By Eli Katz
What has the market learned from the solar securitizations to date? How much do they reduce the cost of capital? What are their potential uses?
The key players in the transactions to date talked about them at the Chadbourne 25th annual global energy and finance conference in late June. The panelists are Stephen Viscovich, managing director, securitized products group, at Credit Suisse, Michael Cheng, director, structured credit, with Deutsche Bank Securities, Andrew Coronios, a partner with Chadbourne in New York, and Xilun Chen, director, structured credit ratings, at Standard & Poor’s. The moderator is Eli Katz from the Chadbourne New York office.
MR. KATZ: Steve Viscovich, we have been talking about securitizations in this market for three to four years, and only in the last year have people finally cracked the code. What happened that made all the stars align?
MR. VISCOVICH: A number of things happened. First, we finally had a solar company with a large enough portfolio of installed rooftop systems with long-term contracts with creditworthy customers to access the rated securitization markets. Second, SolarCity, Chadbourne and we, working with Standard & Poor’s, were able to work out the criteria, a deal structure and a process for getting to a rating. Third, enough data was collected on system performance to allow potential investors to make an evaluation.
If you think about securitization in general, the biggest challenge is trying to figure out what the stress and base case scenarios should be for a 20-year revenue stream when there are only a couple of years of performance data. We tried to make as many correlations as possible to utility receivables, but the comparison is not exact.
MR. KATZ: Michael Cheng, same question.
MR. CHENG: I think of each securitization as building a pipeline from the asset to the capital markets. Focusing on the PACE securitization that we did in March, it took a number of years after enactment of the PACE legislation for the originating entities to build out their IT platforms and their channel partners and work out arrangements with the municipalities involved. After that, it was collecting performance data and working with the rating agencies to work out their criteria.
Once we crack the code on an asset class, then what remains to be done is to optimize the initial structure. We are looking to build out the asset class and broaden the investor base.
MR. CORONIOS: On a legal side, solar securitizations required adapting concepts used by other asset classes for use with solar assets. But what it really took was the commitment of the originator, SolarCity, to devote the resources to get the securitization rolling. The solar company must have enough manpower to handle the asset due diligence and build the IT platform and infrastructure to service customers. People talk about scalability of the pipeline. Once you invest the effort in setting up your internal processes, then you can use that to roll out securitizations on an ongoing basis.
MR. KATZ: As the solar rooftop industry matures, it looks to push down the cost of capital. We clearly see that happening. We see various forms of tax equity, yield cos, REITs and securitizations. What are some of the tradeoffs with choosing securitization as a form of financing?
MR. VISCOVICH: All the sources of capital you mentioned will have to work together harmoniously for this industry to reach its potential.
Comparing securitizations to yield cos, people get too caught up sometimes in the headline numbers and think incorrectly that the dividend yield for a yield co is the real cost. With a yield co, there are also growth expectations. It is far cheaper to raise money through securitizations than yield cos. Securitizations are borrowing in the public debt market. A yield co is a path to raise equity. Debt is always cheaper than equity.
Securitizations are repeatable and scalable and, once you have a process down, it becomes something that is efficient from an execution standpoint. People focus on how long the first deal took. We had a lot of rough months along the way, but I think Andrew Coronios and I spent a total of seven or eight months really focused on the transaction and working with the rating agencies. We closed the second deal in eight to nine weeks. It will take a little longer once you start including tax equity or pairing a securitization with a yield co, but these structures, once they have been worked out, scale rapidly.
MR. CHENG: Securitization is another pocket of capital to add to the capital structure. Having lower-cost debt can improve the returns for the equity.
Advance Rates and Tenors
MR. KATZ: Xilun Chen, what makes a securitization attractive from a pricing perspective, and how should one think about them?
MR. CHEN: We have rated two solar securitizations to date. The maturity date for the debt has generally been beyond the contractual term of the customer agreements that are the source of revenue. The customer agreements generally run 15 or 20 years. As a rating agency, we have to take into account uncertainties related to asset and customer performance as well as regulatory uncertainties.
As for the advance rate on the debt, I think of it as a net present value number. The advance rate is a function of the discount rate applied to the customer revenue streams.
