Merchant gas projects: how many more?
Merchant gas-fired power projects in ERCOT, PJM and ISO-New England helped carry the North American project finance market in 2015. Many banks are feeling flush with merchant gas risk in PJM. Panda Energy Partners is suing ERCOT over losses on three merchant gas projects in Texas in which the company invested $2.2 billion. How much more appetite is there for such projects? Two developers, two bankers and one equity investor discussed these and related questions at the Chadbourne annual global energy and finance conference in early June. The following is an edited transcript.
The panelists are Ravina Advani, managing director at BNP Paribas, Jay Frisbie, managing director at Tenaska Capital Management, Herb Magid, managing partner and co-head of Ares EIF, Michael Pantelogianis, co-head of power & infrastructure finance for North America at Investec USA Holdings Corp., and Scott Taylor, chief financial officer of Moxie Energy. The moderator is Rohit Chaudhry with our Washington office.
MR. CHAUDHRY: The PJM capacity auction results came out on May 24, and they left a number of people very unhappy. Ravina Advani, what were the results and why were people unhappy with them?
MS. ADVANI: The results were abysmal. Looking at the RTO alone, capacity payments went from $164.77 to $100. That was a huge shock to pretty much everybody in the industry. There were a lot of consultants who were predicting 20% to 40% higher in their expectations of where the RTO would clear. So, absolutely abysmal, particularly when from a debt-sizing perspective, those capacity payments represent anywhere from 35% to 50% of the underlying gross margin. It was very unfortunate.
MR. CHAUDHRY: Considering those results, I want to get a sense from each of the panelists whether you think there will be any new plants built in PJM. However, before I do that, I want to get some basic facts and statistics out on the PJM market and how it looks after this auction. Scott Taylor, how many megawatts of capacity cleared in this PJM auction, and how did it compare to prior years?
MR. TAYLOR: This year, 167,000 megawatts cleared, which is maybe 500 megawatts more than cleared last year. What helped to drive down the price is PJM reduced its peak demand forecast by something like 4,500 megawatts just prior to the auction in February when PJM releases its parameters and, on top of that, 185,000 megawatts were offered versus 180,000 last year.
MR. CHAUDHRY: So there was an increase in supply. I think there have been around 5,000 megawatts of new builds.
MR. TAYLOR: Yes. More than 5,000 megawatts of new gas plants bid.
MR. CHAUDHRY: Coupled with a decrease in demand on the PJM market. So in light of these two factors, I want to go around the panel and ask whether you fear an overbuild scenario in PJM.
MR. TAYLOR: It is hard say. The $100-per-megawatt day was terrible. I am not aware of any consultant who forecast that.
There are a couple of offsetting factors that suggest we are not yet in an overbuild situation. I know that was a topic when we were trying to market the Freedom project, and I am sure it will remain a topic for anyone taking a project to the bank market today.
One of the tricky things to decipher is PJM has two types of products. One is for capacity performance and involves more risk and a higher price, and one is a base product. This year, 26,000 megawatts of capacity cleared as a base product that will not be able to clear as base in the next auction, so it will have to convert to capacity performance.
A driver of the build out of gas plants has been the coal retirement projections. Coal retirements will increase in number. Nuclear plants are also being retired, which I do not think was expected several years ago. PJM has been a coal and nuclear market. The two retirements together mean a lot of capacity will fall out of the market and will need to be replaced by something. That something is obviously gas. Put everything together and it is hard to say the market has been overbuilt.
MR. CHAUDHRY: So let me pin you down a little more, Scott. I agree that it is hard to say. Moxie has done three projects, but your company has been quiet for the last few months, maybe even for the last year. Are you looking to come back into the market?
MR. TAYLOR: I have been catching up on my sleep, trying to kick back and relax. There are only five people in our company. It has been a pretty hectic few years. Having said that, we are trying to catch our breath and see how our three deals perform and then figure out what makes sense to do next. There are other solid gas plants in front of us, so I don’t think it would be a smart move just to try to catch up with those deals that are getting ready to come to the market for financing.
MR. CHAUDHRY: Let me ask the others. Herb Magid, are we in an overbuild situation?
MR. MAGID: I do not think we are in an overbuild, but you went from it being an easy analysis to some need for new development to replace coal and nuclear plants being shut down. Some projects that might not have inherent benefits, like gas location or maybe a favorable underlying contract, are probably being pushed off. The ones that will be done are the better projects.
