scale transaction, where you might have significant P90 variability, this asset class has been characterized by expected production being translated into actual production.
And then they started saying: ‘Well, what's tax equity?’ We started doing transactions with multiple tax equity funds, and we started adding concepts like tax loss insurance to mitigate the risk that an investor could have their cash flows diverted by virtue of an issue at the tax equity fund level, regarding the eligible basis of the property. People got comfortable with that.
Something pops up every year. In 2016 it would be net metering. No one is really talking about net metering anymore, but for a period of time in 2016 everyone was wondering what Nevada was going to do. And a certain part of the market that, to Matt’s point, did some very good lending through that period of time, took the view that it was going to be challenging to retroactively change net metering laws, and when they thought about static pools of assets, that was probably a risk that they could price and take.
It's been an incremental process. It's been the better part of a decade of grappling with this and educating investors one step at a time.
The loan side then brought in a whole other pool of people who could come in and say: ‘There’s this transaction that's being done, and if I'm looking at this, then maybe I want to be looking at this other transaction as well.’ When you compare the highly-rated tranches of a loan transaction to other consumer credit, both from a pricing and a default perspective, or you compare a C&I pool and the financing you can do there to a utility-scale asset, there's an opportunity for investors to come in and say: ‘This is a very good place for me to invest my capital, and it's worth spending the time to go through these issues.’
Cohen, T-REX: That's a fantastic macro market perspective. There are different things that prime the market. Loans prime it for leases and PPAs. Resi primes it for C&I. You have belt and suspenders like tax equity insurance, which was a huge deal a few years ago. And now nobody is talking about it. It is not a requirement in every single one of these transactions because the market is mature, because investors understand it better and there's more data. And they also have more methodology, they have their own lens through which they can evaluate these assets, both at issuance and then on an ongoing basis in surveillance and potential secondary market activity.
PFR: So, if those things have all been dealt with—or maybe we are still in the process of dealing with C&I—what's the next hurdle for this market?
Cohen, T-REX: A down credit cycle, because we hadn’t had a major one, really, in the history of this asset class. As Matt was saying earlier, this is a great time to be a solar lender, and it has been, basically, for ten years. Solar lending has probably not even been around for ten years, certainly not en masse. But what happens to this—it’s a probably a bigger question for unsecured consumer lending, the fintech or marketplace lenders, but this is in a similar bucket—what happens to defaults then? What does a default payer look like then? We were talking about the first 18-month threshold, where you have a balloon payment. That’s one thing, but what’s going to happen if we go into a recession?
PFR: Well, what is going to happen? Eric Neglia?
Neglia, KBRA: It’s to be seen, but the companies originating loans for the transactions that we rated do have strong risk management teams. They’re underwriting every loan based on borrower, loan and P.V. system characteristics. They’re making sure that the technology and the P.V. systems are going to last the life of the transaction. But the ultimate performance is still to be determined. KBRA comes up with conservative base case loss expectations, applies conservative stresses and models the cash flows of each structure to withstand a credit downturn commensurate with the rating of each note. We’ll continue to monitor how it plays out, but the ratings we’ve assigned are supportive of the risk with each of the rating categories.
Sunderland, Vivint Solar: This is where you start to get into the value proposition to the customer, and this is probably one of the biggest distinguishing factors between a PPA and a solar loan. With a PPA, the average customer is saving some amount of money, up front, per year, whereas with a solar loan, the customer’s likely not going to be saving money until years 11 through 20. One of the key, foundational ideas on our side is that we are helping the customer save money, we’re not increasing their spending burden. If anything, we’re slightly reducing their energy spend. So in a credit downturn, if they stop paying our bill, that means their energy bill, overall, might be higher. We have a wide range of savings amongst our customers and some customers could be saving more than others, but that is the general thesis overall.
Neglia, KBRA: They look to savings on the loan side as well, especially as loan terms are extended. Longer original term ultimately lowers the monthly payment. We’ve seen system costs go down, causing the initial financed amount to be lower than what it’s been in the past, which makes the loan product more attractive for many customers. From the latest research we’ve seen, 40% to 48% of the market is being financed through a loan, rather than third-party ownership. Longer loan term and rising energy prices help to create a savings component compared to their regular monthly utility bill.
