Cost of capital: 2018 outlook
The tax equity market managed to function all year in 2017 despite uncertainty about what the US tax code would say. Lenders complained about a shortage of deals to finance. More than 2,400 people listened as a group of project finance industry veterans talked in January about the current cost of capital in the tax equity, bank debt, term loan B and project bond markets and what they foresee ahead.
The panelists are John Eber, managing director and head of energy investments at J.P. Morgan, Jack Cargas, managing director in renewable energy at Bank of America Merrill Lynch, Ralph Cho, co-head of power and infrastructure finance for North America for Investec, Jean-Pierre Boudrias, managing director and head of project finance at Goldman Sachs, and John C.S. Anderson, head of corporate finance at Manulife Financial/John Hancock. The moderator is Keith Martin with Norton Rose Fulbright in Washington.
MR. MARTIN: John Eber, what was the tax equity volume in 2017, and how did it break down between wind and solar?
MR. EBER: J.P.Morgan estimates it was $10 billion in 2017, so it was down about $1 billion from the $11 billion we saw in 2016. The total volume breaks down to about $6 billion of wind and $4 billion of solar tax equity. Wind was down about $400 million compared to the previous year, and solar was down about $500 million.
There were 15 investors in the wind market in 2017. That is the same number as in the previous year, although there were three new investors who replaced three who invested in 2016, but did not invest in 2017. There were 27 wind deals in 2017 versus 28 the previous year and 18 different sponsors.
Turning to solar, we saw about $2.3 billion in utility-scale projects, which is almost flat against the previous year, and we saw about $1.5 billion in residential solar as compared to $2.1 billion the previous year. The drop off in residential solar was not a surprise. Many of the residential solar companies were purposely scaling down their growth rates and also promoting sales over power purchase agreements and leases.
MR. MARTIN: You said there were 15 tax equity investors in the wind market. How many in solar?
MR. EBER: Solar is a little harder to track than wind, but our estimate is 15 to 17 in the solar sector. There is overlap, but the investors in the two markets are not identical. We estimate there are about 35 tax equity investors in the market as a whole.
MR. MARTIN: We understand there were 10 other investors in 2017 looking to invest on a syndicated basis alongside more experienced investors. Is that also your count?
MR. EBER: That is possible, but it is not something that we track closely.
MR. MARTIN: Putting the numbers into perspective, I have from past calls annual tax equity volumes of $6.5 billion in 2013, $10.1 billion in 2014, $13 billion in 2015 and, as you said, $11 billion in 2016.
Jack Cargas, was it a surprise that the market found a way to function as well as it did in 2017, given that you had an entire year of uncertainty about what the tax code would say?
MR. CARGAS: We were a little surprised that the market remained as vibrant as it did. There was a lot of competition for specific transactions even though people knew that the hammer was going to fall late in the year.
The only constant in this market over the years has been constant change, and that was certainly the case for 2017. The market found a way to adjust for the tax-change risk. The structures for handling this risk continued to evolve throughout 2017 for the benefit of both sponsors and tax equity investors to the point that projects were essentially protected except for the late curve balls we saw, like the new base erosion and anti-abuse tax and the near miss on the corporate alternative minimum tax.
MR. EBER: One key to why the market worked as well as it did in 2017 is there was growing confidence by late spring that the corporate tax rate would remain 35% through the end of the year. Many of us switched to accelerating depreciation into 2017 by taking the 50% depreciation bonus. That mitigated potential effect of the tax rate change. The rate change was predominantly going to affect the value of the depreciation.
MR. MARTIN: In the past, tax equity investors have not been keen to take bonus depreciation. Do you think they will revert this year to past practice?
MR. EBER: The bonus has essentially increased with the 100% expensing. It will be an even heavier lift for investors to take at 100%.
MR. MARTIN: Jack Cargas, do you expect investors to ask sponsors not to take the bonus?
MR. CARGAS: We are still evaluating our position. The calculus has been complicated by the base erosion and anti-abuse tax, or BEAT. This is an area where investors could try to differentiate themselves.
MR. MARTIN: John Eber, what do you expect in 2018 in terms of deal volume?
