Risk management for solar projects
The Solar Energy Industries Association published a guide called “Best Practices for Solar Risk Management” in September. Jason Kaminsky, chief operating officer of kWh Analytics and the author of the guide, Ed Rossier, a director of project management for renewable energy investments at US Bank, and Mike Mendelsohn, senior director of project finance and capital markets at the Solar Energy Industries Association, talked about the topic during a webinar in November. The following is an edited transcript. Keith Martin with Norton Rose Fulbright in Washington is the moderator.
MR. MARTIN: Jason Kaminsky, what is the difference between an asset manager and a risk manager? The guide suggests the difference is important.
MR. KAMINSKY: The sponsor does asset management. It has the asset. Bankers and tax equity investors do risk management. They want to make sure they will be repaid or reach their target returns.
An asset manager at a sponsor is responsible for supervising technicians, overseeing O&M agreements, managing spare parts and the other physical aspects of making sure the project performs. He or she is also responsible for things like sending and correcting invoices, preparing financials, getting auditing and accounting help, and preparing reports for the financiers.
Risk management is what a banker or tax equity investor does in preparation for and after an investment. That includes things like tracking, monitoring, reporting on, and managing the health of the investments after they have been made, and managing internal stakeholders who have an interest in the financial health of the portfolio.
MR. MARTIN: How did you come up with the best practices you recommend in the guide?
MR. KAMINSKY: First, kWh Analytics works closely with a number of investors. We drew on their insights. Second, I drew on my own experience at Wells Fargo. I was one of the early members of the solar tax equity team, so we basically built our own risk management platform from scratch. Third, various members of the solar energy advisory council at the Solar Energy Industries Association read and commented on the guide, and then we had peer review by about a dozen other reviewers from industry that included other lenders and tax equity investors.
MR. MARTIN: Ed Rossier, you have a huge portfolio of tax equity positions in solar projects. Risk management has to start for you when you are first looking at a potential project. I suspect you end up cataloging all the risks, and then you write into the deal documents which party takes each risk. The ones that US Bank will take have to be quantifiable if you are going to invest.
People sometimes say that it is not important to eliminate all risks. You just have to be able to quantify them. Whose job is it at a tax equity bank like US Bank to identify these risks? Start at the front end of the process.
MR. ROSSIER: Most banks are probably organized with three lines of defense. There are the business lines, then risk management and compliance, and then an internal audit group that makes sure everybody is doing what he or she is supposed to be doing.
Our business line is the world that I live in. These are the revenue-generating individuals at the company.
MR. MARTIN: You originate deals.
MR. ROSSIER: It is everything facing the customer. There is a separate line that is concerned with implementing policies and approving investments. That is separate from the team that originates, closes, and manages the assets. Within that world of the customer-facing business-line function, different banks choose different ways of organizing things.
We are organized into three groups: we have business development officers, who are originators who source new customers and new investment opportunities and take them through the letter of intent. Once the letter of intent is signed, it is transitioned to my team which we call project management, but it is really underwriting, negotiating and closing.
That team will take an investment from signing of the letter of intent through closing on the definitive deal documentation. At that point, the investment is transitioned to our asset management team, which holds it through the life of the investment until we exit.
MR. MARTIN: You are in the middle position. The letter of intent has already been signed. Now the deal has to be documented. You have to understand the risks. You bring in a lot of consultants, including lawyers, to help you understand everything. There is also diligence done by the internal team.
MR. ROSSIER: At this point, the due diligence requirements for most deals are pretty standard, depending on the deal type. Most people will circulate a due diligence closing checklist pretty early in the process that covers the waterfront.
Our internal team will review everything in conjunction with outside counsel. We engage a variety of other experts, including independent engineers, appraisers and accounting firms.
MR. MARTIN: Could someone contemplating raising tax equity from US Bank ask you in advance for the checklist so that he or she can get a head start on setting up a data room?
MR. ROSSIER: Yes, it happens occasionally. A big caveat is that until the specifics of the assets are known, it is hard to cover everything in a checklist.