Hunsberger, Credit Suisse: Just two things to add to those really good points. One is that not every recession is 2008. Recessions are a pretty healthy part of the business cycle. Consensus is that we’ll see one in the next couple of years, so we should be ready for that.
The second is, let’s make sure we’re not making decisions in a boom-time environment that won’t survive some sort of downturn. It’s on everyone in this room and everyone that’s involved in the marketplace to ensure that the safeguards we’ve put in place over the past decade don’t get eroded just because they might not be demanded by the marginal investor on a specific deal. There’s a saying I like a lot, which is: ‘Before you take a gate down, figure out why it was put up.’ That’s relevant here. Why do we require certain documentation or legal protections? Why do we require documentation on how system completions occur? It’s because the data and the performance we’ve had so far, which has been extremely strong, was based on those things being present. It’s not specific to our asset class. It’s in 144a, it’s in 4(a)(2), it’s in term loan Bs. It’s something that you hear a lot in the market right now. When you think about energy finance as a global investment class, it’s present in gas-fired refinancings, it’s present in other types of LNG facilities—anything that you have right now where you’re operating in a prolonged credit bull cycle is going to have safeguards eroded, and it’s just a matter of making sure that we think through each of those. I’m not saying that there shouldn’t be improvements, but that we think through what improvements are warranted and we’ve got data around, and which we should say we actually need some more discovery on first.
Neglia, KBRA: I’ll agree to that. It starts at the finance company, the originators. We’re in a hyper-competitive part of the lifecycle right now. Companies are vying for market share, and origination volumes are growing. So we want to make sure their underwriting standards don’t deteriorate, they still have the same quality loan that they originally started with, they’re not expanding the credit box too much or they’re not failing to verify certain key components of the loan products themselves. We do see the FICO cut off going lower. Earlier on, FICO cut offs were around 650. So we’re making sure that those borrowers are underwritten in a diligent manner. We also see loan terms going out a little bit. Overall, we’re evaluating trends to make sure that originators don’t let asset quality erode just because times are good.
PFR: Going back to the beginning of this conversation, when we were talking about the different formats for residential and small-scale solar securitization, when sponsors and their financial advisers and banks are looking at the options, what are you trying to maximize? Is it the advance rate, the pricing?
Sunderland, Vivint Solar: It’s a great question. I’m sure it’s different for everyone, but as we continue to grow, Vivint Solar’s a very cash intensive business, especially on the PPA/lease side where we’re taking on the upfront cost and then the customer’s buying energy over time, as it’s produced. So we tend to focus on maximizing the advance rate, but we also are very focused on the cost. There were times when we did have to go to more expensive debt markets, but as we’ve continued to mature and grow as company, we’ve started to really drive down the cost of debt. When we closed our securitization in June 2018, we drove down our debt pricing by over 100 bp. So we’re continuing to focus on driving down our costs, but we are focusing on advanced rates first and foremost.
PFR: Do you rank the different products based on what offers the highest rate? Or, when you did a public and a private transaction at the same time, maybe you were able to finesse it and tap two markets in order to maximize the advance rate. Is that how that works?
Sunderland, Vivint Solar: There are also other things to consider. The first time we really entered the debt markets was in the summer of 2016. At that time, the only option to evaluate was the bank loan market. That was the only market that was fully established and had a high degree of execution. That’s one of the biggest factors we look at—the likelihood of execution and how quickly we can get that done.
So we focused on the bank market first. We then entered the private placement market shortly thereafter, and that was our first time working with Kroll. That was a privately rated deal. There is a very large private market. Vivint Solar’s done at least three private deals in the last three years where we got a rating from Kroll in a private setting. We’ve done another one where we have an implied rating.
PFR: What is an implied rating?
