Cost of capital: 2017 outlook

Cost of capital: 2017 outlook

February 15, 2017 | By Keith Martin in Washington, DC

The Trump presidency and the likelihood of corporate tax reform have added an element of unpredictability to the year ahead. More than 2,000 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 for North America for Investec, Tim Chin, managing director and head of power, infrastructure and project finance in North America for BNP Paribas, Jean-Pierre Boudrias, managing director and head of project finance at Goldman Sachs, and John C.S. Anderson, head of North American corporate finance at John Hancock/Manulife. The moderator is Keith Martin with Chadbourne in Washington.

MR. MARTIN: John Eber, what was the tax equity volume in 2016? How did it break down between wind and solar?

Tax Equity

MR. EBER: We are still finalizing our data on the commitments that were issued in 2016, but based on our current analysis, we see a market that was about $11 billion of new tax equity commitments in 2016. This breaks down to a little over $6 billion in wind and the rest in solar.

This is down from last year’s market that was almost $15 billion in tax equity. The big difference seems to be in the utility-scale solar side, which saw a significant uptick in 2015, but was down to a much smaller number in 2016.

MR. MARTIN: Why the drop off in utility-scale solar? This is a significant drop of almost $4 billion in tax equity.

MR. EBER: The interesting thing is the market is about the same size as it was in 2014, so we had a big increase in 2015 of over 40% due predominantly to the utility-scale solar market. The expected expiration of the investment tax credit led to a lot more commitments being sought in 2015. The market in 2016 fell back to its previous level, which was still sizable in historic terms. In 2013, the market was $6.5 billion in total between wind and solar.

MR. MARTIN: So when the tax credits were extended by Congress at the end of 2015, the market was fairly exhausted. People stopped to catch their breath; 2016 got off to a slow start. The pipeline of utility-scale solar projects had thinned and now has to be rebuilt.

MR. EBER: I guess you could say that. Many people thought that might happen. I think you yourself asked me if I thought that was going to happen, and we didn’t know for sure, but the numbers suggest that it did.

MR. MARTIN: Jack Cargas, journalists were asking at year end whether the tax equity market has frozen in anticipation of corporate tax reform. Has it?

MR. CARGAS: It has not. The market continued to function normally after the November election. The market was still robust in 2016 and that continued through the end of the year. Transactions that closed or funded post-election had different structures in some cases, and they ran the gamut from deals that did not change at all from their pre-election assumptions all the way across the spectrum to deals that included rights to restructure completely in the event of a change in tax law or proposed change in tax law. It seemed that most investors found a way to deliver for their clients before year end.

MR. MARTIN: What do you expect this year? We could be in the midst of a tax reform debate for a good part of the year. Do you think the market will continue functioning normally throughout that debate?

MR. CARGAS: We certainly are in the midst of tax reform debate, but we are in a different place than we were a month ago. There is no immediate pressure to fund now that we are past year end. We and many other market participants are taking a hard look at various regulatory scenarios and what kind of impact they might have on structures.

I think the market is generally of the view that there are two areas of potential change that would affect deals. The first is the tax credit regime. Tax reform could lead to a reduction in or even elimination of the investment tax credit and production tax credits. However, I think most market players believe that such changes are unlikely.

The second change is a likely reduction in corporate tax rates, including those suggested by the president elect during the campaign and in the House Republican blueprint. Obviously various proposals have been floated, including reduction of the marginal corporate tax rate from 35% to 20% or even 15%, and assumptions about the magnitude and timing of any such reduction can have significantly different impacts on transaction economics.

So we expect lots more analysis, and we expect structural changes. The structural changes could include things like the election of bonus depreciation and possibly lower investment amounts per transaction, but as we see on a daily basis in the press, expectations about what might happen keep shifting.

MR. MARTIN: John Eber, what do you expect this year during the tax reform debate?

MR. EBER: I think Jack covered it pretty well. People are trying to analyze all the possible permutations. There is the potential for the tax rate, the ITC, PTCs and depreciation to change. We agree with Jack that it seems most of the attention is focused on tax rates.

