Solar electricity is expected to be the cheapest generating source by the middle of the next decade. The winning bids in auctions to procure solar projects were 2.91¢ a KWh last August in Chile and 2.42¢ in September in Abu Dhabi. Upcoming tenders in other countries are expected to draw even lower prices. Five heads of US renewable energy companies talked at the Chadbourne global energy and finance conference in June about what the rapidly falling cost of solar means for the broader US power sector.
The panelists are Tom Werner, chairman and CEO of SunPower Corporation, Tom Buttgenbach, president and co-founder of 8minutenergy Renewables, Gabriel Alonso, CEO of EDP Renewables North America, Rob Freeman, CEO of Tradewind Energy, and Craig Cornelius, president of NRG Renewables. The moderator is Keith Martin with Norton Rose Fulbright in Washington.
MR. MARTIN: Solar surpassed wind last year for the first time in terms of new capacity additions. There were 42% more solar installations worldwide than wind. Solar generates during the day. Wind produces its maximum output at night. Gabriel Alonso, does this ensure an enduring role for both, or is solar about to blow past wind?
MR. ALONSO: The market will decide that. The market will demand the cheapest source of renewable energy that provides the highest value for consumers and the largest amount of flexibility.
I think solar has an advantage. There are many places in this country where wind remains more competitive than solar. From a value perspective, solar produces during the day, and wind produces both during the day and at night. Solar offers the flexibility of offering projects of 100 megawatts, 20 megawatts, five megawatts and distribution-level generation that wind does not offer. If I have to think long term, I can see solar offering a cheaper product with more value for consumers and providing a higher level of flexibility.
MR. MARTIN: Are you shifting resources into solar and, if so, to what degree?
MR. ALONSO: We are. We still see a lot of opportunities for wind. There are still many areas in the US where wind is more competitive, but with every passing year, the map that contrasts wind to solar opportunities is becoming more yellow.
We are shifting resources from wind into solar. We are building solar as we speak. We are entering into solar power purchase agreements for the next year and marketing other projects for future years. We also see a fiscal asymmetry in the early part of the next decade when solar will still qualify for an investment tax credit after the production tax credits for wind have expired. That will give solar a further advantage.
MR. MARTIN: Rob Freeman, is TradeWind solely a wind company?
MR. FREEMAN: We made the decision five years ago to jump into solar. We did wind exclusively for eight years leading up to that. We are focused solely on utility-scale projects. We now have a development pipeline of four to five thousand megawatts of solar. It is about two thirds the size of our wind project pipeline.
MR. MARTIN: Is solar your future?
MR. FREEMAN: Yes, I have been thinking that at some point we will build smaller and smaller amounts of wind each year. Wind will endure, but probably not at 6,000 to 10,000 megawatts a year of new capacity. Most of the consultants are predicting new wind capacity additions will fall into the 2,000 to 4,000 range. I have been expecting a big solar wave to hit. Probably the one thing that gives all of us pause is the Suniva case.
MR. MARTIN: We will come back to Suniva. Craig Cornelius, you see the market perhaps more broadly than the rest of the panel because you work for a company — NRG Energy — that has dabbled in the full range of technologies. What effect do you see the rapidly falling costs for solar having on the broader power sector?
MR CORNELIUS: The effect is already visible. The impact of behind-the-meter solar as well as utility-scale solar injected into the grid in places like California is to cap prices during the parts of the day when the grid is experiencing peak gross load. That has happened with just the solar that is in the system today, and it is very easy to add incremental solar generation. This means that wholesale power prices will remain low for the foreseeable future.
MR. MARTIN: Does solar have an advantage because it alone can continue to supply at lower and lower prices?
MR. CORNELIUS: Yes, but that is not the only thing we see. It was interesting to hear the perspective from some of our fellow wind developers. There are some parts of the US where wind continues to have a distinct advantage. When we look at the US, we see a patchwork as Gabriel Alonso said he sees at EDP. There are parts of that patchwork in ERCOT, the Southwest Power Pool, much of MISO and much of the Pacific Northwest where end-use customers and regulators aim to expand the share of electricity generated from renewable energy and where we expect wind to remain the least-cost best-fit solution. Solar will not necessarily beat wind in these areas. However, both wind and solar benefit from a lack of fuel price risk over the long term.
