Cost of Capital: 2014 Outlook
A group of industry veterans talked in late January about the current cost of capital in the tax equity, bank debt, term loan B and project bond markets and what they foresee in 2014.
The panelists are John Eber, managing director and head of energy investments at JPMorgan Capital Corporation, Lance Markowitz, senior vice president and head of leasing and asset finance for Union Bank, Thomas Emmons, managing director and head of renewable energy finance for the Americas at Dutch bank Rabobank, Raya Prabhu, managing director and head of power and midstream financing at Goldman Sachs, Richard Randall, executive director for North American debt investments for IFM Investors, an Australian-based fund with $48 billion under management, and formerly head of power and project finance for RBS Global Banking, James Finch, managing director and co-head of US loan capital markets for Credit Suisse, and Ray Wood, managing director and head of US power and renewables for Bank of America Merrill Lynch. The moderator is Keith Martin with Chadbourne in Washington.
MR. MARTIN: Tax equity volume in the renewable energy sector hit $6.5 billion in 2013, which is up about $1 billion over 2012. There was roughly an even split between wind and solar. John Eber, can you break it down further: how many wind or solar deals were there?
MR. EBER: We saw 21 wind deals come to the market and receive commitments in 2013. There were 13 different sponsors for 23 projects and about 3,000 megawatts of capacity. In the large-ticket solar market, we saw 27 deals from 18 different sponsors. Of those deals, 10 each were in residential and utility, and about seven in the distributed generation market. In total, it was almost 1,800 megawatts of solar.
MR. MARTIN: What volume are you projecting for 2014?
MR. EBER: I am not in the projection business. Let’s just say we see a sizable pipeline of opportunities in both solar and wind. If you go to the American Wind Energy Association database and look at all the power purchase agreements, you will see the potential to double the amount of megawatts that we saw last year.
MR. MARTIN: Is it your sense that the supply of tax equity is elastic enough to meet whatever the demand will be this year?
MR. EBER: There will be enough tax equity. The market continues to expand in terms of number of investors. The active investors continue to increase the amount of tax equity they are prepared to invest.
MR. MARTIN: Lance Markowitz, yields have remained remarkably stable over the last three years. Where are they currently for wind, utility-scale solar and rooftop solar?
MR. MARKOWITZ: The concept of flip yield is misleading, but you are correct that they have been fairly stable in the wind market. They range between 50 basis points above or below 8% after tax on unlevered transactions.
We have seen a lot of diversity in utility-scale solar. They have been the most aggressively bid transactions, so yields in that market are a little lower than for the benchmark wind deals. Utility-scale solar has an investment tax credit as opposed to wind and production tax credits. People bid those differently. As for rooftop solar, yields tend to be higher than for wind, but no more than 50 basis points higher. We have seen really strong rooftop deals whose yields are well above wind.
MR. MARTIN: So rooftop solar is around 9%?
MR. MARKOWITZ: We have not done a ton of rooftop solar, but we have seen many different pricing files. Pricing depends on what the parties are trying to accomplish. We have seen ranges of a couple hundred basis points due solely to the way the transaction is structured.
MR. MARTIN: John Eber, how many active tax equity investors are there currently? Do you have a breakdown between wind and solar?
MR. EBER: There are roughly 25 active investors between the two markets. Of the 25, 13 invested last year in wind. There were more than 20 investors in the large-ticket end of the solar market.
MR. MARTIN: How far forward will tax equity commit?
MR. EBER: Most equity investors will make a forward commitment of no more than a year. That seems to work for the market. Last year, it was a little different in the wind market with the rule that wind farms had to be under construction by year end to qualify for tax credits. There were more sponsors looking for longer forward commitments than we had seen in a while, so there were a number of us that provided commitments longer than a year, but that is atypical. A year seems to work for the residential solar space. Many of our residential solar clients are looking to raise tax equity within a six- to 12-month time frame.
MR. MARTIN: Is the wind market purely a production tax credit market at this stage due to increasing turbine efficiency?
MR. MARKOWITZ: Yes. I am certain there are a few anomalies, but pretty much everything we see today involves production tax credits.
