Cost of capital: 2020 Outlook

Cost of capital: 2020 Outlook

February 10, 2020 | By Keith Martin in Washington, DC

A group of industry veterans talked in late January about what to expect in the year ahead for tax equity, bank and term loan B debt and project bonds in a widely heard conference call. The US market remains awash in liquidity. There is intense competition among banks to lend. Interest rates remain under downward pressure. Tax equity deal volume is expected to set a record in 2020, making it wise to get financings closed as early in the year as possible as resources needed to finish projects and close deals will be in increasingly short supply as the year wears on.

The panelists are Yale Henderson, managing director and head of energy investments for JPMorgan, Jack Cargas, head of originations on the tax equity desk at Bank of America, Ralph Cho, co-head of power for North America for Investec, Jean-Pierre Boudrias, managing director and head of project finance for North America at Goldman Sachs, and John C.S. Anderson, global head of corporate finance and infrastructure at Manulife. The moderator is Keith Martin with Norton Rose Fulbright in Washington.

Tax equity

MARTIN: Yale Henderson, what was the tax-equity volume in 2019, and how did it break down between wind and solar?

HENDERSON: The market last year was comparable to 2018. The market did about $12 billion in tax equity in 2018. It was in the same $12 to $13 billion range in 2019.

The breakdown was roughly 65% wind and 35% solar, plus or minus 5%.

These figures for market size are fundings and commitments made during the year. The biggest difference we saw between 2018 and 2019 was most, if not all, of the $12 billion was fully funded by the end of 2018, unlike 2019 when not everything got done, and we move into 2020 with spillover.

That means little rest for the tax-equity providers. Everyone has a lot of deals to close in the first quarter.

MARTIN: So a lot of momentum moving into 2020. Jack Cargas, you told me you closed a startlingly large number of deals in the last couple of weeks of 2019. What was the number?

CARGAS: We closed or funded nine transactions in the last two days of the year.

MARTIN: I assume JPMorgan was also extremely busy?

HENDERSON: We did more than $1 billion in fundings in the last week of the year.

MARTIN: What do you expect in 2020?

HENDERSON: More of the same. The market will be much larger: probably close to $15 billion. We are seeing a volume of deals that have already been awarded, but not yet executed, that is significantly higher than during the comparable period last year. We see a lot of big deals in the pipeline still to be done. We hope the deal flow will be spread over the year rather than be back ended. A large number of wind farms commenced construction in 2016 and will have to be completed by the end of 2020 to qualify for federal tax credits. That may contribute to a traffic jam at year end.

CARGAS: We hear observers predicting an increase in demand for tax equity in 2020 by as much as two times. Some of that is due to the hangover that Yale mentioned. There were several hundred million dollars of planned wind tax-equity fundings that were delayed from 2019 into 2020 for sponsor-specific reasons, usually construction delays. Then you have the crush of wind projects that need to be completed by year end. All of that suggests a significant increase in demand this year.

MARTIN: Two times volume suggests $24 billion in demand. Trying to shoehorn that amount of demand into a $15 billion tax equity market may be a challenge.

What percentage of the typical solar project is tax equity as we enter 2020?

CARGAS: For solar, probably less than 40%. The percentage shrank as a result of the so-called tax reforms. A couple of years ago when the federal corporate tax rate decreased from 35% to 21% and tax losses became less valuable to tax-equity investors, the tax equity share of the capital stack declined. It was probably closer to 50% before, and now it is less than 40%.

MARTIN: What about wind?

CARGAS: Similar phenomenon. Wind tax equity had been as much as 60% to 70% of the capital stack and now, after tax reform, it is more like 50% to 60%.

MARTIN: Yale Henderson, do you agree with those numbers?

HENDERSON: Yes.

MARTIN: Many listeners patch into this call to get a better feel for what cost of capital to assume in bids to supply electricity. Tax equity flip yields fell during the past year. They seem to be in the 6.25% to 6.8% range for most utility-scale wind and solar projects. Some larger wind developers reported flip yields a little below 6%. Solar residential rooftop companies seemed to see pricing in the 7% to 8% range. The cost of tax equity is a function of demand and supply. Jack mentioned that demand is expected to skyrocket. In which direction do you sense the cost of tax equity will move this year, if at all?