MR. MARTIN: Tom Werner, do you see solar transforming the US power sector, and if so, how?
MR. WERNER: What we have seen in California is an inverted peak. The utility regulators are responding by changing the electricity rate structures to encourage moving the solar energy generated to other parts of the day. This is creating interest in storage. It will change the electricity delivery infrastructure. The same thing is happening in other states like New York and Hawaii and will happen in Massachusetts.
MR. MARTIN: A few years ago, European solar developers came looking at the US solar market, and some of them turned away. They said the returns were too low. They went back to Europe or to Africa, Latin America, the Middle East. Have solar developer returns improved in the US and where are they in relation to what a wind developer might earn?
MR. ALONSO: We looked into solar years ago and decided not to make the move for two reasons. One was the low returns. We were talking about 150 to 250 basis points difference in unlevered after-tax returns between wind and solar. We did not have endless capital. The wind opportunities were enough to help us keep growing with the capital we had available.
The reason we are now looking into solar is not that we now have significantly more capital or the returns are converging. We still earn a higher return on wind projects. However, solar is becoming more and more competitive. It is more competitive than wind in states where wind had a clear advantage only a couple years ago. Utilities and corporate customers are demanding the cheapest source of electricity, and solar is the cheapest one.
MR. MARTIN: Solar is a way to win a power contract.
MR. ALONSO: Exactly. We are a renewable energy player. We will offer the cheapest source of electricity. If that is wind, we will offer you wind. If that is solar, we will offer you solar.
That is what is driving us to shift resources into solar, even though we still do not see the returns on the two types of projects converging.
The wind companies will eventually have to pull up our socks and accept the fact that some of the traditionally windiest days will no longer be as windy. They will be more sunny. The winter winds will diminish. Wind projects will earn lower returns, leading to a convergence in the future.
MR. MARTIN: What are current returns for wind and solar projects?
MR. ALONSO: The after-tax unlevered returns for both wind and solar are in the single digits, but I still see a spread of 100 to 150 basis points between the two.
MR. BUTTGENBACH: Would you see the same spread if the returns are adjusted for risk? Solar is much more predictable. Our P90 output projections are very close to our P50 projections, which is not the case for most wind farms.
MR. ALONSO: That is a very astute question. When we look at the difference between P50 and P90 risk-adjusted return, taking into account volatility around wind resource on a year-to-year basis, and especially once you start to look at shape risk around projects where the developer is taking basis risk, the returns are more equivalent.
MR. MARTIN: Same gap otherwise: 150 basis points?
MR. ALONSO: Maybe not quite that much but there is still a significant premium for a wind project over solar. One of the things that is often lost in these discussions is the capitalized dry-hole cost for wind development tends to be more significant. If you look at the full life cycle of development and the returns that are realized by developers on successful projects are offset by the costs of projects that are not completed, there is a little more of a convergence in returns for wind and solar.
MR. WERNER: There is another interesting aspect here, and that is the breathtaking drop in PPA prices for solar. The winning bids for solar two years, 12 months, even four months ago look really good for a developer compared to what a developer can get for solar electricity today.
MR. MARTIN: How is that affecting developer returns?
MR. WERNER: The price is a factor in the return. The developer must figure out how to get to those numbers. We do a spreadsheet in order to back into the number. We look at the solar panel prices and balance-of-system cost that will be required to deliver electricity at these prices and what that leaves in terms of developer return.
The solar equipment providers bailed out the developers the last couple years on forward pricing. Can that go on forever? I think that is highly unlikely. We are running out of room to sustain the same rate of solar panel price decline, and yet the bids assume that costs will continue to come down at the same rate. In the next couple years I think things will become really interesting on the solar side.
MR. MARTIN: The business model of many solar developers was to win power contracts by bidding at prices that cannot be delivered today with the expectation that by the time deliveries must start under the contract, solar panel prices will have fallen further, making the contract economic to perform. You are a vendor as well as a developer. Isn’t it in your interest to say the rate of cost decline is unsustainable?