MR. MARTIN: Pay-go structures appear to have made a comeback. In a pay-go structure, the tax equity investor puts in its tax equity over time as the tax credits are earned. Why the revival? Is tax equity provided through a pay-go structure more expensive than where the entire tax equity investment is made at inception?
MR. EBER: Pay-go structures have always been popular. We have been using the structure since we started investing. They work well for deals where the financing is already in place, but the sponsor wants to monetize the remaining production tax credits. They work well for deals that are more risky than average; the pay-go feature can help reduce the risk to the tax equity investor, allowing the investor to commit tax equity to a deal he might not otherwise do. The IRS partnership flip guidelines allow up to 25% of the tax equity investment to be paid in over time as a percentage of output or production tax credits.
The tax equity costs the same. The pay-go feature brings the risk of the investment more in line with a traditional deal. Once you get it risk adjusted using the pay-go feature, then you don’t need to seek additional yield.
MR. MARTIN: As solar projects get larger, they are more likely to need debt as well as tax equity. Yet, tax equity investors doing partnership flip transactions have not been keen on having lenders at the partnership level. Is this changing?
MR. MARKOWITZ: The preference to avoid partnership-level debt is not changing, but, that said, some leveraged flip deals are getting done. Looking at our own portfolio, we were involved in six deals in the last 18 months that required more than $400 million in tax equity, and none of them had leverage.
MR. MARTIN: Is JPMorgan more willing today to do leveraged flip deals?
MR. EBER: We have done some in the past. Fewer than 10% of the deals we have done over the last decade had any debt at the project or partnership level.
MR. MARTIN: Is there anything special about investment credit deals, which is what the solar market is, that makes it harder to do a partnership flip transaction or to combine tax equity with debt?
MR. EBER: No. They are just a very different type of deal, so they will appeal to different investors. The majority of the tax benefit comes at inception rather than being spread over 10 years as it is in a deal with production tax credits. ITC deals have a different income pattern and a much faster payback. They require a lot more tax capacity immediately for a comparable amount of equipment value from the investor, as compared to being able to spread the tax capacity over 10 years.
We like both types of investments, but there are some investors who are more comfortable with one or the other because of their particular tax positions.
MR. MARTIN: One of the difficult issues when you combine debt with tax equity is that tax equity investors want the lenders to agree to forebear from taking the assets after a debt default until the tax equity can reach its yield. The lenders can step in and replace the sponsor in the meantime. There used to be a “market” approach to forbearance, but that seems to have collapsed lately. There are deals that have not gone forward because of forbearance issues. Is it your sense that whatever market consensus there was has now disappeared?
MR. MARKOWITZ: Yes, although I don’t know whether there was ever really a consensus. Over the years, the transactions that took the longest to close were the ones that bogged down over debt and equity issues.
MR. EBER: The consensus was that there were one or two banks that understood the issue and were willing to agree to forbearance. There was never a broad market consensus regarding forbearance, which is why the tax equity market remains dominated by deals that do not have debt at the project or partnership level.
MR. MARTIN: How much is the current yield premium when there is project- or partnership-level debt?
MR. MARKOWITZ: The yield will move up to the low teens to mid-teens, depending on the transaction.
MR. MARTIN: The federal bank regulators came out in late December with a definition of “covered funds” under the Volcker rule. National banks cannot invest in covered funds. Have you been advised by your bank regulatory counsel that the Volcker rule, as the federal bank regulators have now implemented it, will affect your ability to continue making tax equity investments?
MR. MARKOWITZ: I have not. We continue to make such investments and expect to be able to continue doing so.
MR. MARTIN: Wind, landfill gas, biomass, and geothermal projects had to be under construction by December 2013 to qualify for federal tax credits. There are two ways to start construction. One was for the sponsor to “incur” at least 5% of the project cost by the end of 2013. The other was for the sponsor to have started physical work of a significant nature on the project. It does not appear that much physical work was required in 2013. Are you willing to rely at this point on the physical work test?
MR. EBER: We expect to be able to do that. That said, we have not seen many examples of it yet, so we are still feeling our way about where to draw lines. Hopefully when clients bring deals to us, the physical work will be well documented and will be significant enough to fit within the parameters we think the Treasury and the IRS will use to draw lines.