Even there, you have a lot of permutations as to which rate and, even more importantly, when that rate might be effective. If the new rate will not take effect until a future year, then I think the debate in Congress will have a limited effect on the market in 2017. If anything, people might want to accelerate deals to claim tax losses against a higher tax rate.

MR. MARTIN: Both of you suggest tax equity investors are more likely to elect bonus depreciation. They had not been willing to price it into deals very frequently in the past.

MR. EBER: That’s correct. Let’s step back. The risk of tax rate change has been borne by the developers for years, because there is both a potential benefit and a potential detriment from rate changes, depending on when they occur during the term of the deal. If you do not make any structural changes, the flip would move out as the tax rate is reduced. Electing bonus depreciation is one way to mitigate some of the effects of a potential tax rate change.

MR. MARTIN: Will the lower corporate tax rate reduce the potential supply of tax equity, or is tax capacity unlikely to be a constraint since most of the tax equity investors are banks and insurance companies with potentially unlimited tax capacity, even at lower rates?

MR. CARGAS: It is a little too early to tell, but it is possible that the availability of tax equity could be reduced going forward. A few investors could exit the market due simply to the massive uncertainty or to the expectation that corporate tax rates will be reduced with the result that these investors will no longer need as much tax shelter.

MR. EBER: It is hard to tell because most people’s tax positions are confidential, so these numbers are not readily available.

MR. MARTIN: Is tax capacity a constraint for big banks?

MR. EBER: Not for J.P. Morgan.

MR. MARTIN: I assume the answer is the same for Bank of America Merrill Lynch?

MR. CARGAS: I think most financial investors are going to stay in this business. Their business is to provide financing to their customers.

MR. EBER: I agree. More investors are active in the market compared to four or five years ago. New investors came in last year. There are investors who wanted to invest last year who were unable to find deals, so I think that the market on the tax equity side is as healthy as it has been for a while.

But again, there are a lot of different tax rates being bantered around and no transition rules have been announced so, as Jack said, it is difficult to know the potential effect on the market.

MR. MARTIN: The cost of tax equity is a function of demand and supply. I think we heard that supply is not likely to increase. It could decrease somewhat. How do you see the demand curve moving, and ultimately the question is in which direction do you sense yields are moving?

MR. EBER: I think the demand side has been stable. We saw a big spike up in 2015, but 2016 brought us back to a more normal level of demand. The wind side on tax equity has been stable, in the $6 to $7 billion range for the last three years in a row, and I think the solar side has the potential to stabilize as well. We see modest continued growth in demand going forward, assuming no revocation of ITCs or PTCs, and the supply of tax equity seems to be more than adequate to accommodate that growth.

MR. MARTIN: Two more questions. Jack Cargas, what percentage of the capital structure is covered by tax equity today in the typical wind or solar project?

MR. CARGAS: In wind, tax equity usually amounts to 60% of capital stack. We have seen higher and lower figures depending on the amount of cash available for distribution to the tax equity investor, but 60% is pretty typical for us. In solar deals, it has been more like 50%.

MR. MARTIN: John Eber, do those numbers sound right to you?

MR. EBER: Yes they do.

MR. MARTIN: I could swear the two of you said during the tax equity panel we did together at the AWEA fall finance conference that tax equity amounted to 40% to 50% in a typical wind project.

MR. EBER: We do not usually see it below 50% in wind projects. Generally wind is in the 50% to 60% range. Solar is in the 40% range.

MR. MARTIN: Last question. Are there any other noteworthy trends in the tax equity market as we enter 2017?

MR. CARGAS: A trend we have seen the last few years is that transactions have become more and more complex, and we can expect that trend to continue. We have seen the inclusion of back leverage and RECs and hedges and basis risk and environmental issues and corporate PPAs and many other developments. The market will have to evolve further to accommodate new political uncertainties.

We expect investors to continue to add more tools to their toolboxes. We are confident the market will continue to exhibit the adaptability necessary to continue to finance these assets.

MR. MARTIN: John Eber, other noteworthy trends?