MR. KATZ: Let me make this easier for those who do not follow the asset-backed securities market. Suppose I can borrow from a bank 80% of the asset value. What advance rate would I get in a securitization?
MR. VISCOVICH: Part of the challenge is the value may look different depending on who is assigning a value. When we go in for a rating, we are approaching it from the perspective of someone who lives in a “what’s expected” world. The rating agencies live in a “what if” world. What if this happened? What if that happened? There is a bell curve of outcomes. What’s expected is the mid-point of the curve, while the rating agencies, depending on how much history you have with the asset, want to go farther and farther out the tail. So we tell them some of the stress cases they run are ridiculous.
As there is more history, the ratings improve and move more toward the middle of the curve.
The value in a securitization has two components. One is the discount rate applied to the revenues under the customer agreements. You also have 20-year customer agreements with 30-year assets. That is where you will get your advance rate.
Everyone in this market knows that there is value beyond those 20 years of contracted cash flow. There is a renewal value, and it is not being taken into account currently in the ratings. It will take more time before the rating agencies feel comfortable including it.
So we used two terms in the deals. We talked about the “aggregate discounted solar asset balance,” which is just a discounting of the contracted cash flow. We also talked about the “total solar asset value,” which includes some assumption about renewal cash flows. Clearly there will be a power market in 20 years. The only question is what will be the cost to participate in that market. The net renewal revenue has to be greater than zero.
MR. KATZ: Michael Cheng, the securitization you closed in March was a securitization of payments under a PACE program. What was the advance rate?
MR. CHENG: Maybe I should give a little background for folks in the audience who may not be familiar with PACE programs.
PACE, which stands for “property assessed clean energy,” was an Obama White House policy initiative, in conjunction with the Department of Energy, that was implemented in the fall 2010 under which a home or business owner can borrow from a municipality to retro-fit existing infrastructure to improve the energy efficiency. The home or business owner might install solar roof panels. It could be HVAC systems, energy-efficient storm windows or any number of other improvements. The loan is repaid through additional property taxes over time. There is a voluntary additional property tax assessment on the property owner.
From a securitization standpoint, the credit profile is robust. You have very lightly levered property and low loan to value.
The advance rates have been around 10% loan to the value of a residential property. As for the first transaction, on which Deutsche Bank was fortunate enough to act as sole lead and that closed in March, we were able to achieve a AA-rated senior tranche at a 97% advance rate with an 11-year average life note.
The credit spread was approximately 180 basis points over the swap rate. The all-in coupon was 4.75%.
It was well received by the investor community because the asset-backed securities market has a lot of capital to deploy in the current credit cycle. It is very interested in this new emerging asset class. We were overrun with interest when we brought the deal to market. We had more inbound calls asking about the transaction than any other pending deal. We look forward to optimizing the structure on a go-forward basis. We think the market for PACE paper is massive.
MR. CORONIOS: One difference that struck me between PACE and the solar asset securitizations that Steve Viscovich and I have worked on is the difference in the ratings. The ratings are driven by the historical performance data. How were you able to get to an AA rating with the PACE paper?
MR. CHENG: There is really no directly observable collection history for PACE special assessment receivables. We were able to convince the rating agency to use the county’s tax collection history as a proxy for performance. For the property tax jurisdiction where the initial pool was based, property tax collections were greater than 100% of the billed amounts because of penalties and late fees. Everyone pays property taxes eventually or he loses his house or building. That’s why we were able to achieve a 97% advance rate.
MR. VISCOVICH: The PACE loan is a lien on your house that primes the mortgage; talk about something that has hard collateral behind it! The lien also travels with the property. If someone wants to sell the house when there is still a balance remaining on the PACE loan, then the balance will need to be cleared up.
These are two different products. Compare a solar securitization that is really just based on the solar rooftop system and monetizing the electricity it produces over its life to a PACE securitization with less than a 10% loan to value. This has a huge effect on the rating. The PACE securitization is secured by the entire house or building and it sits ahead of other debt that may already have been rated AAA.