MR. MARTIN: What do you think is the minimum physical work required?
MR. EBER: That’s a hard question to answer. It will come down to facts and circumstances. We will make decisions based on what IRS guidance has been issued to date.
MR. MARTIN: In late December, the IRS released new guidelines on tax equity transactions involving tax credits for rehabilitating buildings. Has this so-called Historic Boardwalk guidance had an effect on how you are structuring deals in the renewable energy sector?
MR. MARKOWITZ: No, but I understand that there are a few general principles behind that guidance that people will at least pause to think about when doing future deals.
MR. MARTIN: Tom Emmons, was the big story in 2013 that the banks are back as project finance lenders? The North American project finance bank market was $40 billion in 2011 and roughly only $24 to $25 billion in 2012. Do you have a figure yet for 2013?
MR. EMMONS: That number is hard to pin down, because there are several databases, they don’t have standard criteria and some tallies have not been published yet. I think the consensus is that 2013 was up over 2012. Some of the databases suggest it was up around 20%.
What is more interesting is to look at the sub-sectors within project finance. Oil and gas and conventional power seem to be up. Renewable energy seems to be flat or down.
MR. MARTIN: How many active banks were there in 2013? How many do you expect in 2014?
MR. EMMONS: There were around 40 or 50 in 2012. I expect the final tallies to show roughly 10 more in 2013. There should be even a few more in 2014. We are seeing some US regional banks, smaller Canadian banks and even some northern European and Nordic banks coming in.
MR. MARTIN: Rich Randall, one would think a large number of returning banks would mean downward pressure on margins. Was there? What is the current spread above LIBOR for interest rates? Where do you see it headed in 2014?
MR. RANDALL: For bank deals, the average is probably around 200 basis points over LIBOR. I think there is a lot of downward pressure. We are starting to see some pricing go below 200 on some new deals. With the additional liquidity coming into the market, the downward pressure will continue.
MR. EMMONS: There is a large range in pricing. Pricing has softened over the last year, but I think most of that softening is with large straightforward deals with strong sponsors. The pricing on smaller complex deals has not moved as much.
MR. FINCH: The commercial bank market is the one market where a relationship matters, so unlike all the capital markets, if there is a strong relationship between the sponsor and bank, then the loan will be priced at a discount, regardless of the cost.
MR. WOOD: What we are talking about is deals within a narrow band of risk. There is an implied strong to mid-BB rating, if not higher. While the high-yield market, the institutional term loan market and the commercial bank market are much more liquid than they have been in years, they are still interested only in the low-risk deals.
MR. MARTIN: Current yields are 200 basis points over LIBOR, with some downward pressure. Is there a LIBOR floor tied to the cost of funding and, if so, what is it?
MR. RANDALL: Not in the bank deals. The bank market does not require a floor. However, we are seeing LIBOR floors in the institutional loan market of around 1%.
MR. FINCH: The reason for the LIBOR floor was that when rates were falling, institutional investors were trying to preserve some yield, and so they set a minimum rate to which the spread was added. That is a bit of a legacy that will disappear rapidly in a market where interest rates overall are expected to rise.
MR. MARTIN: What does 200 basis points over LIBOR translate into as a coupon rate?
MR. EMMONS: The six-month swapped LIBOR is around 3.25%, so you add 2% to that. There are often step ups over time for longer deals, but the rate is well under 6%.
MR. MARTIN: How much would you expect the rate to step up ultimately for a 10-year deal?
MR. EMMONS: It goes up typically by an eighth or a quarter percent every three or four years.
MR. MARTIN: What are current upfront fees?
MR. EMMONS: They vary with tenor and other factors, but they are often the same as the starting margin, so in the low 2% range.
MR. MARTIN: We have read a lot recently about manipulation of LIBOR by banks and potential criminal prosecutions. Is the market moving away from LIBOR as a benchmark or is it just adjusting how LIBOR is calculated?
MR. FINCH: LIBOR remains the benchmark.
MR. MARTIN: Bank loans seemed to shorten in 2012 to seven to 10 years with mini-perm features. Where are they today?