MR. EBER: Several sources have recently reported on their estimates of the volume of wind turbine equipment that was delivered by year end 2016, or will be delivered in early 2017, as a means to treat projects built over the next four years as eligible for full production tax credits. One source said costs were incurred in 2016 on enough equipment to treat 30,000 to 58,000 megawatts of projects as under construction in time. Another source had 40,000 to 70,000 megawatts. If you break that down on an annual basis, it is on average maybe as much as 10,000 megawatts a year of wind over the next four years that would qualify for PTCs at the full rate.

From a tax equity standpoint, that suggests there will be a healthy, active market for the next four years. That is really good for the wind business in terms of being able to compete against solar. The numbers do not include potential repowerings or projects on which construction started under the physical work test. The bottom line is that it looks like a vibrant wind market ahead for the next four years.

MR. MARTIN: What does 10,000 megawatts a year of wind translate into in terms of billions of dollars tax equity per year?

MR. EBER: The current market, in terms of tax equity translated into megawatts, is about 6,000 to 7,000 megawatts a year and, of course, not all wind developers use tax equity. Each year, we get a few more players who come in who can own and operate and not need it. The current market would be adequate to service that type of volume.

Bank Debt

MR. MARTIN: Let’s shift to bank debt. We have Ralph Cho from Investec and Tim Chin from BNP Paribas. Either of you, what was the volume of North American project finance bank debt in 2016 compared to 2015?

MR. CHO: From a dollar perspective, 2016 transactions totaled about $35 billion compared to about $60 billion the year before. That was about a 42% drop in dollar volume.

MR. MARTIN: Dramatic.

MR. CHIN: There were a number of huge LNG deals that came to market in 2015, but not in 2016.

MR. CHO: If you go a little more granular, by my count there were 137 deals in 2016 compared to 160 deals in 2015. In terms of number of deals, there was only a 14% drop. Another difference in deal volume is that there were fewer quasi-merchant gas-fired power projects done in 2016 compared to 2015. What was left were renewables deals, especially solar, and they tend to be smaller transactions.

MR. MARTIN: Do you expect LNG to make a comeback this year?

MR. CHO: I do not.

MR. MARTIN: How many active banks were there in 2016, and how many do you expect in 2017?

MR. CHO: During 2016, we worked with 50 to 60 project finance lenders who were actively hunting for deals. I think that this represents market capacity of about $3 billion. For 2017, we expect to continue to see new lenders, especially from Asia and particularly from Korea. They are attracted to the opportunities in the US market, and they are committing large dollars.

On top of that, we also see international commercial banks coming back into project finance, so it is a safe bet to say that there will be an uptick in liquidity in 2017. This is good for borrowers. It will make for more competition among lenders for deals.

MR CHIN: Probably only four or five banks on the list of active banks are leading transactions. With respect to Korean banks, most of that liquidity is being sourced by brokers and advisors. As Korean banks become more familiar with the US project finance market, we will probably see them lending directly or at least through the syndications teams of the global project finance banks.

MR. MARTIN: Ralph Cho, going back to you, we have been hearing for the last two years that the market is awash in liquidity, and you are saying there is more liquidity still coming.

MR. CHO: Correct.

MR. MARTIN: What is a good adjective to describe the next stage of liquidity beyond “awash”?

MR. CHO: More awash? Abundant?

MR. MARTIN: What is the current spread above LIBOR for bank debt, and what is it as a coupon rate?

MR CHIN: The spread on the transaction varies based on the type of financing. For a fully contracted project and plain-vanilla nonrecourse financing for a good sponsor, I would quote a range somewhere around 162.5 to 200 basis points as a margin above LIBOR. If you look at other sectors like quasi-merchant gas-fired power plants, they will obviously carry a higher spread: probably 325 to 350 basis points.

To calculate the coupon from a borrower’s perspective, add LIBOR. The three-month LIBOR rate today is approximately 1%. So add 1% to your spread and that is generally what lenders are receiving in the market. However, keep in mind that banks require borrowers to hedge the majority of their floating-rate exposure. Call it a 5-year average-life swap spread. We are quoting somewhere in the 2% to 2.25% range, so you would add that on top of your margin to calculate a coupon that the borrower would have to pay.