MR. WERNER: The profit and loss performance of the publicly traded solar module manufacturers last quarter was terrible. I don’t mean to sound like a whiner. I am just saying we have done the math, and we do not believe the recent rate of decline is sustainable. Over the next couple years, will costs continue to come down? Yes. Will they come down at the rate that developers are assuming in recent bids? I have my doubts.
MR. MARTIN: Gabriel Alonso, EDP is shifting resources to solar in the US. Is it doing the same thing outside the US?
MR. ALONSO: We are looking at solar opportunities elsewhere, but the European market is not active in either wind or solar, apart from offshore wind. That said, we are open to doing both wind and solar wherever we are.
Returning to the subject of risk adjusted return, we still see higher wind returns on a risk-adjusted basis. We have 5,000 megawatts of wind capacity in the US running at 98% availability. When I go to my board with a solar project and explain that it is just a bunch of solar panels that will remain in place for 30 years with no operational risk, it is not really reducing my incremental operating risk by much because of the way we run our winds farms.
I can understand how others may not look at it that way. For example, the way the two tax credits work — a 30% investment tax credit taken at inception on the cost of a solar project versus production tax credits taken over time on the electricity output from a wind project — can show a higher net present value to the after-tax cash flow from a solar project than a wind project. The point is the wind industry has come a long way to minimize the P50 to P90 spread.
MR. FREEMAN: As a developer to the extent that we have contracted solar assets, we have seen higher returns on our development capital.
MR. MARTIN: Higher for solar?
MR. FREEMAN: Correct.
MR. MARTIN: Which is the reverse of what I thought you and Gabriel said earlier.
MR. FREEMAN: To be clear, not the return that goes into the project pro forma but the return on development capital. The money that is spent to get a project to the point where it is ready to be built.
MR. MARTIN: Gabriel Alonso, you said the reverse.
Barriers to Entry
MR. ALONSO: I am not disputing Rob’s statement. But Rob develops and flips. I own long term.
There are no barriers to enter the solar space. I always tell my team there are two things my grandmother can do: develop a wind farm in Texas and develop solar anywhere in the US. [Laughter]
When you are doing a 200-megawatt wind project in Kansas, you need 40,000 acres. You need to lease a ton of land. There are a lot of studies that have to be performed and a lot of other work that goes into development before the project is marketable. By comparison, very little up-front money is required to develop a solar project. We have been offered deals where people have only a dot on the map. They do not even have an interconnection queue position or easements to access the site and they are already bidding the project into utility requests for proposals and winning PPAs. That is not what happens in the wind industry.
MR. MARTIN: Doesn’t that frighten you? It is the two guys with the Avis car again, signing up contracts that offer prices so low that the contract cannot be performed. Why go into that sort of market?
MR. ALONSO: The line between craziness and being a hero is very fine. A lot of those crazy people five years ago are millionaires today. So I don’t know if I would call them crazy anymore.
We are a renewable energy provider. If the cheapest source of power is solar, then we will have to adjust and play by those rules.
MR. FREEMAN: We are particularly active in SPP and MISO on wind, and we are seeing a lot of interest in solar among investor-owned utilities in what have traditionally been wind-centric markets.
MR. MARTIN: Are the utilities interested in PPAs or in taking the project from you and putting it into rate base?
MR. FREEMAN: Both, but on the solar side, we are seeing more PPA interest today. They are sticking a toe in the water. Our expectation is that we will see solar projects built in places like Oklahoma or pick your state in SPP and MISO where the delivery cost of solar is higher than wind. It may be considerably higher, but solar has distinct advantages.
MR. WERNER: I could not agree more. Utility-scale solar is being built in 40 states. The investor-owned utilities are embracing large-scale solar. There may be a bit of a breather this year because there was such a massive amount in 2016.
Regarding this fine line between a hero and a millionaire, you hear more about the millionaires than you do about the failures. I agree that two years ago, people were astounded at the bets some developers were making to win contracts and those developers did really well, but that does not go on forever. The forward prices that people are projecting for solar have gotten a little out of hand. I went to see a community choice aggregator that had not even hired a procurement person yet, and it said it is expecting to pay electricity prices in the low $30-a-megawatt range.