MR. RANDALL: Seven to 10 years is still the norm. Institutions like ours have the ability to go longer, and that is where we are trying to fit into the market. We see a subset of banks, particularly the Japanese, that are willing to go as long as 15 to 18 years.
MR. WOOD: I think the commercial banks have wanted to keep it shorter for return-on-capital reasons. There has been a big institutional bid for the longer-dated piece. We have seen banks come in jointly with pension funds or other institutional investors so that the sponsor can get the duration it wants by leaning on banks for the shorter piece and institutional money for the longer piece.
MR. MARTIN: Tom Emmons, last year on this call you said, “The shortening of tenors is creating opportunities for institutional lenders and they have been stepping up. I think it is a permanent shift.” Do you still stand by that view?
MR. EMMONS: Yes. As mentioned, I think banks still want to keep their legal maturities under 10 years so borrowers are given the choice of doing a mini-perm with a commercial bank or going long-term fixed in the institutional market. Many borrowers are electing to go long-term fixed. The numbers in the institutional debt market were up last year as well.
MR. MARTIN: What are debt service coverage ratios currently for contracted wind and solar projects?
MR. EMMONS: Wind may be mid-1.40x, and solar is mid-1.30x.
MR. MARTIN: What about new gas-fired power plants?
MR. WOOD: There are not too many of those that come with the same long-term offtake contracts, so it is difficult to say. You tend to have amortization over the contract period, and you are really solving for the merchant loan-to-value. That is how the rating agencies and institutional investors evaluate how much debt gas-fired projects can support.
MR. FINCH: Ultimately, you can get coverage ratios for gas-fired power plants down to 1.0x through the offtake contract period, if the market believes that the project is truly contracted with a creditworthy offtaker. However, the devil is in the details at the maturity of the loan. Does the merchant component of the power plant provide sufficient coverage to merit the investment? That coverage will be higher than 1.0x.
MR. WOOD: A lot of gas deals will have a percentage cash sweep of all available cash flow, anywhere from 50% to 100%. There is a coverage ratio for the mandatory amortization, which tends to be pretty light, and then there is a cash sweep.
MR. MARTIN: Every plant has a merchant tail after a power contract runs out. Does the debt need to be shorter than the power contract?
MR. FINCH: No. Merchant is defined regionally. Merchant in ERCOT is different than merchant in PJM.
MR. MARTIN: What would a coverage ratio be for a merchant plant with a power hedge in ERCOT?
MR. FINCH: It depends on how long the tail is when the loan matures, but the power hedge usually lasts longer than the debt is expected to remain outstanding.
MR. PRABHU: One other factor to keep in mind as you get to the maturity of a term loan B is the loan-to-value. One of the other metrics investors have been using is the out-year value that would be assigned by the M&A market and trying to understand what kind of loan-to-value you have in the base case and downside scenarios.
MR. WOOD: Lenders are assuming a value well below the total capital cost of a new build. This is yet another reason why we are not seeing a lot of new construction. We have seen some in ERCOT and in other places where people have long-term contracts. There is a firm bid for merchant generation, but at a sizable discount to new entry capital costs.
MR. MARTIN: What percentage of project costs can be financed under a construction loan in the bank market?
MR. EMMONS: It depends on the bridgeable capital inflows coming in at the end of construction, and it also depends on each lender’s policy for debt ratios, but it could be as high as 80% to 90%.
MR. MARTIN: Are banks back to full underwriting or are the larger transactions being done as club deals?
MR. EMMONS: In renewable energy, they are mostly club deals. The deals are pretty straightforward, and the borrowers do not require underwriting.
MR. RANDALL: On other transactions, we tend to use institutional markets interchangeably with commercial bank markets. Although it is the same product, there is a different risk appetite among lenders in the two markets.
For the larger deals that need underwriting, to the extent that there is sufficient relationship pull through the sponsor, banks are more than happy to provide significant underwritings to those transactions.
MR. WOOD: The liquidity in these markets lets the relationship banks make an underwriting commitment and have a high degree of comfort that there will be a decent takeout, even if the primary form of takeout falls away. There are so many other secondary forms of takeout with more institutions willing to step in. We have seen some one-off transactions where one bank acted as a bridge lender where time was of the essence and earned an exceptional return for the takeout risk, but it is not the norm.