Sunderland, Vivint Solar: One of the beauties of working with Kroll, not to pump up Kroll too much, is they have a very open methodology. They explain exactly what they’re doing, and we’ve gotten very comfortable with that, as have all the lenders. So every time we’re doing a deal, I believe lenders now essentially do their own implied rating. They understand how Kroll would rate it. They run it through their models and say: ‘We think this is an implied rating of BBB. We’re not going to actually go to the rating agency for this, because we don’t need it to close this deal.’ But they benefit from having that transparency.
That’s one of the reasons why we’ve really enjoyed working with Kroll. That transparency has helped us plan better, because there’s no surprises. And if there’s any questions, they’re helping us understand the methodology.
Hunsberger, Credit Suisse: And what Ben’s saying there isn’t controversial. One of the main aims of the regulatory changes following the last cycle was that for the nationally recognized rating organisations, you have transparent methodology and the ability to compare data and data in and data out from across different rating organizations.
PFR: And the same question to you, Matt Eastwick. Are you looking for the same things as Ben just mentioned?
Eastwick, CleanCapital: Cost and advance rate are 1(a) and 1(b), for sure. They drive our return profile, the thing we have to focus on the most. But increasingly, the flexibility of the market and lenders is also quite important to us. Traditional project finance is very much a check-the-box approach, and that’s not how we look at a lot of these investments. The ability to take emerging investment considerations into account, or things that aren’t necessarily a traditional project finance consideration, is something that we pay very close attention to. And that’s the way the market’s evolving. This is not project finance 1.0. It’s going to be 2.0 and 3.0 going forward. And non-recourse financing is its own marketplace, it’s not just corporate lending. You’ve got to pay attention to all the traditional features of the project finance market, but there’s a lot of value in these assets and these investments that we want our lenders to appreciate.
PFR: Can you play these markets off against each other and go to the bank lenders and say: ‘Hey, we can get this advance rate and this pricing in the ABS market. What are you going to do about it?’
Eastwick, CleanCapital: There’s a lot of time and resource expenditure that goes into preparing these transactions. It’s not like you can just flip a switch and go from one market to the other. Doing the diligence and homework and listening to everyone’s opinion and advice is part of doing our job correctly.
Hunsberger, Credit Suisse: I agree with that. It’s part of the responsibility of doing a job correctly and completely. This is not playing people off each other, it’s not arbitrage. It’s totally the right process people should follow. They’re different markets.
If we require a custodial review of every single asset in a portfolio, but that means the advance rate goes up or the pricing comes down, certainly that’s a good trade off. If the bank says, we’ll give you more flexibility, but the advance rate’s lower or the pricing’s higher, that’s also a trade-off.
The companies and the portfolios that we’re talking about here are owned by private equity companies, owned by public markets, owned by yieldcos, owned by family offices. Those investors all have a different profile that they’re looking for in terms of how much upfront cash they’re raising and how much certain cash flow they’re going to get over time.
Sunderland, Vivint Solar: It’s nice to have that competitive element, where there’s various markets who want this asset class, but what’s driven better advance rates and lower costs is the maturity of these companies. The major players have seven to ten years of operating experience; they have demonstrated a history of operating these assets.
Eastwick, CleanCapital: You also want to look for diversity in your funding. You don’t want to completely rely on one source of financing. You want to be able to tap into multiple forms of financing, make sure it’s going to be there through both up cycles and down cycles.
Sunderland, Vivint Solar: That’s right. We have hit every single market. We’ve also done a forward-flow recently. That’s a healthy thing to do. We don’t want to be dependent on any one market. But what we have seen is, as we’ve started to shift our focus to the securitization market, the other markets have realized what’s happening and they’re trying to find ways to bump up advance rates. The way they’re doing that is essentially taking the same terms we give ABS lenders and pulling them into bank loan deals. You’re starting to see terms that you would only agree to in a securitization slowly start to sink into these other markets. That’s how they’re able to be competitive. They can get to the same level of comfort if they get the same terms. Ultimately, there is a competitive process, but it’s not about Vivint Solar or other sponsors just getting the very best deal, because to get the highest advance rate, you’re going to have to agree to all the same terms as a securitization.