MR. EBER: I expect it to be fairly constant. The number of projects needing financing in the wind and utility-scale solar markets is continuing to grow. Tax equity volume should remain in the $10 to $11 billion range based on what we are seeing.
MR. MARTIN: Jack Cargas, same sense?
MR. CARGAS: Yes. I think we all predicted a slightly higher volume in 2017. There were various reasons for the slight drop off from 2016. We expect the market to remain flat to slightly up.
MR. MARTIN: We enter 2018 with a lower corporate tax rate of 21%. We have a base erosion and anti-abuse tax that could claw back up to 20% of tax credits claimed in 2018 through 2025, and we have a 100% depreciation bonus for assets put in service through 2022 with a phase-out after that. John Eber, going back to you, how do you expect these changes to affect the market?
MR. EBER: That is a tough one. This is my personal opinion and not a J.P.Morgan official view. I expect tax equity capacity to tighten because there are a lot of changes and they trend toward the negative. The BEAT not only exposes 20% of the tax credits you might be taking on new deals to potential claw back, but it also affects the value of the remaining tax credits across your entire current investment portfolio.
That has to cause some investors to think about how much more they might want to invest versus invest at all.
Then, the drop in the corporate tax rate from 35% to 21% is a substantial drop in overall tax capacity, particularly when you consider that there is a 75% limit on how much an investor can reduce its tax liability with tax credits.
These two changes will affect tax capacity. The $64,000 question is how much and to what extent.
MR. CARGAS: There were also some positive effects from the tax bill. The after-tax returns to project owners should be higher due to lower tax rates.
As for the decrease in tax capacity, you have tax depreciation that is now worth 60% of what it used to be worth. This means that investors will have to think about where to allocate scarce tax capacity across the institution, whether you allocate it to low-income housing or other tax credit businesses and renewable energy finance and, if the latter, whether you have a preference for wind or solar or production tax credits or investment tax credits and what to do about the 100% depreciation bonus. The reality is it is early in the year, and investors are still answering these questions for themselves.
MR. MARTIN: Tax equity has accounted in recent years for 40% to 50% of the capital stack for a typical solar project and 50% to 60% for wind. Do you have a feel yet for where these numbers will settle in 2018?
MR. EBER: I had my team run a few examples. These are from half a dozen live deals on which we are working. On average, the upfront investment in a wind deal looks like it will be about 8% less with a 21% rather than a 35% tax rate. Tax equity started in the low 60s and is going down into the mid- to low 50% range.
For solar, we see a smaller impact on the order of a 3% reduction. Some of the deals at which we have been looking were in the low-40% range at a 35% rate and are now down to something like 39%.
MR. MARTIN: What percentage of tax equity investors do you expect to be subject to the BEAT? A small number? Large number?
MR. CARGAS: We are aware of one or two situations where tax equity investors have chosen to exit the market at least temporarily and that is due presumably to BEAT concerns. We do not think BEAT will affect a large number of investors. Our firm, Bank of America Merrill Lynch, lived up to all its client commitments last year, but exactly how and where BEAT will affect the market remains open to question.
MR. EBER: I think that’s right. It could be just a limited number of investors. However, keep in mind that there may be investors who are still in the market, but who may, as Jack was suggesting earlier, moderate how much they put into renewable energy versus other markets like low-income housing.
MR. MARTIN: For investment tax credit deals, the risk posed by BEAT is that a tax calculation later the same year the investment is made will lead to a claw back of 20% of the tax credit. Isn’t this a manageable risk? It is the same risk investors had before of predicting tax capacity for the year of investment. In view of this, do you think investors will give full credit for the investment tax credit in solar deals when deciding how much to invest?
MR. CARGAS: The short answer is yes. Investors are likely to be either in or out on ITC deals and, if they are in, they are likely to give full credit.
MR. EBER: There is a serious question as to whether there will be any carryforward under the BEAT. All the general business credits have carryforwards, which has financial statement implications. This point may cause the market to look at the BEAT risk differently than it did traditional tax capacity risk.
MR. MARTIN: Next issue. It will be possible in 2018 to claim a 100% depreciation bonus for the first time on used equipment. Do you see a move to sale-leasebacks of used equipment as a way for sponsors to raise cash?