MR. MARTIN: Is the upfront fee equal to the LIBOR spread?

MR. CHIN: We are seeing the LIBOR spread a little bit wider than the upfront fee.

MR. CHO: It is more of a coincidence if you see the LIBOR spread and the upfront fee at the same level. Upfront fees reflect compensation for the work that lenders have to do and their balance sheet usage. LIBOR spread is more reflective of the bank’s cost of capital plus the return.

Borrowers should expect to pay an arranger fee and some original issue discount as the two elements of the upfront fee.

The arranger fee ranges between 50 and 100 basis points and is split among the book runners. If a lender is really desperate to be one of the lead banks, it may agree to a fixed fee. That could be less than this amount. It happens occasionally.

The original issue discount is a fee paid to the lenders that are committing to the transaction. The book runners like to offer an OID level that is tied to the amount each lender is willing to commit. For example, for retail-level tickets of $25 million or less, we offer lenders OID somewhere between 100 and 150 basis points, and the highest chair lenders who are committing $75 to $100 million or more are receiving somewhere in the neighborhood of 200 to 250 basis points.

If the borrower wants a firm underwriting commitment, an additional fee would be charged that is generally around 50 to 100 basis points.

MR. MARTIN: Some companies in the wider market are moving to borrow ahead of any move by Congress to deny interest deductions, figuring that existing debt will be grandfathered. Have you seen any greater interest in corporate or construction revolvers as a consequence of this?

MR CHIN: That is an interesting question. We have not seen that trend yet in the project finance market.

MR. MARTIN: Is it still the case that there is no LIBOR floor in the bank market?

MR CHIN: Correct.

MR. MARTIN: What are current loan tenors?

MR CHIN: The quasi-merchant stuff is usually done with a mini-perm structure over construction plus five or seven years. If the project has a long-term power purchase agreement, we could structure the debt amortization over the life of the PPA, but retain the shorter tenor so the borrower gets the benefit of the longer amortization schedule.

MR. CHO: I agree with that. Most of the market is around that five- to seven-year sweet spot. In some fully-contracted deals, you might have heard about longer tenors like construction plus 10 years all the way up to construction plus 18 years. Within the 50- to 60-bank universe that I mentioned earlier, I would say about a third of those banks would be willing to commit on a longer-term basis.

If I had to guess, I would say that the market capacity for long tenors all the way up to construction plus 18-year money would be in the $500 to $700 million range.

MR. MARTIN: Out of $3 billion.

MR. CHO: Correct.

MR. MARTIN: What are current debt service coverage ratios for wind, solar, and gas-fired power projects?

MR. CHO: They are all different. For wind, they are somewhere in the range of 1.4x to 1.45x debt service. For solar, you probably can size it a little bit tighter given the shorter standard deviation in your forecast for irradiation, so call it 1.3x.

We have been doing residential solar deals, and we have been sizing those at 1.5x just because of the nature of the customer agreements. For contracted plain-vanilla gas plants, you are probably looking at 1.4x to 1.45x.

MR CHIN: On the quasi-merchant side, we are seeing base case coverage ratios of a minimum of 2.0x to 2.5x, and an average of 2.5x to 3x, but the way we size the debt reduces the coverage ratio to a range of 1.25x to 1.3x.

MR. MARTIN: What does that mean?

MR. CHIN: Take a merchant gas-fired power plant with a hedge and capacity payments in PJM. We are not taking any merchant energy into consideration when we size the debt. We look only at the hedge and capacity payments. The coverage ratio looking just at those payments is 1.25x to 1.3x.

MR. MARTIN: What are advance rates currently on construction debt?

MR. CHO: Advance rates are generally based on what cash flows you expect after the plant starts operating. If it is a fully contracted plain-vanilla asset with healthy cash flow, the borrower should be able to leverage up to 80% to 85% on a senior basis.

When you start looking at deals like quasi-merchant deals where the cash flow is not fully locked in, then the leverage falls significantly. We are in the market today with one where the advance rate is slightly below 50%. The developer will have to put up a significant amount of cash equity.