MR. BUTTGENBACH: One thing a lot of folks forget is that solar PV is a technology play. It is less of an energy play. When we started our company in 2009, there were 320 developers in California. There are now four, after SunEdison went bankrupt, that have more than 1,000 megawatts of installed capacity. SunEdison showed that if you do not do it right, you can lose a lot of money. We have done well as a company, so it can go either way.
How many of the wind companies have an R&D facility? Do you guys test your turbines?
MR. ALONSO: I always tell my team the same example. When you have a car that breaks, you take it to the shop and you get it back in two days and you do not know what the hell happened, but it is running and you pay for it and you are happy. That is the opposite of how we run. We know our turbines, and we have taught turbine suppliers how the technology performs. We know the technology that we installed, and the goal is to know it better than turbine suppliers. We run tests in partnership with the turbine suppliers. There is no other way.
MR. MARTIN: You may be unique because you were head of wind turbine vendor Gamesa’s operations in the US before moving to EDP.
MR. ALONSO: We are looking at technology ideas that we share with turbine suppliers. We have collaboration agreements. We work together on improvements. Do we have an R&D facility where we are testing blades? We do not, but we think we have a cheaper and smarter approach.
MR. CORNELIUS: The major turbine vendors do it themselves, and we make use of data that comes from those R&D facilities to calibrate performance expectation.
MR. BUTTGENBACH: A lot of people do not understand this, but what is happening in the solar industry today is it is becoming vastly more complex. There are a lot of technologies that have been around for quite a while, but they have not been deployed because the focus was on plain-vanilla PV modules. Today, we are looking at heterojunctions. We are looking at bi-facials. We are looking at all kinds of new technologies. We run an R&D facility with 52 different modules that are being tested and evaluated. We are ahead of the industry so that we can remain competitive and understand the materials that go into every module. We monitor that at the factory level. You have to have that kind of insight to be competitive.
I hear the talk about the forward curve in the last couple years during which solar panel prices came way down. Will this trend continue? As a developer, I have to make a well educated guess. If you want to compete in Texas for sub-$30 PPAs, you have to know where the technology is headed and what you can and cannot build and finance.
MR. MARTIN: Let me take this in a different direction. People expect energy storage to be transformational when it becomes economic to install. Which technology will be helped more by storage: wind or solar?
MR. CORNELIUS: Probably solar based on what we see in the markets we service with both. The reason is the greater ability to boost the effective load carrying capacity of an incremental solar generator with cheap lithium ion batteries than is the case for wind. When we think about how a state can move from a 33% renewable portfolio standard to a 50% RPS or more and the capital that will have to be deployed to produce those megawatt hours, solar will have an outsized share only if it has a cheap means to store solar generation during the sunniest hours for release later in the day. We see a forward trend that leads us to believe that will be possible. We look at developing projects in anticipation of being able to offer firm blocks of power to utilities by making use of batteries. We see fewer opportunities to make wind more competitive with storage.
MR. WERNER: The math for utility-scale storage does not work today.
MR. MARTIN: When will it work?
MR. WERNER: It works today in the commercial and industrial sector where storage can be used to eliminate demand charges by utilities. We are seeing the beginning of a massive ramp up of storage attached to solar systems for the commercial market in California. We will triple the attach rate next year compared to this year. As you get scale, the costs will come down, and then you will start to see adoption in other market segments as well.
MR. MARTIN: Tom Buttgenbach, I read that 8minutenergy Renewables is now focusing on storage. You see it as a growth area. When do you see yourself routinely adding large batteries to utility-scale solar?
MR. BUTTGENBACH: We have 1,000 megawatts of storage projects currently in development. We are working on the first deployment of batteries with lithium-ion technologies in the next two years. Construction starts in 2018 with deployment by the end of 2018. We are talking about 100-megawatt batteries. We see them as already competitive in the right mix. You do not have to have a 100-megawatt battery with a 100-megawatt PV plant so you can right size them. You can solve a lot of problems for the utilities by providing ancillary services that solar companies are not in a position currently to provide without batteries.