MR. CARGAS: We do not expect to see much of this in the renewables market.
MR. EBER: I agree. I don’t think sale-leasebacks are likely to come back. The lease-buy analysis is not likely to tilt toward lease at this stage given the low interest rates.
MR. MARTIN: The cost of tax equity is a function of demand and supply. John Eber, you said you think there will be a contraction in supply of tax equity because of the reduction in the corporate tax rate. Jack Cargas, you said there was a lot of competition last year for deals, suggesting that there is low deal flow. Given this, in which direction do you sense yields are moving?
MR. EBER: I hope they are going to stabilize. As Jack said, it was a competitive market last year in which there was more than an adequate supply of tax equity. If my concern plays out in terms of contraction, it is really just a question of how much and how severe. Hopefully, the two curves will remain intersected near the current equilibrium.
MR. MARTIN: If anything, there was downward pressure on yields last year, so stabilize means no further fall.
MR. EBER: The question is answered by the yet-to-be-determined degree of impact of the BEAT.
MR. MARTIN: Are there any other noteworthy trends as we enter 2018?
MR. EBER: We continue to see a lot of corporate PPAs and hedge deals, and one thing we have been noting is that basis risk continues to be a major issue. We think sponsors are underestimating the size of the risk. It is growing in many markets. The issue is in deals with settlements at hubs instead of what happens with a traditional PPA where the power is taken at the bus bar.
MR. MARTIN: Basis risk in this context refers to a difference in pricing. The electricity is delivered in one place, but the price is established in another. Sponsors usually end up with that risk. How does that end up being handled in a tax equity deal?
MR. EBER: It affects the cash flow to the partnership which affects the cash flow that is available for the back-levered lenders and the tax equity. As between the tax equity and the sponsor, the sponsor bears the ultimate risk. However, with electricity prices so low, there is not a lot of excess cash to begin with, and that is putting pressure on a lot of these deals.
MR. MARTIN: Let’s move to Ralph Cho and bank debt. What was the volume of North American project finance bank debt in 2017 compared to 2016?
MR. CHO: Deal volume in 2017 was basically flat compared to 2016. Volume according to Thomson Reuters was about $40 billion spread across 124 deals. This compares to 2016 where we saw about $39 billion spread across 136 deals. The average deal size ticked up a little bit, but the volume was pretty much flat year over year. The market is down about 40% from the 2015 deal volume, which was the last strong year.
MR. MARTIN: How many active banks were there in 2017? How many do you expect in 2018?
MR. CHO: A lot more players came into the bank market in 2017. We estimate there are around 70 to 90 lenders with a number of new players from South Korea. I am including grey market lenders in this count. I expect the number of lenders to remain unchanged in 2018.
There are a lot of lenders looking for these types of deals. Total market capacity, meaning the size of power deal the market can handle at the upper end of the range, is probably about $5 billion and I would even go a little more specific and say of this group, maybe 20 to 30 lenders can probably book loans longer than 15 years and there is enough market capacity on 15-year debt to cover deals of up to $1 billion in size.
MR. MARTIN: LNG project financings were running $10, $11 and $12 billion.
MR. CHO: True. I was talking about power, but lenders definitely can hold much larger tickets on LNG deals.
MR. MARTIN: What is the current spread above LIBOR for bank debt?
MR. CHO: The spread has tightened from a year ago because of the intense competition among lenders for deals.
Plain-vanilla loans have probably fallen from LIBOR plus 1 3/8 to 150 over. Short-term construction loans could even be priced tighter at LIBOR plus 100. That is in the US. In Canada, we have seen pricing as tight as LIBOR plus 125. For whatever reason, Canada seems to get much more aggressive terms.
Quasi-merchant deals have remained amazingly stable at LIBOR plus 325. However, we are starting to see those levels cracking as we enter 2018, and we expect to see them come inside of the 325 level. More interesting to me is the plain-vanilla holdco loans, which traditionally price at LIBOR plus 400 and higher. Today you can probably get those deals done at LIBOR plus 300 to 350 basis points. We are aware of a couple situations where the spread has even been inside of 300. A lot more commercial banks are diversifying and increasing their risk appetite to take on those types of deals.