PJM is a market where you have to channel your inner Bernie Sanders. It is kind of a rigged system for the utilities and the incumbents. New entrants have a hard time. I would imagine the capacity payments will move up and down over time, so good projects will continue to be done. It is unlikely to get into an overbuild and as banks get taken out by refinancing in the public debt markets, they will look for other good projects. I think it is a sustainable market.
MR. CHAUDHRY: Jay Frisbie?
MR. FRISBIE: A couple thoughts. First, auction results are notoriously hard to predict. The auction is a black box. When we try to do an analysis around where prices might land, we have a pretty wide range. We had the bottom of our range where it actually came out. It was certainly possible to do.
Second, it is important to keep in perspective that this was one auction and one auction result. These are long-lived assets. They are going to continue to go through these auction processes, and the results move up and down. You have to be able to stomach that as an investor. It also means that you have to have a keen focus on your operational capabilities because there are going to be leaner years when the capacity prices are not as strong and you will be relying more heavily on your merchant generation for revenue.
As for whether there is an overbuild, it depends. There has been a lot of new build out in Pennsylvania and Ohio where the prices were strong for a long period of time. In the last two auctions, Commonwealth Edison has broken out quite significantly suggesting there may be an opportunity for some new build there.
PJM is a pretty broad ISO. You have to analyze where we are in an overbuild situation by looking at specific locations within the PJM grid.
MR. CHAUDHRY: You mentioned ComEd and Ravina mentioned the RTO price at $100 a megawatt, but ComEd was significantly higher than that. What was the ComEd price?
MR. FRISBIE: It was over $200, but I don’t remember the exact price.
MR. CHAUDHRY: It was $203.
MR. FRISBIE: It was even higher than that at the previous auction. If you have generation in that region, you were not disappointed with the results.
MR. CHAUDHRY: Tenaska recently closed on the financing for the Westmoreland project in PJM. Are you pursuing new opportunities in PJM or, like Scott, are you just recuperating?
MR. FRISBIE: We are always looking at opportunities, and not just within PJM. It has been in the press that we have some projects in ERCOT. We are looking at ISO-New England, but it is an asset-specific case-by-case situation.
The Westmoreland project was the first time we have built a project like this. It was quasi-merchant. Historically, Tenaska has always been about long-term contracted generation, whether it was gas-fired 10 or 15 years ago, or some of the utility-scale solar that we have purchased or built in the last five years. It is a step outside our comfort zone, but with the long track record and long history, it is a calculated risk and it seems the right thing to do. We will continue to look for opportunities like that. ERCOT is challenging, and PJM is going to be challenging.
MR. CHAUDHRY: Mike Pantelogianis, how does the bank view the results of the capacity price auction? How did you react to it?
MR. PANTELOGIANIS: A higher capacity environment is always viewed more positively. We expected a downturn. We looked at the results from last year. There were a lot of omissions in terms of new build. When we compared what got bid versus where the financing calendar was, we did not expect as healthy pricing for the general region as last year.
I agree with Jay Frisbie. The analysis has to be more focused on specific locations.
When you look at PJM, it is hard to generalize on an overbuild. We looked very closely at expected retirements. We are trying to get comfortable that capacity additions are definitely at or below the level of retirements. We have made seven or eight investments in PJM projects in the last three years. We try to do that analysis and make sure that supply and demand and retirements and additions reflect a healthy balance.
MR. CHAUDHRY: Ravina Advani, you started with the comment that the prices are abysmal. This panel seems to have a very balanced, relaxed view about what will happen as a result of that. What is your take?
MS. ADVANI: On an overbuild situation, I think it does really depend. As a lender, we have really aggressive budgets to meet. I am hopeful that we are not in an overbuild. We see a handful of transactions still coming to the market, despite the recent auction results. In addition to some that my fellow panelists are pursuing, there is the Quantum transaction, an Invenergy transaction and a CPV transaction, so there seems to be a pretty robust pipeline for the rest of the year. I am hopeful.
MR. CHAUDHRY: Scott Taylor, the other part of the story besides capacity payments is spark spreads. How have they reacted to the movements in capacity prices.
MR. TAYLOR: Spark spreads are still strong. It is that fact as well as the expected retirement of coal and nuclear plants that is driving the continued development of new plants.
I agree with the comment that the capacity auction price is a one-year event. Last year, it was $164. That is a big drop in one year, but these are 30-year assets, and you are not developing these projects based on a one-year result. A bigger driver is spark spreads. Spark spreads this year are down compared to last year for a bunch of reasons not worth getting into, but the spark spread forecast is still solid and, as long as it remains solid, you will see continued development in PJM.