CARGAS: We are always happy to discuss details around after-tax IRRs with our customers, but perhaps on a less public and more one-on-one basis. What you said about the direction in which yields moved last year is accurate. The future is notoriously hard to predict. While it is accurate to say the primary driver on tax-equity yields over time has been demand and supply, the cost of capital to investors has also become important as well as a few other factors such as the amount of competition for deals. The most attractive projects draw a lot of interest from potential tax equity investors.

Structures are changing significantly enough for there to be some level of yield premium for transactions that the market views as more risky, such as residential solar portfolios with low FICO scores, solar projects presenting basis risk and a whole host of other things. Other factors besides demand and supply have been playing a bigger role lately in the cost of tax equity.

HENDERSON: I agree with Jack. As we have talked about before, flip yields are not created equal. The nature of the project, its location, the shape of the cash flows, the offtaker or hedge counterparty credit are just a few of the factors that play into the after-tax IRR. Some after-tax cash flows are more valuable than others in terms of how they flow through the book return for an investor.

The range you cited is pretty close. Some numbers are higher than that range depending on the project attributes and the length and source of the contracted revenues.

MARTIN: Deficit restoration obligations — DROs — last year were 40+% in many deals. Do you see any change going 
into 2020?

HENDERSON: No. There should not be that much structural change, but the less cash there is in a deal due to low electricity prices, the more pressure there is for higher DROs. The maximum DRO size is one piece of the puzzle. The timing and reversal of the DRO are important elements in how high an investor will be prepared to go.

MARTIN: Both of you mentioned that this year is the deadline for wind projects that started construction in 2016 to be in service. If construction started in 2016 by incurring at least 5% of the cost, the developer can buy more time from the IRS by proving continuous efforts were made to advance the project after 2016. Will we see the tax equity market willing to accept continuous efforts to finance projects in 2021 that started construction in 2016?

CARGAS: That’s a tough question. We ran into something similar in the past, and it proved difficult for the sponsors involved to provide full evidence to document the continuous efforts. It is not clear yet what will happen for projects whose construction slips into 2021.

We encourage sponsors to make sure projects are in service before the cliff so that we do not have to have this discussion. We do not know what logs ought to look like. Should they be monthly, daily, hourly — that’s just one of many questions in this area.

MARTIN: Yale Henderson, I think you said something similar at an offshore-wind conference this fall.

Have either of you seen any tax-indemnity claims made on renewable energy projects and, if so, around what risks?

HENDERSON: Actual claims? I am not aware of any, other than the well-documented issues around tax basis under the Treasury cash grant program.

CARGAS: We have made no such claims on our portfolio.

MARTIN: Do either of you have a rule of thumb for how large a step-up in tax basis you are willing to accept above the cost to construct?

CARGAS: We do not have a fixed rule of thumb. We are aware that the Treasury suggested during the section 1603 program that step ups should normally not exceed 10% to 20%. That does not mean that we can go straight to 20%. The facts and circumstances may say that it ought to be 10% or 12% or 15% or 23%. Twenty percent is not necessarily a cap either. It comes down to the specific facts and circumstances.

MARTIN: Every project seeking tax equity at this point had to be under construction by a deadline. I know you prefer that developers incur at least 5% of the cost. However, many developers have moved to a less expensive approach of relying on physical work on such things as transformers. Are you financing projects based on transformers? How much work do you need to see on the transformer before the construction-start deadline?

HENDERSON: We are willing to work with transformers. Obviously more work is better than less. Certainly if you just have a conservator tank or a radiator, that is more challenging and will depend on a lot of other facts and circumstances. We are very interested in the contract to buy the transformer. What is the time period for performance? How much has been done on the transformer before the construction-start deadline. When will it be delivered? What outs does the sponsor have to follow through on the purchase? How real is the contract? We have not set any bright lines. We look at the totality of the deal and the information available and then make a determination once we know all the facts.

MARTIN: Jack Cargas, same answer?

CARGAS: We have been willing to rely on physical work on more than just transformers, but basically our approach is the same. We want detailed, time-stamped recordkeeping by a credible third party documenting the work that was done before the deadline. There is not any bright line.

MARTIN: Does a single new nacelle work for either of you?

HENDERSON: I don’t think we have ever faced that question, so I don’t have an answer, but it seems a little light.

MARTIN: What other noteworthy trends do you see as we enter 2020?

CARGAS: We see a major squeeze on human resources, especially in the fourth quarter of 2020. This past year was the heaviest year from that perspective ever, and it was almost intolerable. The squeeze is not only with respect to tax equity investors, but also sponsors and consultants like financial advisers, modelers, independent engineers, appraisers, lawyers — the list goes on and on.