MR. MARTIN: Are you describing advance rates for term debt or construction debt?

MR. CHO: We do construction plus term. We would start lending you money from day one when you first break ground.

MR. MARTIN: You would expect the equity to fund during construction on a pari passu basis, and the entire construction debt would roll into term?

MR. CHO: We have seen the equity fund in two ways. The equity can fund first or the equity can fund last as long as there is a letter of credit or some other kind of credit support behind it.

MR. MARTIN: For quasi-merchant gas, at no point would you be out of pocket as a lender for more than 50% of the capital cost.

MR. CHIN: I concur with that.

MR. MARTIN: One of the big stories in the last three years has been the increase in volume of back-levered debt in the renewable energy market. It is rare to see project-level debt in that market. How do coverage ratios, tenors, and pricing change as the debt moves upstairs behind the tax equity?

MR CHO: In 2016, approximately $15 to 20 billion of the volume was in back leverage. At Investec alone, we probably moved $1 billion just in residential solar, which was all back-leveraged debt.

To be honest, I have not seen much difference in terms of coverage ratios. Back-leveraged debt is still being sized against 1.3x in solar. Tenors are still five to seven years. Pricing is 175 basis points over LIBOR, give or take, based on the situation. In general, lenders are willing to take flip risk, but they want to put in structural mitigants to protect them from the flip risk.

The tax equity structures are slowly becoming more accommodating to lenders. The tax equity investors are structuring in protection of lenders’ principal and interest payments before sweeping in full for indemnity claims.

The new thing that we are starting to hear about is we see some creative structuring where tax equity investors are allowing the back-leveraged lenders to have a lien. That actually might take away back leverage. What the tax equity guys might ask for is for lenders to agree to forebear for five years and carve out preferred distributions to the tax equity investors in exchange for giving them a lien. Let’s see whether or not that becomes a trend.

MR. MARTIN: Very interesting. A 12.5 basis point premium for debt that is behind the tax equity in the capital stack. That is not much of a premium.

MR. CHO: No.

MR. MARTIN: The dollar has appreciated by 4% since Trump was elected. It is up 25% over the last two years. What effect, if any, does this have on participation by foreign lenders in the US market?

MR CHIN: I can only speak for my institution. I have not seen any effect.

MR. CHO: I agree with that.

MR. MARTIN: Are there any other noteworthy trends in the bank market as we enter 2017?

MR. CHO: The slowdown of GDP growth in Korea and the general maturation of the economy is causing Korean banks to look for higher-yielding opportunities in international markets.

Even though the US deal pipeline in general looks weak, our bank sees some potential areas of growth in 2017. We expect to see more residential solar aggregation facilities as well as aggregation takeout financings. We expect to see some consolidation of commercial and industrial solar financings through portfolios. We see utility-scale renewables activity ramping up in Mexico. Our institution is also chasing storage and infrastructure. We see these as the primary areas of growth for our business.

MR. MARTIN: Tim Chin?

MR CHEN: Banks will probably start talking more about being overexposed to PJM. That will lead to more use of hybrid debt structures, such as including a fixed-rate debt tranche as part of a larger floating rate financing. We also see more Korean debt appetite for these transactions. We would like to see more infrastructure deals coming to market.

Term Loan B

MR. MARTIN: Jean-Pierre Boudrias, let’s talk about the term loan B market. For our listeners who don’t know what a term loan B loan is, it is basically debt papered as bank debt, but sold to the institutional market. That means that there are fewer occasions when one needs to come back to the lenders for approvals.

What was the term loan B volume in the North American power sector in 2016, and how did that volume compare to 2015?

MR. BOUDRIAS: Last year, we saw $11.6 billion of volume across 11 transactions. That was a significant increase from the year before when we saw about the same number of transactions, just one less at 10, but the B loan volume in 2015 was only $3.3 billion, so there was a significant increase in B loan volume in 2016.

MR. MARTIN: It sounds like the market did bigger deals. What types of deals accounted for the increase in dollar volume?

