We followed with great interest this week as SolarCity unveiled their MyPower residential loan product. Many articles have already covered some of the unique aspects of the loan, including that it is available to customers with a 680+ FICO and comes with a 30-year warranty, production guarantee, and monitoring service package. While we don’t yet know the particulars of exactly how the loan is structured, we think the structure raises a few interesting topics. We discuss below those elements that we have not yet seen covered by other articles. In particular, we think the loan highlights a central thesis held by our firm: the likelihood of customer payment in any solar financial product is a function of both credit risk and technical risk.
The linkage between technical risk and customer payment
The SolarCity MyPower loan makes explicit one of our core beliefs that is usually underlying other emerging loan products: a customer’s likelihood to pay is a function of both the quality of their credit, and the ability for their system to produce electricity. We bucket financial structures into one of two broad categories:
For financial products where repayment is a function of electricity production, such as a PPA or SolarCity’s MyPower loan, the customer payment is a direct function of the system’s electricity production.
For financial products where repayment is not a function of electricity production, such as a lease or traditional loan, the investor is implicitly taking production risk within their portfolio.
For PPAs and the MyPower loan, SolarCity and its financial investors are directly taking performance risk. Unlike almost every other consumer asset class, which has fixed payments regardless of the quality of the underlying product, members of the financial community need to understand and monitor asset performance to appropriately risk-weight and evaluate their portfolios. While this is generally accepted among the investment community, it is difficult to do in practice across many portfolios among different solar developers, many of which have proprietary monitoring and reporting platforms, due to lack of standardized reporting or investor asset management platforms.
For leases and traditional loans, we often hear that the consumer is taking all of the technology risk. We view this differently. Consumers are sold on the fact that solar technology will save them money; to the extent that they are not saving money due to underproduction (or a poorly structured contract), we believe that they have a higher likelihood of strategic default. At minimum, as soon as the customer realizes they are not saving money, the solar bill falls to the bottom of the list when a homeowner is short on cash and needing to prioritize their bills. And we do believe that, over time, customers will become more savvy in understanding their utility bills, and we expect that they will know when they are not saving money. It’s not difficult to conceive of a simple website that allows a homeowner to quickly evaluate their existing PPA or lease to determine savings.
As a loan-that-feels-like-a-PPA, the MyPower loan is a useful case study in linking the technology risk of a PPA to the financial structure of a loan. At kWh Analytics, our tools are designed to make more transparent the hidden connections between asset performance and the financial quality of the underlying portfolio, and we are helping our partners uncover insights about consumer behavior via our investment management platform.
In addition to highlighting one of our central theses, we think there are other interesting implications for SolarCity:
Impacts on long term risk
We think the shift toward a full-service lease is a trend that is here to stay. For SolarCity in particular, the 30-year loan should take away investors’ concerns around renewal risk at year 20, and removes contingent liabilities for system removal at the end of the system’s life, which is required under both SolarCity’s lease and PPA products. Assuming the O&M services and production guarantees are included in the consumer’s lease payment, however, the loan may add unfunded O&M obligations to SolarCity’s balance sheet. Alternatively, if SolarCity is allocating costs to their servicing platform, their implied interest rates are even lower than stated to the homeowner.
SolarCity allows their customers to prepay any or all of the loan, at any time, without penalty. We view this as a big benefit to the homeowners and is in their best interest. While this is a common structure in the mortgage industry, it leads to investor uncertainty regarding repayment risk: as interest rates rise, consumers are unlikely to refinance or pay down their loan, and the investors are stuck in a long-dated but low-yielding investment. As interest rates drop, homeowners are more likely to refinance their investment just at the time when the investors are looking for yield. While people have developed models to predict this structure in other asset classes, we believe repayment uncertainty may lead to higher financing costs for SolarCity.
The tax elements of the loan
We believe that SolarCity may sell their systems to homeowners at the Fair Market Value (FMV) of the system- i.e., what they believe it is worth, as opposed to what the system costs. They make the economics work for the homeowner with their very competitive interest rates and innovative pricing structure. According to SolarCity’s most recent investor deck (slides 14 and 15), the FMV of the systems are $4.75 / W but cost $3 / W to build. While we won’t expound on their current disputes with Treasury on how to evaluate the value of a system, we think the product enables an interesting situation: while it should be defensible that a homeowner paid FMV for the system, and it would be complicated for the IRS to challenge the tax returns of each individual, the FMV price would only be palatable to a consumer in situations where the homeowner takes the MyPower loan (given that competitors can sell cash systems for $3.50-4 / W). It could be viewed that SolarCity is using excess margin to buydown the interest rate on the loan. We don’t know how the Tax Man would respond to this, but suspect SolarCity may draw additional scrutiny if there is a big difference between their sales prices and/or interest rates as compared to the broader market.
We think the SolarCity loan is an interesting product to watch develop. They have proven they can execute on new structures, and have a knack for shaking up the market when they innovate.
(Disclosures: We are not tax lawyers, so don’t rely on our comments on tax risk. Also, the authors of this post do not have any financial positions in SolarCity stock. Image by Gray Watson courtesy of CC.)