MARTIN: So get your deals done as early as possible this year. Yale Henderson, other noteworthy trends?

HENDERSON: Offtake arrangements keep evolving and putting additional pressure on the cash flows from projects. There is less cash due to falling electricity prices. Projects with virtual power contracts or that sell into organized markets with hedges have electricity basis risk. This can reduce the cash flows further. We are spending a lot of time looking at how tightly projects are structured in terms of the ultimate ability to cover operating expenses given how low pricing has gotten to date. We are paying careful attention to project fundamentals.

CARGAS: One other phenomenon that I don’t think this market has seen before is the tax extenders bill in late December gave wind developers an incentive to rescind or cancel construction-start arrangements in 2019 and restart in 2020 to qualify for higher tax credits. There were media reports this morning referring to this as a “last-minute curve ball.”

MARTIN: Wind developers had to slam on the brakes. There was a lot of activity between December 17 and year end. While you were closing deals, the lawyers were doing what you described.

Let’s move to bank debt and Ralph Cho at Investec. What was the volume of North American project finance bank debt in 2019 compared to 2018?

Bank debt

CHO: The preliminary data suggests that dollar volumes were down, but the number of transactions increased, suggesting a move from large gas-fired power plants and LNG terminals to renewables.

The dollar volume was down 15% in 2019 compared to 2018: $59 billion in 2019 compared to $69 billion in 2018.

The number of deals was up 58% to 306 transactions compared to 194 in 2018.

Renewables financings, which tend to be smaller, are dominating the flow.

Another interesting statistic is that if you zero in on power, bank volume was $36 billion in 2019 compared to $43 billion in 2018, so down 17%. This reflects again the move from thermal power to renewables.

MARTIN: How many active banks were there in 2019 and how many do you expect in 2020?

CHO: I like this question because I am very interested in new sources of capital. It is where I spend a lot of time. My estimate is we start the year with 80 to 100 lenders searching for deals, with 40 to 50 of them highly active. It is always interesting to see renewed appetite from institutions as markets are continuously adjusting. By and large lenders are expanding their definitions of what is an acceptable risk profile and what are acceptable economics. Of course, the majority of activity is always dominated by the top 20 banks just given their costs of capital and access to a broader range 
of borrowers.

The lender count includes grey-market institutions or non-bank lenders, and it also includes a lot of new South Korean institutions that have established a strong presence in our markets. If you combine this with the 15% drop in transaction volumes, you can see why you continue to hear the same broad theme of lender demand outpacing the supply of projects to finance. It has been hyper-competitive this year among lenders. It remains a great time to be a borrower. Pricing continues to tighten. In such a market, people come up with creative ways to increase leverage on deals.

MARTIN: What is the current spread above LIBOR for bank debt, and to what does that translate as a coupon rate?

CHO: It depends on the type of deal. Plain-vanilla loans are pricing at LIBOR plus 125 to 137.5 basis points. Short-term construction loans are pricing at LIBOR plus 75 basis points. We have been refinancing operating quasi-merchant gas deals at anywhere from LIBOR plus 225 to 275 basis points. Greenfield gas assets always get a little bit of a premium for the lender, so call it LIBOR plus 287.5 to 300.

It has also been interesting to watch the spread compression, or difference in pricing between Opco and Holdco loans. The excess demand to lend leads to tension among different classes of lenders, essentially forcing everyone to reevaluate the risk that each lender is willing to take and how much the lender is willing to compromise on yield just to get its capital put to work. What used to be a 200 to 250 basis-point spread differential is really compressed into 100 to 150 basis points for any lender who wants to be competitive.

There is also a growing pool of capital available from the stretch senior Holdco type of lenders that is creating a sweet spot at LIBOR plus 350 to 500 basis points based on the underlying risk. If you price something risky at that level, there is a ton of money that will jump in.

As far as the all-in coupon, the three-month LIBOR is about 180 basis points today. Last year at this time, it was 280 basis points, so it is down 100 basis points. Banks swap LIBOR, and LIBOR swaps are coming in around 175 to 200 basis points today. To get your coupon, just add the spread above LIBOR to that rate.

MARTIN: Explain the difference between a Holdco loan and an Opco loan and is the spread compression you just described more relevant to the gas market than the renewable energy market?