In five years, depending on the market, I think the old model of “I build you a PV plant and you have to buy my power whenever I produce” will be dead. I think the utilities and other customers are going to demand power when they need it.
Storage for solar is a very good combination. The more predictable output from a solar project means you can right size the battery more easily with solar than with wind.
MR. MARTIN: Gabriel Alonso, Rob Freeman, that’s smack talk. [Laughter]
MR. ALONSO: They are right. I agree that storage will help solar first more than wind because you can postpone the dark portion of the output curve for two to three hours to great effect. You are shifting a much steeper curve than postponing 2 a.m. wind production all the way to 8 a.m.
We are interested in storage. The costs are coming down quickly. While the benefits will be greatest for solar in the short term, it will eventually help put wind projects in a position to provide ancillary services, which is important in a market where we are entering into PPAs with prices in the teens and twenties.
MR. MARTIN: Providing ancillary services will help make up for lost revenue.
MR. ALONSO: One or two dollars more in revenue is another 10% to 20%. It is huge. If we had spoken about this years ago, I would have said you can keep the extra dollar for yourself.
MR. MARTIN: That may be what makes storage economic ultimately.
MR. ALONSO: Even if the economics of storage cannot be made to work, we are working with turbine suppliers on turbines that will enable us to provide ancillary services. The extra one to two dollars can make a big difference to our competitiveness in the market.
MR. MARTIN: We are growing short on time. I have two other subjects I want to tackle. One is the Suniva petition. Suniva is a solar cell and panel manufacturer in Georgia. It has asked the US government to impose a tariff of 40¢ a watt on imported solar cells and a floor price of 78¢ a watt on imported modules. Tom Werner, how will the market be affected if tariffs at anything close to these levels are imposed?
MR. WERNER: Terrible effect. The idea that we will impose tariffs to save a small number of jobs and put in jeopardy about 10 times the number of other jobs is ridiculous. It makes no sense.
It is already having an effect. Companies are rushing to buy solar modules to put in storage. This is already putting upward pressure on prices. I don’t know what good comes of it. The costs go up for everybody.
MR. MARTIN: Do you think a tariff will ultimately be imposed? Trump has the final word. He announced the other day that he wants to finance the wall on the Mexican border by putting 1,250 megawatts of solar panels on top.
MR. WERNER: Imagine you are the CEO of a solar company and you have to plan around that scenario. [Laughter] There are a number of variables that you would have to take into account for purposes of planning. The interational trade commission will look at this first and then it goes to the President. We will know whether a tariff will be imposed by the end of this year or early next year.
MR. MARTIN: I did the math. If you take Suniva at its word about the percentage of US solar panel manufacturing capacity it represents, we are talking about 979 jobs being saved against some significant share of 260,000 other jobs in the US solar sector put at risk.
MR. CORNELIUS: When we look at addressable market, even with average panel prices floating up to 45¢ a watt, a tariff at these levels would eliminate at least half the incremental capacity additions that folks had expected to see happen over the next three years and puts them out of reach. What we have heard over the last few weeks as conventional wisdom from some suppliers, which is absolutely faulty, is that the industry can maintain returns at a constant level by increasing PPA prices to accommodate something like a 45¢-a-watt panel price. What is setting the prices in new PPAs today is not solely the competition among developers. We are competing against utility avoided costs. We are at multi-decade lows in real terms for wholesale prices because of how efficient new gas-fired generation or wind is. Since load is not growing, those prices are not going up.
MR. MARTIN: Are we already feeling an effect of the Suniva petition? You are bidding for PPAs in the solar market. How do you bid not knowing whether such a large tariff will be imposed?
MR. BUTTGENBACH: Carefully. [Laugher]
MR. CORNELIUS: Agreed.
MR. FREEMAN: I think the assumption is that people are going to look for a way out. They are going to have to have some ability to terminate contracts.
MR. MARTIN: Question from Jack Cargas, managing director of the tax equity desk at Bank of America Merrill Lynch.