MR. MARTIN: There are three different business lines involved in moving the deal to completion and then afterwards during the operation phase. One is a group that gets it through the letter of intent, then you come in as part of another team that does the diligence and documents the deal. Then the baton is passed to an asset manager.
MR. ROSSIER: Not to confuse terms, but that is the term that we use. At our bank, the asset managers process all of the post-closing fundings and then any amendments or issues that come up in deals. They manage through exit.
MR. MARTIN: How many solar projects does US Bank have in the asset manager stage at this point?
MR. ROSSIER: Something like 250 investments.
MR. MARTIN: How many asset managers do that many projects require?
MR. ROSSIER: Our team is around seven, and then there is another group of portfolio analysts that supports them, which is about five or six people. There is some permeability between people in different business lines, and there is a lot of cooperation, because every investment that we do is either with a repeat customer or a customer that we hope will become a repeat customer.
The information from each group has to be shared with the others because the originators need to know how projects that already closed are performing, and underwriters need to understand what might have gone wrong so that they can incorporate that information into their underwriting. We all sit together. There is constant communication among the groups.
MR. MARTIN: What does a portfolio analyst do that the asset manager does not?
MR. ROSSIER: They collect everything that is required to be delivered and then dig into financial statements, operational statements, tax returns and the like and are available to help do a deeper dive into any asset that might be troubled or where we want to pull a little more data out of our portfolio.
MR. MARTIN: That’s a lot of work. What is left for the asset manager to do?
MR. ROSSIER: That’s a good question. [Laughter].
MR. MARTIN: There are seven of them compared to five or six of the people you just described.
MR. ROSSIER: The bulk of their time is spent on fundings. We do a lot of residential solar and portfolios of small utility-scale solar, and each of those investments will have multiple fundings. You could have monthly fundings or twice-monthly fundings all year long for one investment. There is a lot to review in connection with each funding.
After that, most of the work is annual and quarterly reviews of the portfolio and then, of course, any project or sponsor that has distress of any kind will draw the bulk of their remaining attention.
MR. KAMINSKY: One of the surprises to me when I first joined Wells Fargo is that you have the business line, as Ed mentioned, and then there are lots of other internal stakeholders who have an interest in what you are doing. A bank has an accounting group that is working on the bank’s accounting. It has a tax group. You might work with an equipment leasing group. You have internal audit.
A lot of the time is spent directing traffic and making sure that all those other teams have what they need to do their jobs.
Current hot buttons
MR. MARTIN: Ed Rossier, you said the portfolio analysts and asset managers sit pretty close to the deal originators so that you can learn from each other. Is there a formal process? Are there regular meetings? Can you think of anything that has been passed to the deal originators recently from experience on the asset manager’s side?
MR. ROSSIER: Yes. Our company made a conscious decision to integrate the teams by having people from different teams mixed together. We don’t have silos. We think communication is important.
For example, the tax equity investor often has an outside completion deadline, and if you have a lender that is bridging the tax equity investment, it will want a cushion between the maturity date and the outside completion date on the tax equity commitment to ensure it is not left without a takeout. The outside completion dates are negotiated during the term sheet phase by the business development officer. They are essentially a conversation between the business development officer and the customer about an outside date to complete the project that the customer feels confident will be met.
The deal then transitions over to my team. Maybe there are eight to 12 weeks of negotiation, underwriting and closing and, during that time, the schedule might move, but the outside completion date might not. That can create a problem if asset managers are having to ask for extensions after closing because the lender extended its deadline in order to accommodate the construction schedule. The tax equity investor gets asked to extend as well. The delay might be due to a new schedule the utility imposed for interconnection, for example. The delay was not incorporated into the internal credit approval.
This causes heartburn for sponsors and lenders and creates work for asset managers, so we try to learn from the experience and come up with a different approach for setting future outside completion dates at the front end.
MR. MARTIN: Make sure all the dates synchronize. Jason Kaminsky, you worked at Wells Fargo for a while in a similar capacity as Ed Rossier. Are there other ways that you have seen teams organize themselves?