CHO: You can create Opco-Holdco loans on any asset. The Opco loan is basically a loan that is secured at the asset level. The Holdco loan is a loan higher up the ownership chain whose repayment is subordinated to the Opco loan.

MARTIN: Correct me if I am wrong, but there is no yield premium currently for back-levered debt that sits behind tax equity in renewable energy deals: no yield premium compared to what the lender would charge if the loan were at the asset level.

CHO: Correct. The back leverage is basically sitting upstairs, but there is really no loan at the asset level other than the tax equity. When we back lever renewables deals, the lenders do not put a premium on that. I am not sure this makes sense, but not only is there no premium, but the fact that the loan is back-levered also does not change the profile of how lenders are willing to size it. The Holdco loan in that context would be another loan that sits behind the back-levered debt.

MARTIN: A few quick questions. Is there a LIBOR floor in the bank market currently?

CHO: No. If there is a LIBOR floor, it is currently 0%.

MARTIN: What are current debt service coverage ratios for wind, solar and quasi-merchant gas projects?

CHO: Debt service coverage ratios for contracted wind farms are 1.35x for a P50 forecast. Solar is 1.25x for P50, given the lower standard deviation on resource forecasts. Solar output is more predictable. To be competitive in a typical gas-fired asset, the lender would have to be at around 1.3x to 1.35x over the life of the PPA. There is downward pressure on this because there a lots of banks that would love to be part of these types of deals. We do not see many PPA deals in the gas market. Holdco consolidated coverage ratios are as tight as 1.1x for sizing on debt service.

The other market segment is quasi-merchant gas. These deals are slightly more complicated. They are where most of the action on refinancings has been occurring. Lenders size at 1.15x revenue from the capacity price and revenue puts. If there is a heat-rate call option, sizing is based on a debt-service coverage ratio of 1.4x.

We have been assuming flat capacity forecasts in areas such as PJM and the New England ISO. We are seeing increasing use of cross commodity net-back hedges, which basically lock in a spark spread so long as commodity prices trade within a band. We size loans based on this type of cash flow at 1.5x based on a conservative downside case. Sometimes we are open to giving credit on a conservative merchant energy revenue forecast. We would probably use a 2.0x to 2.5x debt service coverage ratio. The issue becomes what balloon repayment levels we are willing to accept at loan maturity. The answer varies by location, age and technology.

MARTIN: So you will assign value to the merchant tail after the power contract ends for how many years at 2.0x to 2.5x?

CHO: It depends on where the project is located and the age of the asset. We certainly do not want to go past the remaining useful life of the asset.

MARTIN: Right, but you may go out to the reasonably expected useful life?

CHO: We will probably want a cushion, and we are using conservative downside price forecasts.

MARTIN: Some banks have been willing to accept only 8% rather than 10% sponsor equity. Where do you think the market is?

CHO: No lender wants to lend on an asset where there is no equity value. The sponsor has to have skin in the game. That level sounds right, especially for renewables, but not so much for thermal assets. When it comes to renewables, there is clearly a halo effect where lenders are driven more by ESG and less by economics.

MARTIN: So here is another phrase that will go into the annals of this industry. The earlier phrases were “wall of money,” “satchels of Euros” and now “halo effect” for renewables.

Are there any other noteworthy trends in the bank market as we enter 2020?

CHO: There are a lot of new trends, but I will just name a couple. Seoul, South Korea is a hotbed of capital, whether you are looking for senior debt, mezzanine debt, limited partner commitments, equity investments, every one of these types is available and open for business. It is a super-efficient source of capital. Capital is available in all sizes from small to large.

The delay potentially for another year of the PJM auctions of 2022 and 2023 capacity could sideline a slew greenfield gas-fired power projects that had been expected to come to market. However, the bank refinancing markets appear to remain open. About $2 billion in term loan refinancings are scheduled for closing over the next 30 to 45 days.

I expect continued capitalizations as borrowers take advantage of the different competing sources of capital I mentioned. It can go from A loan to B loan, B loan to A loan, and A loan to private placement, as capital moves from one pocket to another.

The last point is that structures in the renewable energy market will continue to evolve as a growing number of lenders accept merchant exposure. Lenders are always going to be looking for ways to drive higher-yield products.

Term loan B

MARTIN: Good list of trends. Let’s move to Jean-Pierre Boudrias at Goldman Sachs. What was the term loan B volume in the North American power sector in 2019? How did that volume compare with 2018?