There is an interesting tension in the market between the solar rooftop companies, which are offering customers a solar rooftop system through a lease or power contract, and the PACE providers who are making loans to help customers buy the systems. The tension is over which business model will prevail in the long run.
A homeowner has to consider whether he wants to put a lien on his entire house or just have a payment obligation secured solely by the solar rooftop system.
MR. KATZ: So PACE is the easier one to securitize, but it is a much smaller market, right? How large is the PACE market? How big do you think it can get? Do you think the growth in securitizations will be with the SolarCity model or the PACE model?
MR. CHENG: Keep in mind the PACE legislative framework has only been in existence for three and a half years. It takes time to build out a functioning asset platform. The first deal, which was a pilot or proof-of-concept securitization, was $100 million.
Our clients have suggested they expect origination within the jurisdictions where they have coverage in California to exceed $1 billion this year, with substantial growth thereafter.
As background, 31 states have adopted PACE statutes. The program is most active in California, but there are active programs as well in Connecticut, Florida, Louisiana and New Jersey among other states. Thus, the market is still developing. We expect that there will be a healthy flow of capital into the space and very brisk development of the asset class as it rolls across the country.
Commercial and Industrial Market
MR. KATZ: One challenge the solar industry faces when trying to do securitizations is to have standardized customer agreements. Xilun Chen, how much of a challenge was this for rating the SolarCity transactions and how big a challenge will it be when other solar companies try to do the same thing?
MR. CHEN: I can only speak to the two SolarCity transactions that S&P rated. In those two transactions, the customer agreements have many similar economic terms, including the lease and PPA rates and some form of prepayment, purchase option and inflation adjustment and varying production estimates and guarantees. We felt that some of the differences in contractual terms may not have much of an economic impact, but nonetheless are probably quite important, such as access rights to site locations for maintenance and other purposes.
MR. CORONIOS: One of the things that made the first securitization successful was you had an integrated developer with robust standardized forms, and that let the rating agency look at a pretty standardized process. As for the rest of the market going forward, the Department of Energy has a group called Solar Access to Public Capital, or SAPC, that has put in a lot of effort into trying to standardize documentation going forward. The group has posted a residential power purchase agreement and residential lease to its website. It is working on similar agreements for use with commercial and industrial customers. These forms will be a huge help.
It would be prohibitively expensive to do a deal with thousands and thousands of contracts with different terms. The due diligence would be impossible.
MR. KATZ: Do you see solar securitizations moving next to the commercial and industrial sector?
MR. CORONIOS: Commercial deals are a more difficult market, but the developers in that market very much want to see it so, yes, the forms will be standardized through a group like SAPC and the bigger developers will have to keep their forms as standardized as possible. In the residential market, the companies are better able to impose standardized terms on customers.
The deals are bigger in the commercial and industrial market and the terms are more likely to be negotiated on a deal-by-deal basis. Every customer will have a lawyer look at the agreement. Once you get lawyers involved, trying to keep to a standardized document becomes a real challenge.
MR. VISCOVICH: The effort by SAPC will be very helpful for smaller developers who lack the resources to develop their own contracts. The market will not work with 100 different types of contracts.
However, if sponsor A has its standard contract and sponsor B has a different standard contract, that’s okay, provided each contract has the important terms: for example, it is enforceable, the customer cannot just walk away without making the full payments, etcetera. Each company has its own sales pitch to customers. Contracts are bound to vary at least to that extent. The market should be able to handle different solar business models.
Minimum Deal Size
MR. KATZ: What is the minimum size pool of assets required before a company can do a securitization?
MR. VISCOVICH: We look at the issuance size as opposed to the asset pool. The first SolarCity deal was $54 million, which is really small for securitized markets, but it was almost a pilot offering to prove the market. It is expensive to do your first deal with all of the diligence, setting up the legal structure and going through the contracts. Once you get it set up, then efficiencies kick in on future deals. You probably do not want to go smaller than $50 million to make it economic after the transaction costs. We used to say $75 to $100 million was a good sweet spot, but it depends on what your alternative source of capital costs you.