MR. CHAUDHRY: I want to ask both lenders what impact all of this is having on your existing financings. What did you project for instance in capacity prices in your base case models in the deals that have closed? What is the impact on financings that are still ahead this year?
MS. ADVANI: It really depends asset to asset. We have probably a dozen assets that are located in PJM. It really depends on the capacity forecast that we used in sizing the original debt. In a lot of cases, we assumed an overbuild situation and, in others, we assumed a base case forecast. By and large, most of the projections, particularly for the 2019 to 2020 delivery period, were in excess of where we currently cleared. We have one asset where we assumed $180 a megawatt in the 2019 to 2020 delivery years, so that is one project where there could be a default on interest or principal.
On the flip side, these transactions have been structured pretty well with a dynamic target debt balance to deal with this very instance, so the result may be we end up with a higher dollar-per-KW metric at maturity. It was expected in most of these deals to be in the $350 to $375 range. It may end up higher in most of the deals if the capacity prices do not turn around.
MR. CHAUDHRY: On the first panel this morning, the bold statement was made that some banks will end up taking losses. Do you agree?
MR. PANTELOGIANIS: Wow. I certainly have a lot of respect for my colleagues on that panel and their views.
MR. CHAUDHRY: You don’t need to. [Laughter]
MR. PANTELOGIANIS: But — the but was coming — I have the benefit of having lived through the first round of merchant expansion, which was just before the merchant meltdown and the Enron bankruptcy. We did deals differently then. The hedging markets were not as developed as they are today. Our financings today include a lot of stability in terms of spark spreads and put and call options that are entered into by our clients to hedge price risk. The transactions that are done today are much better in terms of credit quality from those that were done 10 and 15 years ago.
We look at it bottoms up. We do not want to rely too heavily on capacity payments. We look at our portfolio. We start with the spark spreads and try to understand what the heat-rate call option or the put provides for in the context of a stable spark spread.
What we have noticed in the last three years is that the hedge counterparties are charging more for certainty and delivering lower spark spreads.
Projects that have a high cost of capital and that might have gotten done in a stronger market have a bigger nut to crack as capacity payments fall. Our clients have a lot of equity in these deals. They are not making these investments without getting comfortable with the downside. Like Ravina said, we are in a volatile merchant market. Our clients are prepared for that. We are a lot more sophisticated in our understanding of merchant risks today than we were 10 or 15 years ago.
MR. CHAUDHRY: Do you expect capacity prices to rebound, stay the same or continue moving down, and why?
MR. PANTELOGIANIS: They will be different.
MR. CHAUDHRY: I need something more than that. Scott Taylor, you go first.
MR. TAYLOR: I have been wrong every year, so . . . . [Laughter]. If I was going to guess an over-under for next year, I would put them at $140.
MR. CHAUDHRY: What is the driver for that? Coal and nuclear plant retirements? Something else?
MR. TAYLOR: One of the unknowns is that you have 26,000 megawatts of base capacity that can’t be “base capacity”; it has to take on the capacity payment risk. It has to convert, so the question is how much of that will actually take on that risk or just go away. You also have the retirement story, and then there is the question how many of these new projects will actually get financed.
To go off on a tangent for a second and talk about the lenders, I do not think there is any big risk under which the lenders will risk taking a haircut, at least on the deals that I have seen. The lenders have done a very good job of structuring to cover themselves on the downside.
MR. CHAUDHRY: It looks like you are trying to finance a new project, Scott. [Laughter]
MR. TAYLOR: No, I am not. [Laughter]
MR. CHAUDHRY: Let’s hear from Herb Magid and Jay Frisbie on where they think the capacity price will be after next year’s auction and why?
MR. MAGID: Scott is probably in the right range. There are four or five things that move around. We shop consultants because you try to see what all the experts are saying. The consultants fall into two camps. One camp says the capacity price increase over time, and the other camp believes it will remain flat at $100 and it is the spark spread and the total cost of energy that matters. Both groups are persuasive. A big wild card is whether the states or cities where the nuclear plants are located will make a special effort to keep them open.
I second Scott’s opinion that there is no way the banks will lose money given how the deals have been structured. It is just the equity at this point that is exposed to losses.
MR. FRISBIE: There is logic as to why we should expect to see the prices increase in the next auction, but I go back to my comment about the process being a black box.
As Herb said, there are many other factors that affect the auction price. The bidding behaviors by each individual bidder can be so diverse and have a profound impact on what happens with the auction results. They may change from auction to auction. We have pretty detailed and long discussions when each auction approaches as to what price we will bid.