BOUDRIAS: In 2019, we saw $4.6 billion of term loan B lending across the power sector. That means that the market was essentially flat from 2018 to 2019. Last year, about half of the volume was in the form of repricings. We did not see any repricing in our market last year, so volume remained flat versus 2018.

MARTIN: How many transactions were there in 2019?

BOUDRIAS: We saw 12.

MARTIN: So the number of transactions was up if one focuses solely on new money. Last year, there were 18 transactions in total for $8.25 billion, but about half of that 
was repricings.

BOUDRIAS: That’s right.

MARTIN: When should a CFO turn to the term loan B market rather than the bank market?

BOUDRIAS: It is really a question of either wanting more leverage or looking to finance projects that will have more merchant exposure.

MARTIN: You just heard Ralph Cho say that the banks are salivating over higher yields and looking for merchant exposure. Will that cut into the term loan B market this year?

BOUDRIAS: The total B loan market is $300 billion in size. We have always been talking about a relatively small subsector of the market, so the market is unlikely to be significantly affected by what banks are willing to do. While banks can invest in the types of transactions that tend to seek financing in the term loan B market and in some cases they do, I do not see bank loans and B loans competing directly.

For example, the term loan B market has been a better conduit than the traditional bank market for acquisition financing just because of the efficiency and ability of some of the underwriting participants who underwrite transactions quickly and then distribute the paper efficiently.

MARTIN: To be clear, the term loan B market is basically bank paper that is sold to institutional lenders. B loan deals are set up so that there will not have to be a lot of future interaction between the holders of the term loan and the borrower.

BOUDRIAS: Correct.

MARTIN: There is also a term loan C. What is it?

BOUDRIAS: The phrase term loan C refers to things that should have been provided either in the form of letters of credit or bank guarantees. What market participants do is create a funded tranche that can be sold to the market with that tranche marketed to institutional investors rather than being provided by banks.

MARTIN: One of the interesting stories last year was that bank rates were falling. There still seems to be downward pressure on interest rates as we head into 2020. Last year, term loan B pricing was moving in the opposite direction from the bank market. Rates were going up. What do you expect in 2020?

BOUDRIAS: Last year was a tale of two markets. The relative attractiveness of the fixed-rate high-yield market relative to the leveraged loan market meant that the loan market on a relative basis was not super busy until probably November or December. Now we are in the midst of a wave of repricings of the kind we have not really seen since 2018.

MARTIN: What volume do you expect this year?

BOUDRIAS: My sense is if we are going to have repricings, we should be closer to what we saw in 2018, so closer to $8 to $10 billion compared to the $4.5 billion we saw last year.

MARTIN: Pricing at this time last year for BB credits was around 350 to 375 basis points over LIBOR with a 1% floor and 1% OID. Single B credits were pricing at LIBOR plus 450 to 500 basis points. Where do you see pricing today as we start 2020?

BOUDRIAS: It is difficult to go tighter than 350 to 375. That said, I am convinced that we will break into the 200-basis-point level at some point this year for BB credits. For single B, we are a lot tighter than we were last year. A new single B would issue in the 400 to 425 range today.

MARTIN: So pricing is improving for borrowers. Are B loans still for seven years?

BOUDRIAS: Yes.

MARTIN: Advance rates in the B loan market have been in the mid-60% range. Has there been any change as we enter 2020?

BOUDRIAS: There is no magic formula that solves for advance rates, but my guess is advance rates will remain 60% to 70%.

MARTIN: How does a developer determine how much he can borrow in the B loan market?

BOUDRIAS: It is similar to what Ralph Cho talked about for the bank market. B loans tend to be sized on what we call lender cases. We will assume certain merchant energy prices, assume capacity prices will remain flat or are declining and then look for enough cash available to pay down a portion of the debt over the life of the loan.

For example, targeting 50% of the loan to be repaid over seven years usually means, as a rule of thumb, that we will lend six to six and a half times EBITDA.

MARTIN: How large a transaction must one have to make it worth the trouble?

BOUDRIAS: There is not as much liquidity for loans below $500 to $750 million, but with that said, debt as small as $225 to $250 million can be placed in the institutional market.

MARTIN: Are there any other new trends in the B loan market as we enter 2020?

BOUDRIAS: One of the biggest trends is the arrival of direct lenders. This started across the middle market and now it is spreading to larger transactions. Institutions are announcing transactions that they are handling on a sole basis that can be just shy of $1 billion. I talked about coming to the term loan B market for $225 to $250 million at a minimum. I would say that $125 to $200 million is a sweet spot for these direct lenders.