MR. KAMINSKY: Organizations that do not have the volume and head count of US Bank sometimes divide up by customer. This affects the way such a bank makes investments and evaluates risk. The same person takes the deal from start to finish and also acts as the asset manager.
Where the investment group sits within the banks also influences its perspective. Sometimes, like at US Bank, it sits alongside groups handling low-income housing and other tax-credit investments. Other times the group sits within a principal investing group, or an energy lending group or a leasing practice. This influences everything else about the accounting, the oversight and the risk perspective. We see people come at solar from different starting points, and I think it influences the way they evaluate the deal.
MR. MARTIN: Ed Rossier, you mentioned a lot of people who have to sign off and, Jason Kaminsky, you said that a lot of the work is acting as a traffic cop by steering things to people who will need them in order to sign off. The people who sign off include the tax equity business unit, the credit committee, internal and external auditors, the bank regulators, and maybe the tax department.
You probably get to a point fairly quickly where you know the hot buttons of each of these groups. Name a couple of current hot buttons that are getting a lot of attention.
MR. ROSSIER: It can change from year to year and from month to month, depending on the topic du jour. Risk mitigation tied to tax reform is a hot topic today. It is requiring a lot of coordination among departments within the bank.
Another one that is new this year is module supply security and tracking where modules are as a direct result of the Suniva tariff case. The potential for import tariffs creates pricing uncertainty.
Other ones that are popular this year include interparty terms of lenders where debt or back leverage is added to investments after closing, and post-closing changes in sponsor ownership. There have been a fair number of upstream acquisitions of developers, and that has created a lot of work for asset managers.
MR. MARTIN: The risk management guide is really just a compilation of best practices for managing risk. It breaks these practices into three categories. There is risk measurement and monitoring over time. Then there is comparison against industry benchmarks, and finally there is compliance. The guide recommends a dozen best practices. Let’s talk about some of them.
Let’s start with operating risk. This is the first place to focus because it determines the cash flow from the project. The guide recommends tracking a weather-adjusted performance index. Jason Kaminsky, what is that?
MR. KAMINSKY: It is comparing actual output to the projected production in the project pro forma. Is the project or the portfolio hitting its target? If not, is it because of something that the sponsor can control? The main variable is usually weather. If it was poor sun, then it is not usually the sponsor’s fault. On the other hand, if there is an operational issue, we will want the sponsor to address it.
Backing out the influence of the weather — that’s essentially the weather adjusted performance index — can isolate any operating issues.
MR. MARTIN: Who maintains this index? Is it the sponsor, the tax equity investor or the lender?
MR. KAMINSKY: It is usually monitored by the sponsor and then delivered to the various stakeholders as part of the reporting package.
MR. MARTIN: Does the sponsor come up with its own model or is this something that is purchased from outside?
MR. KAMINSKY: It is a blend of both. External information is sometimes used to support an internal model.
MR. MARTIN: How do big data and industry benchmarks play a role?
MR. KAMINSKY: We see a lot of requests for benchmarks. That gives investors a sense of whether assets are underperforming in relation to the broader market. If so, that might be a sign of a bigger problem. The other area we have had a lot of questions about is a significant weather event, like a big snowstorm, fire or hurricane. These events seem to be becoming more common.
MR. ROSSIER: I am a big fan of aggregating data and standardizing data fields. We participated in Orange Button, a US government effort, when the group was determining the taxonomy. I think more standardization is helpful at both the data level and the contract level.
MR. MARTIN: How does standardization help you?
MR. ROSSIER: There are two sides to this. One is we are an active syndicator. We will try to raise about $500 million of third-party tax equity next year and deploy it alongside our own bank’s tax equity. As part of our reporting to third-party investors, the more we can benchmark, the better the product we are delivering to our third-party investors.
The other side is accessing new market segments that are not currently served by tax equity. Two great examples are low-FICO customers, like subprime homeowners, and the commercial and industrial solar market, which has people scratching their heads over how to tackle it.