The fact that the auction is moving to 100% capacity performance will put a lot of pressure on bidders, particularly demand response, and that should take supply out of the market. But who knows? These bidders could come up with creative ways to get comfortable and feel good about bidding into such a capacity performance market. It is always hard to predict.
MR. CHAUDHRY: Let’s look at this from a different angle. There are three key stakeholders in all the projects. There is the equity. There are lenders. There are the commodity hedge providers. All have exposure to these projects. Among these three, who is truly going to regulate the market to make sure there is no overbuild? Just one word answers. Of the three, who do you think would be the true constraint that controls this market? Banks? Equity? Hedge providers?
MR. PANTELOGIANIS: I think it is probably a combination of banks and hedge providers. Developers are always going to develop, but if the capital is not there to get the projects done, or the hedges are not there, then it will be difficult to move forward.
MR. MAGID: I would say it’s hedge, then equity, then debt.
MR. FRISBIE: None of them.
MR. CHAUDHRY: None of them? So it is going to be a chaotic situation?
MR. FRISBIE: Everyone thinks his project is the best and has a rationale. If you are looking for a safety net, it is not there.
MR. CHAUDHRY: Ravina?
MS. ADVANI: I’m going to go to the other end of the spectrum and say, “All of the above.”
MR. TAYLOR: I would have expected hedge providers because they have to take a true view on spark spreads. Hopefully it is actually the developers, but Jay is probably right. Hedge providers, but it should be the developers.
MR. CHAUDHRY: Let’s focus on the equity part of this equation. What return to equity can investors expect in these PJM projects?
MR. TAYLOR: I think it depends on the type of investor. You have some investors who are relying on after-tax consequences. Bonus depreciation plays into it. You have some investors that are pension funds who have no ability to use tax benefits.
Generalizing, it is fair to say that people are looking for returns from the low to high teens when developing a project. You build hoping for a return at the upper end of the range, but are prepared to accept something at the lower end. You manage your downside risks as best you can.
MR. CHAUDHRY: What is interesting to me is that these returns are dramatically higher than what people have said the leveraged returns are on renewable energy deals. The returns mentioned there are in the high single digits. When you compare the two, merchant gas still sounds like a good bet.
MR. FRISBIE: That’s true. It reflects the fact that these assets lack the long-term contracts that renewables have, and they do not have the aura of social good, so there are many investors who want to invest in contracted renewables and those investors see themselves holding those contracted renewables for a long time so, even if it earns a single digit return, the multiples could be three or four times over a long period of time. It is harder to have a long-term hold in the merchant gas market. Returns are in the low teens to low 20s.
MR. CHAUDHRY: Scott Taylor, are valuations for these projects on the way down or holding tight, notwithstanding the capacity auction results?
MR. TAYLOR: There is one project that is in the process of being sold. From what I understand, it is getting attractive bids. That reflects the strong spark spreads in PJM and the fact that newer assets have a competitive advantage when they are located in the right area. Maybe some more projects will be sold this year so that we can give you a better answer next year.
MR. CHAUDHRY: Let’s move to the bank side of things. When this market started, it basically started with term loan B lenders providing the debt. Then at some stage, banks became flush with cash and started lending to these projects and the term loan B market was not doing so well. The last few deals were done when the bank market has been somewhat constrained. Projects have just about made it over the finish line, lining up the entire book. How constrained is the bank market now for PJM debt? Do you think a bank market can fill the entire debt piece?
MR. PANTELOGIANIS: I still think good deals get done. We just came off closing a transaction from Macquarie called Lordstown and that question was asked.
MR. CHAUDHRY: There was no issue with Lordstown. We were sponsor’s counsel in Lordstown. [Laughter]
MR. PANTELOGIANIS: Exactly. And we were able to oversubscribe the transaction and get to a good sell-down position. I anticipate the financing calendar will offer us a lot of opportunities to pick and choose what we want to play in.
Is there a little deal fatigue? I think so. I think it’s probably related to some of the institutions that approach the sector in a careful way. A lot of these institutions remember Enron and I think that still sits in the back of people’s minds, and so they are just trying to be portfolio managers and actively look at their exposures and say, “Is this enough? What could throw it off?” The question is out there. There is noise in the market, but it has not kept deals from getting done.
MR. CHAUDHRY: Jay Frisbie, you guys were also able to get there this year on another deal.
MR. FRISBIE: Yes, we were able to do Westmoreland, and it was well received in the market. But we did hear noise that there are some lenders who are up to their exposure limits in PJM. It was not surprising. That is where all the activity has been. It is a risk thing for them.