MR. MARTIN: To what do you attribute the large number of additional banks coming into the market last year? A year ago, you and others warned that there is a wall of money chasing deals. That does not appear to have deterred anyone from joining the fray.
MR. CHO: Focusing on the South Korean lenders, they look at the yield opportunity in the US compared to what they can earn in other places. They like the yield. We began seeing them in 2016 in quasi-merchant gas deals, and they liked LIBOR plus 325. But as those deals have started to dry up, they are looking to diversify.
As LIBOR has started to increase, they have been able to do tighter spreads and still earn attractive yields.
MR. MARTIN: What are current loan tenors? You mentioned 20 to 30 banks are willing to go 15 years and perhaps even to 18 years.
MR. CHO: Typical loan tenors today are five to seven years with mini-perm features. We see that for a lot of the acquisition financings. Construction loans are obviously much shorter. But for plain-vanilla financings, we have seen banks go as long as construction plus 18 years, especially if the project has a good long-term PPA.
We have even seen back-levered deals go out as long as 15 years. Competition for deals remains strong, and banks want to do these types of deals.
MR. MARTIN: The competition helps. What are current debt service coverage ratios for wind and solar?
MR. CHO: There has not been much change. For wind, the typical DSCR is 1.4x on a P50 basis and 1.0x on a P99 basis for contracted projects. DSCRs for hedged projects are slightly higher. In solar, the ratio is 1.3x on a P50 forecast, and usually you have no problem hitting the 1.0x under the P99 numbers.
MR. MARTIN: Are you seeing any merchant solar deals?
MR. CHO: There is definitely talk about merchant solar where sponsors are looking for value in a residual tail beyond the PPA, especially because people have to take a view if they want to be competitive when bidding to acquire these types of assets. Given the competitive landscape, more and more banks appear willing to accept some amount of merchant tail after the PPA.
MR. MARTIN: A merchant tail, but not an entirely merchant or even quasi-merchant deal?
MR. CHO: A fully merchant deal will still be tough.
MR. MARTIN: Most debt in the renewable energy market is back-levered debt. It is behind the tax equity in the capital structure. You said tenors for back-levered debt are running out as long as 15 years. I assume the coverage ratios are the same as for front-levered debt?
MR. CHO: There is really not much difference between back-levered and front-levered debt at this point. Lenders have not really asked for much of a premium to lend on a back-levered basis. We talk about this among the bankers, and I am not sure it makes sense, but that is where we see the market.
MR. MARTIN: Last year, you said banks have been requiring a 12.5-basis-point premium to lend on a back-levered basis. We heard from some other banks that the premium is 25 basis points. Now it sounds like there is no premium at all.
MR. CHO: You might be able to pick out a couple instances where lenders try to collect a small premium, but for the most part, I don’t think a premium is required.
MR. MARTIN: Some lenders during 2017 were considering crediting two to three years of additional revenue past the PPA term as a way of justifying increasing advance rates on loans. That is what you referred to earlier as crediting some amount of merchant tail. In what other ways is the intense competition playing out besides the pressure on yields and merchant tails?
MR. CHO: The pressure is not just in the renewable energy sector. We do this on gas. It is not just PPAs, but also hedges. Banks will take a view based on what they believe a good balloon value can be for the asset. We have seen leverage levels increase as lenders agree to somewhat more aggressive assumptions.
For some banks to be a little more competitive, we have even seen clever ways to increase leverage by shrinking scheduled amortization amounts while maintaining some level of acceptable coverage. The rest of the loan is paid down using cash sweeps. It is a very competitive market for any bank that wants a part of club or syndicated deals.
MR. MARTIN: Are there any other noteworthy trends as we enter 2018?
MR. CHO: M&A and re-pricings will be a significant part of the 2018 deal volume. Greenfield quasi-merchant gas deals — especially in PJM — will be slow. Not zero, but slow. In order to diversify and keep doing deals, lenders have expanded how they define infrastructure projects to create more opportunities for themselves.