These are numbers for a broadly distributed deal. There is also the option of a single investor deal where one investor buys the whole thing. The minimum size for a single investor deal is probably around $40 million.
MR. CORONIOS: This product is still in its infancy, so the fact that the first deal was a publicly-rated deal was highly unusual. Before the 2008 financial crisis, this kind of asset class would have started by emerging through commercial paper conduits or bank-sponsored financing where the players and the market already understood the asset.
MR. VISCOVICH: And that step normally helps accelerate the growth of the market and acceptance of the asset class. However, we see this being a multi-billion dollar issuance asset class in the next couple of years based on the projected number of residential and commercial PV installations. There is a lot of buzz in the market about the deals that have been done to date. Doing the rated deal early and getting it out to a broader universe of investors will help the market because now you have a larger investor base looking to absorb the projected volume.
MR. CHENG: The analysis of the optimal size for securitization is no different from one asset class to the next. Two key factors are the amount of leverage you can carry and the cost of that leverage, including transaction costs. The legal fees will be the same whether it is a $10 million deal or a $100 million deal. Bankers will charge what they will charge regardless of the deal size. The first PACE transaction was a hair over $100 million. Again, that was proof of concept. The remaining deals this calendar year will probably be three or four times that size.
MR. KATZ: All PACE?
MR. CHENG: All PACE, and mostly on the residential side. To convert that into a data point for this audience, that is $120 million of actual solar product financed through PACE.
MR. KATZ: Xilun Chen, what do you think is the pipeline of securitizations in the sector?
MR. CHEN: We can’t comment on discussions with clients, but we are constantly hearing from market participants. Some of the challenges we see are the amount of data on asset and customer performance and regulatory uncertainty. An overarching focus for us is the value proposition itself: how much are the customers who use the product expected to save over time on their electricity bills?
MR. VISCOVICH: The industry needs to drive down the cost to be able to continue to compete with utility rates in a post investment-credit period. Everybody makes electricity payments. I do not know anyone here who fails to pay his utility bills. The question for the future of this industry is: who are you going to pay, your solar provider or your utility? It will come down to which one is cheaper.
Combining With Tax Equity
MR. KATZ: Most solar rooftop developers have been using tax equity to finance their systems. How will securitization fit into the tax equity structures?
MR. VISCOVICH: It has to. It is still a work in progress among all of the parties involved. If you go back and look at the securitization market in general, we always talk about two different worlds: pre-crisis and post-crisis. It is not new to combine securitizations with tax equity. If you go back to the late 1990s and early 2000s, we did deals where there were leveraged sale-leasebacks, and what we ended up securitizing were rents paid on portfolios of auto and other equipment lease portfolios. The tax equity was effectively the lessor, and the deal was done at the lessor level.
It is really just a matter of getting people to sit at the table and understand the real risks as opposed to just saying no. There is plenty of precedent with other securitization asset classes for combining securitized debt and tax equity, and we think you will see the same thing in this market.
MR. CORONIOS: The tax equity investor will want to remain in place and not suffer any recapture loss. The securitization party has to be able to make sure he can get his hands on the cash flow and the assets that are supporting the financing. The first deal combining tax equity with securitized debt in the solar market will be an inverted lease structure, which is the easiest to do because there is no transfer of title. Partnership flips and other structures may get a little more complicated.
MR. KATZ: That is clearly the next frontier, right? Let me see if the audience has any questions.
MR. GREENWALD: Steve Greenwald with Credit Suisse. Are the payment obligations from the home owners unconditional or are they in any way conditional on the panels working?
MR. VISCOVICH: It depends on the form of customer agreement. Some customers lease the systems for a fixed rent each month. Some sign power contracts and pay for the kilowatt hours of electricity produced.
MR. SILVESTRINI: Mike Silvestrini with Greenskies Renewable Energy. Are some forms of tax equity transactions more easily combined with securitizations than other forms?
MR. CORONIOS: Any structure that requires a transfer of the asset out of the tax equity vehicle is a non-starter because the transfer would trigger recapture of investment tax credits. So that makes partnership flips the hardest ones to make work. The inverted leases are the easiest because the legal ownership remains with the lessor.