MR. CHAUDHRY: Ravina Advani, how many active banks are lending to the PJM market? How does that change after the capacity auction? I know some French banks have retreated somewhat.
MS. ADVANI: There is still a healthy number of banks that are active in PJM and just generally active in the quasi-merchant space. The number is between 10 and 15.
We are seeing banks be more selective in terms of the opportunities they pursue from a sponsor perspective, from a remuneration perspective and based on the underlying credit profile of the asset.
There have been a couple refinancings of projects. The Newark transaction is one, and that obviously helped banks recycle some of their capital. LS Power refinanced its Seneca pumped storage project in Pennsylvania. This has freed up some bank exposures.
The capacity in the bank market depends on the underlying size of the transaction. Once you start pushing $700 or $800 million, assuming the capital costs are higher than that, you really need to consider back-filling the balance and alternate markets. The term loan B market is much more robust. We have seen a number of re-pricings come to market. We have seen a number of dividend recaps come to market. The term loan B market remains a viable outlet for some of these financing.
MR. CHAUDHRY: So you expect there to be more hybrid deals, with a bank loan tranche and an institutional debt tranche with someone like Prudential providing the institutional debt tranche?
MS. ADVANI: Yes. I think we will start to see more Opco-Holdco structures and a lot more hybrid transactions.
MR. CHAUDHRY: What kind of complexity does that add to the deal?
MS. ADVANI: There are obvious inter-creditor issues in such structures, but nothing that is insurmountable.
MR. CHAUDHRY: How do you see some of the metrics for how the debt is structured and the bank market changing in light of lower auction prices and concern about spark spreads? Where do you see leverage going?
MS. ADVANI: I see leverage potentially coming down on these transactions. When lenders look at these transactions, they are looking at a capacity forecast. Most of these deals have been levered between 50% and 60%. If anything changes, it will be leverage.
In terms of pricing and structure, I don’t see much movement, at least in the near term. And in terms of remuneration, it depends on the sponsor and the transaction. We have seen arranging fees range from $360,000 to $1 million. Up-front fees have remained pretty stable north of 2%.
MR. CHAUDHRY: Mike Pantelogianis, anything to add to that?
MR. PANTELOGIANIS: I agree. People are going to be creative for larger deals in the current market.
I think there could be an incremental uptick to the required equity because the hedges are costing a lot more money either up front or on heat-rate call option premiums. In order to provide for that incremental cost, there could be a slight uptick. Having said that, equity is typically between 40% and 50% of the capital structure, but I still think that equity probably goes as high as 50%, but not more than that.
MR. CHAUDHRY: We do not have a hedge provider on the panel, so maybe I will ask some of the developers. How constrained is the hedge market? How many hedge providers are there? How easy is it to get a hedge? And are banks getting overexposed to certain hedge providers?
MR. FRISBIE: When we went through Westmoreland, we had a fairly competitive process. I would not say there was a large number, but certainly enough to make the process competitive. It is potentially a big constraint. The number of hedge providers ebbs and flows. There will not be as much capacity in PJM as the developers would like. Other markets could also eventually prove difficult.
MR. TAYLOR: Don’t take any contingency fee deals based on closing. The reason I say that is good sponsors will get their deals done, but one of the tough parts about this business is the lenders create structures that work for them and they might involve a certain hedge price and that also drives into the equity structure. One of the challenges is that you do not know what the final deal is with the hedge until you get to the day of closing.
For some reason, the numbers you get from hedge providers never increase on the day of closing. I don’t know why that is, but it just seems to work that way. If you have a tight deal where you are counting on X, you had better have a lot of confidence in your hedge provider that it will be able to deliver X.
The hedge market still seems to be strong. Prices may have increased, but there may come a day where the hedge number that makes everyone happy cannot be delivered at closing and there will not be enough room between the equity and debt to make up the difference.
MR. CHAUDHRY: Last question, as we are running out of time. Other than PJM, where do the other opportunities lie?
MR. FRISBIE: In places like ISO-New England and New York Zone J. That is a little different type of market, but those essentially cleared markets are the most attractive to us at this point in time.
MR. CHAUDHRY: Herb Magid, your take?
MR. MAGID: I agree with that, but I think there is an interesting opportunity for at least equity investors on smaller deals and it was mentioned on the prior panel. A lot of corporate customers and manufacturers are returning to the US. We are seeing large steam users who are looking to invest in their facilities. They might have old oil-fired or coal boilers. There may be an opportunity to sign long-term contracts with such offtakers.
These are smaller deals, not billion dollar projects, but I think you will start to see some of those in the market, more of the old inside-the-fence kind of projects.