We have touched on growing merchant exposure to renewable energy assets. Clean tech is the other trend. These deals still tend to be on the small side, particularly ones with battery storage applications, but we are starting to see a lot more of these types of opportunities.
Term loan B
MR. MARTIN: Moving to the term loan B market, J-P Boudrias, what was deal volume in that market in 2017, and how did it compare to 2016?
MR. BOUDRIAS: Last year, we saw about $10 billion in volume in the term loan B market in the US power sector. That compares to $7 billion in 2016, so there was an increase. But the increase was timid when compared to the overall B loan market.
The US leverage loan market saw total issuances last year of $938 billion, and that number was up almost 100% compared to the previous year. That number includes $434 billion of repricings. Primary issuances were $503 billion, which was 50% more than the previous year on aggregate. Despite the increase in volume, lenders and investors were looking for product and, as a result, gave permission for a lot of transactions to get repriced.
MR. MARTIN: There were 11 B loan transactions in 2016. What was the number in 2017?
MR. BOUDRIAS: The same count and roughly $6 billion of the $10 billion were repricings. Only $4 billion were new-money deals.
MR. MARTIN: The gross figure of $938 billion was the entire US B loan market?
MR. BOUDRIAS: Correct.
MR. MARTIN: How many 2017 deals were merchant gas?
MR. BOUDRIAS: Most of them were in that category. There was one coal transaction, and there were only three renewables transactions.
MR. MARTIN: Were all of the merchant projects in PJM or ERCOT?
MR. BOUDRIAS: The last ERCOT deal was in early 2015, so almost all of the volumes have been for transactions in PJM, New York and New England.
MR. MARTIN: What do you expect in 2018?
MR. BOUDRIAS: We expect the same as 2017. When one thinks about the source of new transactions in our market, it is refinancing of projects that were not financed originally in the B loan market and financings of new assets that will enter the market via M&A. The M&A market does not have a very strong backlog at the moment, so we may be looking at continued tightening of pricing that may lead to more refinancings of deals that were originally done in bank market as they come into the term loan B market.
MR. MARTIN: Pricing a year ago for strong BB credits was around 350 basis points over LIBOR with a 1% floor and 1% OID. Single B credits were pricing at 425 to 450 basis points over. We heard Ralph Cho say that, in the bank market, quasi-merchant is getting 325 over LIBOR, trending down. I assume the term loan B rates have come down as well. Where are they today?
MR. BOUDRIAS: BB credits are probably 325 over. A number of independent power producers that tend to be treated more like corporates have all repriced in December, taking between 25 and 50 basis points off their spreads.
It is not unforeseeable that the 325 I just gave may be on the high end and that you may see some of the deals that were repriced in the fall come again for potential repricing that may tighten further to the tune of another 25 basis points.
MR. MARTIN: That is 25 basis points improvement over last year, so 325 but trending down.
MR. BOUDRIAS: That’s right, and single B is probably the same thing. The right range for single B credits is probably in the 400 to 425 range.
MR. MARTIN: Explain what a B loan is for anyone who is unfamiliar with the term.
MR. BOUDRIAS: It is a loan that is documented the same way as a bank loan, but the lenders are institutional lenders. The biggest difference will be in the amount of refinancing or merchant risk that lenders will be willing to take, and obviously it is a capital markets execution, so the transaction when it is launched generally will get done over a 10-day period, during which the debt participations are allocated and then it trades. For example, a normal $500 million transaction will probably have something like close to 40 lenders. B loans tend to be used in riskier projects like quasi-merchant deals.
MR. MARTIN: Are B loans still for seven years?
MR. BOUDRIAS: Yes.
MR. MARTIN: How does a developer determine how much he can borrow in the B loan market?
MR. BOUDRIAS: We have not seen new construction in our market for some time. The last new-build deal was in the fall of 2015. Advance rates historically have been close to the mid-60s. Obviously it will depend on a variety of factors. I may have touched on them last year if anyone has access to the February 2017 NewsWire. The US government issued leveraged lending guidelines in 2012 that set a limit on the amount of leverage permitted for most transactions. In general, deals do not exceed six times leverage, require 50% repayment over seven years and have a loan-to-value ratio of 75%. That is what the federal bank regulators are looking for, and that is generally where you see transactions cap out in the institutional market.