MR. MARTIN: The data and the benchmarks get you comfortable eventually that a low FICO score is not a problem.
MR. ROSSIER: In theory they could. Or they could tell you the opposite. For now, we don’t know. We recently began working with kWh Analytics to help us aggregate our data, and we use the aggregated data to support our syndications business and underwriting practices. For example, Fannie and Freddie publish a lot of mortgage data. This allows people to do analysis and draw conclusions based on the data. It is much more difficult in solar because every sponsor holds its data as a proprietary source of value for its own use.
MR. MARTIN: How could data help you get comfortable with the C&I solar market where the problem seems to be lack of standardization for the offtake contracts?
MR. ROSSIER: It will not help with the issue of offtake contract standardization, but it does help when you are trying to evaluate default risk or credit risk generally of offtakers or the likelihood that solar equipment will remain in service. There may be ways to look at it on a portfolio level and draw conclusions about default rates or periods of lost revenue. Banks might not start there, but there might be an opportunity for insurers to be first movers in that market.
MR. MARTIN: Next question. The guide says, “Nearly every large portfolio to date has seen the insolvency of a vendor or sponsor.” That is referring to rooftop solar, correct?
MR. KAMINSKY: It is true across the solar market, but more so in the C&I and residential segments.
MR. MARTIN: You go on to point out in the guide that there is also regulatory risk as net metering policies, tax law and renewable portfolio standards sometimes change, sometimes with retroactive effect. The guide tackles this by recommending investors track exposure at both the project and portfolio level by looking at how many dollars are exposed to different equipment types, geography and what else?
MR. KAMINSKY: I guess the best way to answer this is with a few examples. With the recent wildfires in northern California, investors are trying to scope very quickly what kind of exposure they have to assets in that region. Zip code is the key metric. Investors who have their assets divided by zip code can determine their exposures quickly.
In addition to geography, assets might be tagged by installer, equipment vendor, servicer, offtaker and offtaker credit. A robust data set helps an investor respond quickly to the changing market.
MR. MARTIN: So having a matrix on your computer screen or a big spreadsheet showing exposures is a powerful risk management tool. One of the uses is being able to respond quickly to questions from credit. Are there other uses for such a matrix?
MR. KAMINSKY: Yes. Sometimes within a bank you are managing exposures to risks like geographic concentration, low-FICO customers, lots of battery storage, for example. There is a misconception that if there is a problem, you can’t really do anything about it.
I learned very early in my career that problems are okay, but surprises within a bank environment are not.
If you see a problem on the horizon, often if you get a lot of smart people together, you can come up with a pretty good mitigation strategy. That might include financial support. It might include alternative O&M strategies. Using this sort of matrix allows you quickly to scope your risk and decide how you want to manage that risk.
MR. MARTIN: Offtaker credit risk is another issue, especially in the residential solar portfolios and the utility-scale projects with corporate PPAs. What are the best practices to deal with these kinds of risk?
MR. KAMINSKY: You can’t manage what you don’t measure. Today in the residential solar sector, we don’t even have a clear definition of what a default is. We are still figuring out what is the right time to call a default and how to track different performance metrics across the portfolio. The challenge is how to normalize the data or metrics so that people like Ed can make better investment decisions.
MR. ROSSIER: The challenge on the residential side is the market is currently structured and willing to serve the prime homeowners, meaning a FICO score of at least 680 or 700. There is often an allowance for some portion of the portfolio to be unrated with either partial prepayments or ACH requirements to mitigate that risk. That’s just where people have been comfortable, and in the absence of any broad data-based conclusion, I don’t see that changing much.
In the C&I sector, it is really just commercial credit underwriting for the offtakers on a credit-by-credit basis. This is why I think that market has not really grown much because the only time we have been successful doing these types of portfolios are ones where they have, for example, a FedEx contract covering 50 FedEx locations so that you can underwrite a single offtaker and a single form of PPA and it is all standardized. The PPA wraps some of the site-control risk.