MR. BOUDRIAS: It is important to remember where the term loan B market has traditionally been active. It has been used to support new M&A activity. We saw larger such transactions tap the market last year. There were also more refinancings than we saw the year before. Those refinancing volumes are generally easier to place when one thinks that the lenders already have exposure to them.

MR. MARTIN: What percentage of the 2016 deals — you said there were 11 — were merchant gas-fired power projects?

MR. BOUDRIAS: I would not limit it to gas projects. Merchant projects were probably 50% of the mix, and the balance was a mix of small contracted portfolios and larger retail-oriented companies.

MR. MARTIN: You heard Tim Chin say just a moment ago that one trend in the bank market may be a sense of growing overexposure to PJM for merchant gas deals. Do you sense that as well in the term loan B market?

MR. BOUDRIAS: I would describe the trend in the term loan B market as follows. We had a group of investors who are in almost all the transactions in 2012, 2013, 2014 and, to a lesser extent, 2016. When you look at the performance of a lot of these financings — a large component was merchant, and obviously you overlay what happens to natural gas during the same time period — you can see a fair amount of underperformance. As a result, in 2016, some of these investors decided to stay on the sidelines. They were the people who were probably overexposed to the sector broadly. We saw a group of new investors enter the market in 2016 for power transactions.

MR. MARTIN: Very interesting trend. The dollar volumes in the B loan market were $11 billion in 2013, $9 billion in 2014, $3.3 billion in 2015, and we bounced back up to $11.6 billion in 2016. What do you expect for the term loan B market in 2017?

MR. BOUDRIAS: My suspicion is it will be largely driven by M&A volumes. Some transactions are already known. LS Power is purchasing certain assets from TransCanada that obviously will be part of that volume. It is probably reasonable to expect that we will see a lower dollar volume than in 2016, but larger than what we saw in 2015, so probably in the $5 to $6 billion range. If we see a few large acquisitions, the dollar volume could increase above this range.

MR. MARTIN: Pricing for B loans tends to be higher than for bank debt. Pricing a year ago for strong BB credits was around 425 to 450 basis points over LIBOR, and B credits were 575 to 600 basis points over. Where do you see rates today?

MR. BOUDRIAS: For a BB name, probably around 350. That could even move lower when the markets open in 2017. For B credits, we are probably around 425 to 450 basis points over.

MR. MARTIN: We are in a market where people expect interest rates generally to increase and yet the term loan B rates are going down. Why?

MR. BOUDRIAS: It is important to remember those were all spreads. Unlike the bank market, there were LIBOR floors in 2016 in most deals of 1%. My suspicion is we will probably see LIBOR floors start to disappear in 2017.

MR. MARTIN: Because of competition or because the underlying rates are rising?

MR. BOUDRIAS: Investors were demanding LIBOR floors because, unlike banks, they do not fund in the floating-rate market. The larger investors in the term loan B market are what we call CLOs that are essentially structured vehicles who purchase loans. The structure of their funding tends to be fixed, so they have limited ability to deal with extremely low underlying rates. They tend to require that LIBOR have a floor so that they can service their own liabilities.

As LIBOR moves higher, we expect the floor to disappear given the expectations that LIBOR will continue to increase.

MR. MARTIN: Are B loans still for seven years?

MR. BOUDRIAS: That’s right.

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

MR. BOUDRIAS: It is really driven by repayment. The expectation has to be that a little more than half the debt principal will be repaid by the maturity date for the loan in the downside case. For acquisition financing, investors generally want minimum equity of between 30% and 40% depending on the profile of the asset.

MR. MARTIN: What upfront fees are required? We heard about the fees required by banks.

MR. BOUDRIAS: Similar to banks, there is a component that goes to investors. Generally it is around 100 basis points. The amount fluctuates depending on market conditions. Then there is compensation for the underwriters. The amount fluctuates depending on the nature of the underwriting.

MR. MARTIN: Are the total upfront fees something like 100 basis points, 150, 200?

MR. BOUDRIAS: It depends, but probably between 100 and 150 basis points for best efforts, and about 100 basis points on top of that for underwritten transactions.