Some of these participants may be grey market lenders that Ralph Cho included in his count. It is something that we had not seen as much of before.

Project bonds

MARTIN: Let’s move to project bonds then and John Anderson with Manulife.

Project bonds are long, cheap, fixed-rate money. The project bond market does not tend to do well when the bank and term loan B markets are wide open and looking for product. That seems to be the case this year. Interest in project bonds revives when people fear interest rates are headed up.

It did not seem like there would be much of a market at the start of 2019. Was there one last year?

ANDERSON: The project bond market is a subset of the private placement debt market. It was pretty stable at about $100 billion in 2019. It was about a $100 billion the year before.

About 20% of the flow last year was infrastructure debt, the new term for project debt and public-private partnerships or PPPs.

MARTIN: Are these US or global numbers?

ANDERSON: Project bonds that syndicated in US dollars. It includes some international projects.

MARTIN: But mainly US projects?

ANDERSON: That’s right. It will be mainly US. We are doing about $12 billion a year of debt and infrastructure equity. That is about 60% in US dollars, 20% in Canadian dollars and 20% euro, sterling and Australian dollars. The dominant opportunity, whether it is in the US or Latin America, is US dollar flow. Even within that, Latin America is relatively small.

Of that, $20 billion was infrastructure. That is syndicated project finance debt. We know that there is a large amount of direct lending also occurring in smaller projects that does not come into the syndicated market. The size of the market is opportunity constrained rather than investor constrained.

To your point, we are the guys at project conferences with the badge that says, “Talk to me about long cheap money.” We look for sponsors who want to lock in long-term base rates. They are looking at a community that will lend 30+ years against contracted cash flows. We give you a spread for life. It is the coupon throughout the life of the loan. There is no escalating spread in order to incentivize a refinancing that you see in some other markets.

We are seeing contracted power projects clear at spreads of about 175 to 190 over treasuries today. If 10-year treasuries are at 1.8% today, you are looking at coupons on project bonds in the 3.5% to 3.75% range.

The market is showing some willingness to look at partially merchant cash flows or clearing capacity markets. The market has evolved beyond the traditional utility PPA to more and more corporate PPAs.

Every flavor of project finance can be done here: wind, solar, biomass, geothermal, hydro. We saw some distributed generation deals done this year.

MARTIN: How many active investors are there?

ANDERSON: Since we are not a syndication agent, we do not track that as closely as some other people do. We talked a couple years ago about there being 25 active investors. I think the number is higher today because I don’t see any life insurance companies that play in this market dropping out, and we are seeing European insurance companies becoming more interested in North America.

MARTIN: How large a transaction does one need to make it worthwhile to borrow in the project bond market?

ANDERSON: Syndicated transactions work best at $250 million and higher. If you come in below that, you are probably looking at doing something clubby in individual tickets of $25 to $50 million. A lot of our peers might do a smaller deal as a single-investor transaction. Thus, $25 to $50 million and up will work, but it just depends on how broad an investor base you want.

MARTIN: How long does it take to close a project bond deal from start to finish?

ANDERSON: It depends how fully baked it is. We have seen fully baked syndicated transactions make it to market in four to six weeks. The bond private placement market can move at least as quickly as, and sometimes more quickly than, the bank market. It takes longer where the transaction is not fully baked. Many lenders would be happy to work with you before all of your documents are done. Partly baked deals require a longer process. The key is to get a lead investor involved early as an anchor.

MARTIN: Must a borrower be an investment-grade credit?

ANDERSON: The market is deepest for investment-grade projects. The borrower need not be rated. It helps in a broad syndication, but project bonds can be placed without a rating if the lenders think the project is investment grade. The market will bid on BB senior-secured project finance paper. It is a sub-set of the broader market.

MARTIN: What is the loan tenor? Is used to be a year short of the PPA term.

ANDERSON: You can think of it essentially as the length of the PPA because project bonds are amortizing debt with stable debt service coverages throughout. By the time you get to the last year of the PPA, you are probably not putting a balloon payment at the end of the loan.

MARTIN: Are there any new trends in the project bond market as we enter 2020?