MR. CARGAS: What does the panel think about risk allocation tied to the potential solar tariffs? Tom Werner says the possibility of a tariff is already causing market disruptions. Others are saying that maybe developers will ask for the ability to terminate contracts. But what about deals that are getting done currently that are expecting to have solar panels delivered shortly after the potential imposition of tariffs? Who is taking that risk? Are developers taking that risk? Are sponsors taking that risk? Financiers will not take that risk.
MR. FREEMAN: People are locking up their panel purchases. There has been a big run on panels and maybe that capacity is gone at prices that can support current bids.
MR. CARGAS: My understanding is that the president will decide in November whether to impose tariffs, and the imposition will be immediate.
MR. MARTIN: Trump has 60 days after the US International Trade Commission makes a recommendation on November 13.
MR. CARGAS: So what happens if and when the tariff is imposed, a project is expecting panels, but the panels have not crossed the US border.
MR. ALONSO: If that happens, you have a problem. I mean, what can I tell you? You know the answer but are looking for someone to say something better. [Laughter]
MR. MARTIN: So the financiers are not taking the risk. Developers are not taking the risk. Tom Werner, I guess it comes down to you.
MR. ALONSO: Actually, developers are. If President Trump slaps a tariff on wind turbines and I have a project at which I plan to use the turbines being built next year with a PPA in place that I cannot terminate and I am locked into the turbine purchase, then I have a problem. I cannot deny it.
MR. WERNER: Somebody loses for sure if this happens. There are ways to mitigate the risk without breaking your balance sheet. You can only buy so many solar panels. If you can buy a couple years of solar panels, I definitely want to see you after this session. I have a deal for you. Don’t pay attention to the tariff figures in the Suniva petition. If you are Suniva, of course you ask for the sky and the stars. We will see what happens, but if tariffs end up anywhere close to what Suniva is requesting, then somebody is going to lose for sure, and we don’t need to negotiate live who that will be because, looking at two developers on either side of me, I don’t like my odds right now. [Laughter]
MR. CORNELIUS: We are only in week three after the US International Trade Commission announced it would launch an investigation into whether solar panel imports are causing enough injury to domestic manufacturers to warrant tariffs or other import relief. We had stop-start experiences around fee and tariff reductions in various countries in Europe throughout the 2000s. If there’s anything that the photovoltaic supply chain and industry have demonstrated over these last 15 years of growth, it is an ability to be nimble and adjust with commercial and supply chain structures that sustain growth.
For our near-term projects that are entering the financing phase, we have solved the problem already. We have the solar panels in hand. For projects that are not yet at that state, we plan to wait to assess options and figure out commercial structures that accommodate whatever happens with the tariff. We hope ultimately that we will see that the public policy and economic interests of hundreds of thousands of employees will supersede the financial interests of a few investors who made a bad investment in a small solar panel manufacturer.
MR. MARTIN: Last question. I will go across the panel. Just give me a short answer. What is your single biggest challenge as a developer?
MR. FREEMAN: The transmission system.
MR. MARTIN: Not enough transmission.
MR. WERNER: Risk-adjusted buyer IRRs in Mexico.
MR. MARTIN: Say more.
MR. WERNER: Mexico sort of blew up post-Trump putting lots of supply projects that won contracts by bidding low prices in auctions in a more challenging position. The developer IRR assumptions changed significantly. They did not change as much in some other Latin American markets. Managing that is an issue.
MR. CORNELIUS: Discipline among fellow market participants in the contract structures.
MR. MARTIN: So your competitors are bidding unrealistically low prices. Tom Buttgenbach?
MR. BUTTGENBACH: The Suniva petition keeps us awake at night even though we have hedged for all projects we are building in the next 24 months.
MR. MARTIN: Hedged by bringing in panels already?
MR. BUTTGENBACH: And by having contracts with US manufacturers that are not affected by the tariff.
MR. MARTIN. Gabriel Alonso, you get the last word.
MR. ALONSO: There is additional basis risk that we are taking with most of the newer corporate PPAs. That and grid congestion are the two biggest concerns. The inability to upgrade existing transmission lines or to build new ones in this country is a serious problem.