MR. MARTIN: Final question. How large a transaction must one have to make it worth the trouble to go to the B loan market, and how long should one assume the transaction will take compared to a bank loan? What is your sense of how long it takes to close a bank loan versus a B loan?

MR. BOUDRIAS: For a new transaction or new borrower, $250 million is probably the minimum amount borrowed that is efficient for both the market and the borrower. For a new borrower, it is probably a 12-week process, most of which involves getting a rating from the rating agencies. Once a deal goes to market, investors won’t really see the deal until two weeks before commitments. Closing occurs relatively quickly thereafter.

For an existing issuer, the process is probably compressed down to a week or week and a half. In a weaker market, rating agencies will work faster, and there is an ability to get additional dollars relatively quickly for companies that have already had transactions rated. In such cases, a borrowing of $100 million or more would be economic to do.

Project Bonds

MR. MARTIN: Let’s move to project bonds. John Anderson, the project bond market does not do well when the bank and term loan B markets are wide open and looking for product. You heard the bank market is awash in liquidity, and the term loan B market was pretty healthy last year but is expected to drop somewhat this year.

How many deals were there in 2016, and what are you expecting in 2017?

MR. ANDERSON: I think some of the liquidity in the bank market translates to the fixed-income markets as well. We had another year of record issuance in the public bond market in high-grade debt, and the private placement market had a very high level of issuance as well.

Last year was a marginally stronger year for project bonds than 2015. In

2015, there were about half a dozen syndicated deals. In 2016, there were close to a dozen if you look across wind, solar, public-private partnerships and transportation and an LNG deal. So there was better volume in 2016, although still not a ton, but that was just the syndicated flow.

There are roughly 25 participants in the project bond market. Maybe eight to 10 of them are anchor investors. The anchor investors drive the market. They tend to be life insurance companies with larger staffs. They are also doing smaller direct deals, one-investor deals, that do not show up in the broadly syndicated numbers. The direct deals are harder to track.

MR. MARTIN: When you say there were close to a dozen syndicated deals in 2016, that means in the public market and not privately placed deals, correct?

MR. ANDERSON: Correct. I am giving you numbers for syndicated deals. Generally you see project finance deals placed in the private placement market, and not in the public bond market. Sometimes you will see some in the 144A market that, again, is limited to institutional investors.

My default expectation is a flat market in 2017 compared to 2016. The 2017 numbers will depend on supply of investment capital and what needs to get financed. As rates tick up a bit, more treasurers will say, "Base rates are still pretty low on an historical basis. Maybe it would be best to lock in the current rate before it goes higher." The project bond market is a fixed-rate market. You can lock in 20- to 25-year money at current rates, which some people may find more attractive to do in 2017 than they did last year.

MR. MARTIN: There was only one deal in the project bond pipeline last year at this time. How many do you see today?

MR. ANDERSON: We see five. That number pales by comparison with the bank market numbers that our earlier panelists were talking about. But it is a healthy flow.

MR. MARTIN: Project bonds price off treasuries. So you have a fixed rate that is a spread off treasury bonds. What is the current spread?

MR. ANDERSON: The 10-year treasury is 2.4% and the average life on a lot of project deals will be more like 12 years or a tad longer, and we see spreads of 200 to 300 basis points over. So if you look at what that turns into as a coupon, the range is probably 4.5% to 5.25%. Investors in the project bond market generally do not receive an upfront fee. They are compensated through the spread.

MR. MARTIN: The tenor is generally as long as the PPA or perhaps one year short?

MR. ANDERSON: We generally go the length of the PPA. I see that in syndicated deals frequently.

MR. MARTIN: Can project bonds be used for merchant or quasi-merchant gas projects with hedges?

MR. ANDERSON: That gets more into a BB-type credit quality. The project bond market is an investment-grade market.

MR. MARTIN: So BBB?

MR. ANDERSON: Exactly right. If you have fixed capacity payments and the other payments are merchant, lenders in the project bond market will lend solely against the capacity payments and other contracted revenue stream.