PFR: What’s the benefit of doing a forward-flow agreement?
Sunderland, Vivint Solar: Third-party-owned solar is a very cash intensive business, so we spend a lot of time focusing on working capital and forecasting it. On the traditional financing path, you raise tax equity but you get those proceeds at installation, then you raise some type of temporary debt, and you get that close to the time the system is placed in service, and then, once you’ve accumulated enough assets, you can hit one of these more permanent take-out markets, whether it’s ABS, bank loan or private placement. That whole process, from the system getting installed to the final take-out, can be an 18- to 24-month process. It starts to really drag on working capital. So the idea is, with a forward-flow arrangement, you can condense that capital cycle down to getting all of those proceeds before placing in service or at placing in service. It reduces the strain on the business, but you are giving up the upside, because you have to give away some of the residual. That’s the balance you’re constantly focused on. How much upside to you want to give up to help manage your cash position?
Hunsberger, Credit Suisse: When we think about forward-flow more broadly, in the context of loans as well, there’s another point I think we’re just starting to crack now. One of the points of feedback I hear a lot from issuers on the loan side is, they not only like the price and the volume and the terms they might get under a forward-flow arrangement, but also—one of the other points Ben was talking about—selling loans outright during the origination period is less cash intensive.
And also, because of risk retention in an ABS deal, sponsors might not be able to bring in third-party proceeds equal to what the total third-party bid could be on the asset class absent risk retention. That 5% wedge, that classic risk retention, is a material wedge when you think about trying to originate and sell. So this other benefit that we see on the forward-flow side is, (1) the originator of the loan is not going to be subject to retention rules for securitization, and (2) that means that the person who is going to be holding that loan, once it’s purchased from the originator, can either hold it on the balance sheet or securitize it themselves. They might be a more efficient risk retention owner over time than somebody who needs every dollar of capital in the door on day one.
PFR: My next question was going to be, ‘What else could be holding back the market?’ But maybe the market’s doing great. What is the limit? Is it just how many assets can be originated?
Hunsberger, Credit Suisse: If I were to make a prediction for the year ahead, it would be that this type of syndicated institutional investment, whether it’s 144a, 4(a)(2) or otherwise, is going to apply to more asset classes than we’re talking about today. You can apply ABS methodology, CLO methodology, all these types of diversified pool methodologies, to the C&I assets Matt was talking about, and to even larger assets as well, portfolios of utility-scale assets, or mixed portfolios that include utility-scale assets. We’re actually just getting to a point where we’re bringing efficient capital markets into the broader power and energy landscape. You see it on the energy side with PDP [proved developed and producing oil and gas asset] securitizations that have started to occur. You’ve had investors be forward-thinking on securitizing underlying leases, on thinking about efficient ownership and financing of transmission assets. So if the limitation is the total amount of assets that come into play, I would expect that pool of eligible assets to grow over a year or two, as we’ve seen in the total growth in new energy infrastructure in the U.S.
PFR: Does anyone else want to make any predictions for solar securitization in 2020?
Cohen-T-REX: I think we’ll see a continuation of 2019 versus 2018, which is more of each of these different types of transactions, 144a, 4(a)(2), and forward-flows of all types. I do agree that it’s in a broader energy mix, in the same category with oil and gas. The more you have investors that have appetite for securitized products, on the whole, that’s good, and that trickles down to this as it does to adjacent asset classes. An interesting question is what happens with the tax equity sunset. What’s going to happen around that? It’s early days but maybe securitization would actually have to take off to fill in the gap of institutional capital coming into the capital stack for every individual system, whether it’s a residential system or a commercial system.
Hunsberger, Credit Suisse: It’s a really interesting point. There’s an argument that PPAs—whether utility-scale, C&I or residential—are being supported through the tax regime for a long period of time and that’s not going away overnight. But, at least right now, as we speak today, it’s in the sunset phase. If you follow that argument through, as those subsidies go away, the break-even price, absent material future changes and what the cost curve looks like, is going to be a higher-priced PPA and a higher cash price paid for the asset, if you’re taking away your tax price paid for the asset. That means more debt finance is going to come into the asset class and that means, to Benji’s point, solving across all those different streams in a larger amount.