MR. MARTIN: We talked a little about merchant projects. They were another big story in 2013. Gas-fired power plants and some wind deals were financed on a merchant basis in PJM and ERCOT. Were all these deals done in the term loan B market? Are banks getting more comfortable with merchant deals?
MR. RANDALL: It depends on the market. PJM is where most of the activity occurs. It is the most mature and transparent market, and the easiest in which to get a deal done. The supply-demand economics work well.
MR. MARTIN: Will banks get comfortable with merchant deals?
MR. EMMONS: I prefer to call deals either contracted or un-contracted. Contracted can mean a power purchase agreement or a hedge with a strong counterparty. Commercial banks typically lend against contracted cash flow, whether under a PPA or a hedge with a strong counterparty. There is no magic minimum number of years for a hedge, but shorter hedges support less debt, and the balance has to come from equity or junior debt.
MR. WOOD: All but possibly one “merchant” deal over the last 12 to 18 months has involved a power hedge. A counterparty agrees to a fixed-price offtake for 10 to 12 years off of a P90 wind resource scenario. It may even be a lighter production estimate than in the peak summer months, given the volatility in ERCOT.
Most such merchant wind deals have been in ERCOT. The load-serving entities have little interest in signing long-term contracts. There are anywhere from 2,000 to 4,000 megawatts under construction or being planned. Most of the projects have power hedges, the banks are coming in for construction debt, and tax equity has been available. A handful of players are providing the hedges. It will be interesting to see how the current discussions in Washington among the federal bank regulatory agencies about the extent to which banks should be allowed to trade commodities will affect Wall Street’s ability to continue providing those hedges.
The same type of coverage ratios apply to deals with power hedges. The banks plan to be taken out with the tax equity or back leverage at the end of construction.
MR. MARTIN: Raya Prabhu, Goldman Sachs led many of the most prominent recent financings of merchant gas-fired power plants in the term loan B market. Do you see merchant gas as an expanding market?
MR. PRABHU: The bulk of the activity will remain in PJM and ERCOT. That is largely driven by the fact that these are mature markets with very strong underlying power fundamentals. Other drivers have been the low cost of natural gas and the expected coal retirements over the next few years.
We led most of the projects in the term loan B market this past year. We found a great reception to them from a wide range of investors. A lot of that was driven by tightening yields and spreads for operating assets. People who are looking for total return are moving to riskier asset classes, like project financings.
Term Loan B
MR. MARTIN: Term loan B debt is papered like bank debt, but it is sold to institutional investors looking for yield. It tends to be used to finance projects riskier than one might be able to finance in the regular bank market. Any idea what the term loan B volume was last year?
MR. FINCH: Rather than being papered like bank debt, I would say it is bank debt. You are simply selling it to different investor groups. Whether the buyer is a commercial bank or an institutional investor, it is still a bank loan. It is particularly attractive in a rising interest rate environment.
Last year, there were about $455 billion of B loan issuances, and that was an all-time record. The previous record was in 2007 at $387 billion. Money continues to flow into the term loan B market to the tune of about $750 million to $1 billion a week. As you see a lack of new M&A-driven issuances, investors are looking for new places to invest money. Project financing is becoming more and more attractive to them as they become more familiar with the construct and the risks they are being asked to take.
MR. MARTIN: Last year, this panel estimated that the combined term loan B and project bond market for North American project finance in the power sector was about $4 to $5 billion. Is there a comparable breakdown for 2013?
MR. PRABHU: Focusing strictly on the term loan B market for greenfield projects, I would venture to say that in 2013, between the various ERCOT and PJM financings and other deals, the figure was probably in the $2 to $3 billion range. We have not seen a lot of greenfield project financings in this sub-sector of the market. Most deals have been quasi-merchant.
MR. WOOD: One thing that hurts the project bond market is that banks are so comfortable with solar and wind projects that benefit from the 12- to 20-year PPAs that the loan-to-value, spread above LIBOR and the flexibility of being able to call at par is of greater value to sponsors than going to a non-call, long-duration project bond that has to be rated by both rating agencies and that locks them into a fixed yield.
Sponsors are m