Those work, but in terms of my brother-in-law’s gym getting solar on the roof, that is probably not going to be financed by US Bank because there is too much work to do to get comfortable with the credit. A handful of companies are trying to solve this problem using technology and web-based underwriting platforms to make that underwriting a little more systematic. They might be successful, but ultimately that gets to a second regulatory area for banks, which is third-party risk management. If you are relying on a third party to underwrite your risk, then you have to be comfortable with the creditworthiness of the third party, and the bank regulators will hold it to the same standard as the bank. That means there is another door to get through.
MR. MARTIN: It sounds like a lot of this lends itself to artificial intelligence.
MR. ROSSIER: Yes. Our tax equity subsidiary was a community development corporation that was originally focused on affordable housing and new markets tax credits. We were always compliance-focused. Tax credit recapture was our first risk category. Other banks may start with the asset-level risk and then go up the chain.
MR. MARTIN: Is this a case where the more you know, the more risks you see or the more you know, the more the list of risks shortens?
MR. ROSSIER: I don’t know that it is either. With any new investment type, you uncover some risks that maybe you did not think of at first. The risks in solar are pretty well known at this point. Performance is the unknown.
If we could go back 10 years and rebuild our platform and money were no object, every bank would develop a proprietary underwriting database that could tap into other sources of data, allowing you to make the case to your credit or risk management folks that you can underwrite 100 different commercial offtakers in a single fund. That is something that artificial intelligence might eventually make possible.
MR. KAMINSKY: Even if you have identified a risk, it doesn’t mean a bank is always best suited to wear it, especially if it leads to volatility in cash flows. Currently within the lending market, there is a lot of capital, but a finite number of deals. I think in that market, the question is how either to stretch your risk appetite or to find innovative ways to structure around risks. We have secured insurers to accept untraditional risks. In our case, insurers take production risk, and we are finding lenders are able to underwrite projects more aggressively with these policies in place. That is important in today’s competitive market.
MR. MARTIN: Let’s work Mike Mendelsohn in here. The US Department of Energy’s Sunshot Initiative office has been working with the solar industry on an Orange Button data standard. Ed Rossier mentioned it. The standard is described in the guide as a taxonomy for solar data transfer and reporting. What is it? When will it be available?
MR. MENDELSOHN: Essentially it is a dictionary of all relevant data terms so that there is a consistent name and definition for each term. That allows for interchangeability of data and facilitates greater liquidity in the financial market. We expect to see something in the next couple of quarters.
MR. MARTIN: You worked at the National Renewable Energy Laboratory. That is where most of us in the solar market got to know you. The Solar Energy Industries Association hired you away from NREL. NREL had run out of budget for your program. You were working mainly on model contracts for solar deals. Now that you have moved to SEIA, you have also come up with sets of best practices for installation and O&M and for consumer protection. Is the focus of these efforts rooftop solar or are you working more broadly?
MR. MENDELSOHN: We started with distributed generation. We are now thinking about doing standard contracts for the utility-scale market. We recently improved on the commercial and industrial PPA that is designed for relatively small C&I projects. We are in the process of bolting on some different components to that: for example, a PACE addendum and a storage addendum. We are trying to build out our suite of contracts to include other technologies and other finance models.
MR. MARTIN: Where can people find these contracts?
MR. MENDELSOHN: Search in Google for “SEIA model contracts.” Or navigate through the SEIA website. You can always contact me as well.
MR. MARTIN: You suggested your next focus as a group may be on utility contracts. Is there any other potential focus at which you are looking?
MR. MENDELSOHN: We are looking across the investment ecosystem. How can we facilitate reduced transaction costs or allow project cash flows to be pooled into structured finance products?
We were talking a little bit earlier about the C&I sector and how difficult it is to finance. A lot of that is because the originators are too small to come up with a contract portfolio of sufficient scale to reduce risk across that portfolio. Building portfolios across originators has been too difficult to date because of the risks inherent with that. I think we will be able to solve for a lot of those issues and develop portfolios that are large enough where any single project within the portfolio does not represent too much concentrated risk. That will be a valuable barrier to overcome.