Henne, KPMG: It may not be next year, but the first C&I deal will be a big market mover. And then as the tax equity investors take a smaller and smaller piece, securitization will probably fill that gap, or other marketplaces will fill that gap. Speaking to some of the underlying assets that are available to be securitized, both residential and C&I, they don’t have conduit deals to aggregate them, at least not as plentifully as they do in some more mature asset classes, but the underlying assets themselves, the raw materials, the residential contracts—whether it’s loans or third-party ownership—they’re still growing. SEIA just put out their Q3 market outlook. In 2018 the resi market in the U.S. grew 8%. Year-over-year in Q2 it grew another 8%. But most importantly, within that growth in Q2, approximately a quarter of all the new installations came from states outside of the top ten, which is a boost for diversification. Looking at some of the PPA deals, third-party ownership deals, California is regularly 40% of the collateral pool. But Maryland has just increased their renewable portfolio standards. Illinois has increased their inducements. That should boost some of the collateral that’s available for securitization.
Neglia, KBRA: The way we predict volume on the loan side is to start with what headwinds or tailwinds are in place to either increase or decrease originations. In February 2018, everyone was talking about tariffs and how that’s going to be a strong headwind—it’s going to drive up costs to the detriment of origination volume. But it hasn’t had the impact most people were predicting, mainly because the panels and the equipment represent a small share of the total costs. We’ve seen system costs continue to decrease. The other thing that people like to talk about is the phase-out of the ITC. At the end of 2019 it’s going to go from 30% to 26%. But the ITC has accomplished what it was set up to do—help drive investment in solar financings. I don’t think that dropping the ITC is going to have a material impact on residential originations. We’re already starting to see some finance companies create products in advance of this step-down. Mosaic for instance, rolled out a straight amortizing loan that doesn’t reamortize or require a prepayment. Overall, we believe originations are going to increase, for a variety of reasons. Consumers are more comfortable with the technology itself, they see their neighbors starting to have it. Additionally, as you start combining solar and battery, it appeals to a different customer base. Several years ago, you had early adopters that were looking at it for a green standard, but now, customers are going solar for savings and to get off the grid. Especially in California with some of these systematic power-down times to prevent wildfires, being able to produce their own electricity is appealing, which can help drive origination volumes.
PFR: On the second Mosaic deal this year, there seemed to be more tranching relative to previous 144a deals. Is that something that is likely to become more of a feature of future 144a deals?
Neglia, KBRA: I believe so. KBRA rated four classes of notes for Mosaic’s 2019-2 transaction, AA- (sf) all the way down to B+ (sf).
Hunsberger, Credit Suisse: It's interesting. This touches on one of the things we talked about—this isn't an esoteric asset class anymore. How do we think about differences between investors? There's been historically more tranching on the loan side than on the lease and PPA side. The most tranches that lease-and-PPA has seen is two. And sometimes you have an unrated tranche as well. Whereas, on the loan side, you can have four-tranche issuances and there won't be a residual buy beyond that. We have started to get some feedback from investors saying: ‘Can we think about time tranching these more? Can we think about even floating-rate structures?’ As volume grows, as you need to think about the marginal investor’s preference, to keep the total cost of capital down, we should explore some of those tools, whether it’s on the loan side or third-party-owned side, and go out to market and get some feedback on how people think about some of those variabilities in how they would price a specific part of the asset class. You do have investors saying: ‘I really like this space and it's great from an ESG perspective. But even with an eight-year ARD, it's too long for me.’ If we can do a two-year tranche, like you might see on an auto deal or on a credit card deal, that could be very attractive to them. That's something that is collectively on us to think about as we continue to solve for the most efficient path of execution.
 At the end of 2018, the average delinquencies in a Sunrun portfolio represented 0.09% of total billings for receivables more than 120 days past due, according to a KBRA pre-sale report dated June 6, 2019.