MR. MARTIN: How large a transaction does one need to do a project bond?

MR. ANDERSON: If there is only one investor, $30 to $50 million can make sense. If you’re doing a larger syndicated deal, it probably should be at least $100 million.

Summary

MR. MARTIN: Let’s summarize. Let me tell you what I took away from this. The preliminary figures are that renewable energy tax equity was an $11 billion market in 2016. That was a little over $6 billion in wind and a little under $5 billion in solar. The $11 billion total was down from an adjusted $14 to $15 billion in 2015. On this call last year, we had said 2015 tax equity was about $13 billion.

The reason for the smaller tax equity number seems to have been a drop in the number of utility-scale solar projects in 2016 compared to 2015. Both our tax equity investors said they see the market functioning normally despite the tax reform debate starting on Capitol Hill. The one change is they expect to see tax equity investors more interested in taking bonus depreciation in order to use up tax capacity at the higher tax rate in effect this year.

They see tax reform having a limited effect on the market: perhaps a little less tax equity might be raised in deals. Tax equity accounts for 50% to 60% of the capital cost of the typical wind farm and 40% to 50% of the cost of a typical solar project today.

In terms of noteworthy trends, Jack Cargas said he sees a trend toward more complex transactions. John Eber provided some very interesting numbers. Wind developers stockpiled enough turbine equipment in 2016 to justify a build out of anywhere from 30,000 to 58,000 megawatts of new wind farms, according to one estimate, and 40,000 to 70,000 megawatts, according to another estimate. That is in a market where the total installed wind capacity today is around 82,000 megawatts.

Turning to the bank market, what we heard was a significant drop off in dollar volume of transactions in 2016 compared to 2015. It was a 42% drop, according to Ralph Cho, from a $60 billion North American project finance bank market in 2015, down to $35 billion in 2016. There was only a 14% drop in the number of deals, from 160 down to 137. The reason was there were massive LNG projects in 2015 but not in 2016. Neither of our bankers expects to see LNG make a major comeback this year.

There were 50 to 60 active project finance lenders in 2016. Both bankers are expecting new lenders, particularly from Korea and perhaps other places in Asia, to come into the market in 2017. The bank market is expected to remain awash in liquidity.

Spreads are currently 162.5 to 200 basis points above LIBOR for plain-vanilla deals with good sponsors. That translates into a coupon rate of 2.6% to 3%, but then you need to add the cost of a swap. The bank market offers floating-rate loans. For merchant gas projects, the spread is 325 to 350 basis points over LIBOR.

Debt service coverage ratios are currently 1.4x to 1.45x for wind, 1.3x for utility-scale solar, and 1.5x for rooftop residential solar. Contracted gas-fired power plants are 1.4x to 1.5x. Current advance rates on debt are 80% to 85% for contracted projects, but a little below 50% for merchant or quasi-merchant projects.

Turning to the term loan B market, we saw a significant rebound in that market in 2016. The numbers are $11 billion in 2013 falling to $9 billion in 2014, falling to $3.3 billion in 2015, rebounding to $11.6 billion in 2016. About 50% of the 2016 deals were merchant, and not just merchant gas-fired power plants, but also other types of merchant deals.

J-P Boudrias expects to see maybe $5 to $6 billion in B loan volume in 2017. The final volume will depend on how vibrant an M&A market there is. The spreads above LIBOR for term loan B debt fell from 2015 to 2016. We start 2017 with a spread of around 350 basis points for BB credits and 425 over for B credits. There is a 1% floor currently for LIBOR, but that may disappear during the course of the year.

About the project bond market, we heard that it, too, had a significant rebound in 2016. It is an investment-grade market. There were probably a half dozen syndicated deals in 2015, but close to a dozen large deals in 2016. There are a lot of active players, probably 25 as we enter the year, and eight to 10 anchor investors. There are five deals already in the pipeline in the project bond market compared to one at the same time last year. Spreads above 10-year treasuries are about 200 to 300 basis points. Ten-year treasuries are 2.4% at the moment, so that translates into a 4.5% to 5.25% coupon.