ANDERSON: We talked two years ago about the market turning away from coal in North America. We may be in the same position now internationally. We saw a few coal-fired projects get done in Asia over the last few years. Such projects have become much harder to do. There is strong demand for project bonds in the US. The ESG tailwind that Ralph Cho mentioned is definitely true in our space as well. If an insurance company finances a wind farm, the transaction goes up on its home page. People talk about it at the holiday party at the end of the year. There is a lot of feel-good around it among our employees.

Audience questions

MARTIN: Let’s move now to audience questions. Ralph Cho, someone asked, “How do you define North America in your deal volume results: US, Canada and Mexico or just US and Canada?”

CHO: US, Canada and Mexico.

MARTIN: Here is a question for the tax-equity investors. “How is the tax-equity market dealing with the merchant portion of deals? How much merchant are you doing?”

CARGAS: We are doing quasi-merchant wind and solar projects with hedges to put a floor under the electricity price. It would be difficult to do a fully merchant deal without a hedge. A few such transactions were done some years ago that have not performed terribly well.

MARTIN: There are a number of questions about community solar. Ralph Cho, how interested is the bank market in lending to community solar projects?

CHO: Community solar falls into the renewables category. There is appetite for it. Any well-structured, well-priced transaction will get a lot of traction. A limited number of banks have been looking at community solar deals, and I would throw other transactions like CCAs — community choice aggregators – and corporate PPAs that we have talked about offline, Keith, into the same bucket, but there is growing interest from banks in these types of transactions.

MARTIN: Tax equity guys, people are asking, “What pricing is available for community solar?” I know you will not comment on pricing, but perhaps you can say something about your general willingness to do community solar.

HENDERSON: Community solar is tougher for us to do. It does not have the volume that we are looking for, and the time and effort it takes to underwrite a community solar transaction is not very efficient from the standpoint of making best use of our people resources. Given the market volumes expected, particularly in wind, community solar could be a tough sell this year.

MARTIN: Question for all three lenders: “Will credit cover insurance unlock new transactions for unrated or lower shadow-rated credits?”

CHO: There is so much liquidity in the bank market that its role is fairly limited. No matter what level on the risk spectrum a project sits, there is a lender to fill that spot. How much does it is enhance the credit? It depends on the cost of that credit wrap and how much cheaper can I find a lender to lend at that level.

ANDERSON: I agree. Most investors in the project bond market would rather underwrite and price the underlying risk and get paid for that. We would probably see the underlying project economics as more durable than the wrapping financial institution would since the underlying project economics have good forward visibility for 10, 15, 20 or 25 years. We saw in the global financial crisis that many of the credit enhancement agencies kind of went away.

MARTIN: There are a lot of questions about energy storage. Let’s start with Ralph Cho. “Are you financing standalone storage, and if so, what are the debt-service coverage ratios and debt equity ratios?”

CHO: We are certainly open for business in financing energy storage. However, here is the issue with energy storage on a standalone basis. There are two storage models. A lender wants to see cash flows. If you give me very firm cash flows, the sizing coverages are going to be very tight and the pricing will be very tight.

On the flip side, we have also seen where the revenues are all over the place. It almost feels like we are being asked to take equity-type risks. If it looks like equity risk, it will not work in the bank market. If you can box the risk and show that the worst the project can do is X and the cash flows could go as high as Y, then maybe we can work with it. In deals with potentially volatile cash flows, the coverage ratio will be wider: call it 2.5x. We would have to get very comfortable with the underlying cash flows.

MARTIN: One of you mentioned that tax equity structures are changing. The question is, “Could you please elaborate?”

CARGAS: The days are gone when projects had 30-year PPAs with investment-grade investor-owned utilities. We have had to develop different ways of dealing with varying credits of offtakers for the electricity. There may be credit enhancements in some cases. There may be insurance. We have seen some fixed-flip transactions. There is so much more variety today in structures in place of what once was a commoditized structure.

MARTIN: Next question. “Has BEAT” — the base-erosion and anti-abuse tax — “caused any fallout in the tax equity market?”

CARGAS: It did a couple of years ago when a number of tax equity investors determined that they were subject to the tax. People exited the market. It was a problem for a number of sponsors.

I think many of those investors have remained out of the market, but one or two have determined that they are no longer subject to BEAT and have re-entered the market, which may be a positive for sponsors in this heavy-demand year.

MR. CHO: Keith, you know this subject better than we do, but my understanding is that the IRS guidance that came out on BEAT at the end of last year was fairly helpful for financial institutions and may have given some people who were concerned a little bit more breathing room. My sense is BEAT is not a significant issue currently in the market.