MR. MARTIN: Are there any other new trends in the B loan market as we enter 2018?
MR. BOUDRIAS: Demand for B loans in the power sector has not kept pace with the supply of capital that has been assembled to pursue the opportunity. The volume of power-sector B loans has grown less rapidly than the volume of B loans in the broader economy.
MR. MARTIN: Let’s move to John Anderson and project bonds. The project bond market does not do well when the bank and term loan B markets are wide open and looking for product. Ralph Cho said 70 to 90 banks are now chasing deals. On the other hand, interest rates seem headed up. The yield on 10-year treasuries spiked yesterday at a little over 2.6% before backing off slightly. How many project bond deals were there in 2017, and where do you see the market headed this year?
MR. ANDERSON: The project bond market is long, cheap, fixed-rate money. The long loan tenor is its competitive advantage, and we tend to see people come to this market when they are looking to lock in inexpensive base rates for fear that rates are on the way up. In terms of rates, notwithstanding some of the movement we are seeing this week, the overall cost of capital for project bonds declined last year as did rates in the broader corporate investment-grade market. High-grade fixed-rate credit came in by about 20 basis points year over year last year.
The 10-year treasury has remained flat over the last year at 2.5%, and the 30-year treasury is actually in by 10 basis points from 3% to 2.9%. Put all that together and combine it with increased investor appetite for the asset class, and we have seen spreads on project bonds come in a similar 25-plus basis points year over year as has occurred in the bank and term loan B markets.
MR. MARTIN: What is the current spread above 10-year treasuries?
MR. ANDERSON: We are seeing spreads of 175 to 200 basis points for high-quality US deals, maybe a bit tighter in some cases, and that puts the coupon rate in the 4%-to-5.5% range, with most of it in the 4%-to-4.5% range for clean projects.
There has been a pretty stable flow of projects. We have seen gas-fired units used to back renewables. We have seen wind. We have seen solar.
Of the roughly half a dozen syndicated deals that we saw last year, the large gas units placed well. You have good visibility in the syndicated market for them. Some wind and solar financings are getting done on a syndicated basis. We saw both last year. But some projects, particularly community solar transactions, are too small for the syndicated or public market and are being done with one investor or in club deals in the private market.
From our own activity and talking to other lenders, we think volume was flat year over year. The cost of capital fell both in terms of base rates and general corporate spreads. New investors came into the market. These are the same themes you heard about the bank and term loan B markets.
If you had 25 or so keen investors a year ago, you probably have 30 today. There has been increased interest from overseas investors. Some such investors have teams in the US. Some are using US advisors to find opportunities for them to lend.
In terms of trends, I agree with Ralph Cho’s comment. To clear renewable energy at an attractive return, you have to bid something for the merchant tail.
MR. MARTIN: Meaning to win the deal you have to give some value to the merchant revenue?
MR. ANDERSON: Exactly right. You cannot just count the contracted cash flows and get to something that works. The other thing that is interesting is it felt like 2017 was the year that the market went explicitly no-bid for coal or pretty darn close to that for coal-fired power plants. Granted that no new ones are getting built today, but in terms of secondary sales of project bonds that were already used to finance such plants, it felt like there was a chilling pull-back in terms of investors that were willing to buy such bonds last year.
MR. MARTIN: Very interesting. How many deals are in the pipeline as we start the year?
MR. ANDERSON: Probably about the same as last year. We have about half a dozen between things in the syndicated and club or direct markets that we are looking to work on. Our transactions tend to move quickly, so a point-in-time snapshot of the deal pipeline ends up being not that representative of what happens over the year. In our market, four weeks to place a syndicated transaction is plenty of time. You do not always get much forward visibility on what is coming.
MR. MARTIN: You said there were around six syndicated deals last year. Do you have any feel for the volume of private transactions?
MR. ANDERSON: It is anyone’s guess because there is not good data on the private market. In general, you probably have more capital flowing through the syndicated market and less capital through the club market, but probably a higher number of transactions in the club market because they tend to be smaller. ¥