August 21, 2026
Returns for project developers have increased to the low double digits to compensate for increased volatility in government policy. As many as 170,000 MW of wind and solar projects were under construction in time to lock in federal tax credits. How much of an electricity price increase will be required for projects that no longer qualify?
The US is projected to install a record 86,000 MW of new utility-scale electric generating capacity in 2026, with 93% of it coming from solar, wind and energy storage. How quickly will gas gain market share? Share prices for gas turbine, nuclear and some fuel cell manufacturers have gone up dramatically this year.
Four CEOs and the head of project finance lending at a large bank talked about these and other subjects at our 35th energy finance conference outside San Francisco in mid-June. The panelists are Hunter Armistead, CEO of Pattern Energy, Cassidy DeLine, CEO of Linea Energy, Declan Flanagan, CEO of Bluestar Energy Capital, Greg Hutton, head of project finance for the Americas at Dutch bank Rabobank, and Kevin Smith, CEO of Cypress Creek Renewables. The moderator is Ben Koenigsberg with Norton Rose Fulbright in New York.
Post-July 4
MR. KOENIGSBERG: Kevin Smith, let's start with you and work across the panel. How do you expect the development pipeline for renewables to evolve after the July 4 deadline this year to start construction of solar and wind projects? Do you expect no change, a drop-off in new projects or a reset in how renewables projects are priced and financed?
MR. SMITH: We heard earlier today that about 170,000 MW of projects are expected to have started construction for tax purposes by July 4. I have heard estimates of between 100,000 and 250,000 MW. People will be in head-down execution mode for the next several years. They have four years to finish after the year construction started.
It will be interesting to see what happens over the next couple years as newer projects that do not qualify for tax credits come to market.
MR. FLANAGAN: I agree with what Kevin said. The industry has been dealing with the construction-start rules for more than 15 years. There is a huge backlog. Historically, power contracts rather than construction-start dates were the limiting factor. That could flip for the first time. The deadline to complete projects on which construction starts in the first half of 2026 is the end of 2030. If there is no extension of the tax credits, in my view 2030 will be the big transition year.
MR. HUTTON: I see a phased ramp down to projects with tax credits. There will a bulge of projects that did not get under construction by July 4, but can still qualify for tax credits if they are completed by the end of 2027.
Those projects will require a lot more diligence around completion risk to get financed.
It gives me PTSD to remember the days of the one- and two-year tax credit cliffs. That is the most challenging and interesting part to me as a lender, and then we expect the repricing of risk and capital that will happen after 2030 that Declan just mentioned.
MS. DELINE: I have a slightly different take. There will be plenty to build. The cliffs and deadlines have put an opportunity cost on what gets built. People are being forced to determine the highest and best uses of high-voltage transformers, capital and team resources. We are starting to see a culling of pipelines that frankly needed to happen and needs to continue to happen.
MR. ARMISTEAD: We are all anxiously awaiting the time when we do not have to be so sophisticated in how we start construction, where we procure our equipment and how we finance projects using elaborate tax equity structures. Right now we are playing a jigsaw game, and I am excited about that going away for freaking ever.
MR. KOENIGSBERG: We are just going to have AI take care of it.
MR. ARMISTEAD: This is beyond AI. AI is not there yet. Trust me.
MR. KOENIGSBERG: Hunter, let's continue with you and then swing back across the panel. Power prices are already trending up. If post-July 4 economics for renewables deals pushes power prices higher, how much higher do they need to go for solar and wind to work economically?
MR. ARMISTEAD: They are already high enough for solar and wind to clear. Solar and wind are the most economic and lowest-cost technologies. For the longest time, we had no load growth. The only thing we were competing against was gas prices times heat rates, and we were trying to undercut the marginal cost of gas projects. That was it. The gas industry did not love that the tax credits made solar and wind more economic. We are able today to compete without the tax credits.
MS. DELINE: I agree. Solar may not be competitive in every state, just as wind may not be competitive in every state, but certainly one of them is.
MR. ARMISTEAD: The real cost is not the solar or wind. It is the cost of connecting to the grid and of the grid upgrades required to accommodate the additional electricity. That, rather than tax credits, is more relevant to what gets built.
MR. HUTTON: We have an in-house research team. Our calculations suggest electricity prices would have to rise at least 30% to 50% in the post-tax-credit world to pencil out. I am curious what the guys actually holding the pencils see.
MR. ARMISTEAD: Not that much. If we can eliminate the current need to figure out how to finance things 18 different ways with starting construction and tax equity and monetization, it will reduce costs. Our weighted average cost of capital is probably overstated by more than 150 basis points.
I think prices are going to go there anyway, but that is a different point.
MR. FLANAGAN: Renewables are way cheaper than gas. Relative to liquid metal-cooled small modular nuclear reactors, they are infinitely cheaper.
The bigger issue on the consumer side is that we still have a capex inflation problem, both with regular inflation and imposed tariff inflation and imposed risk.
We cannot as an industry be complacent while the ultimate consumers of electricity are forced to reset electricity from all power sources to more than $100 a MWh. Renewables will remain competitive in a post-tax credit world against all of the alternatives, but this is not reflected yet in what ratepayers are being asked by utilities to pay. That is the coming challenge.
MR. SMITH: Our calculations are that a $60 per MWh power price in MISO with 40% tax credits goes to $85 or $90 without tax credits, plus or minus depending on interconnection costs and other factors. That is still substantially below natural gas, which is in the $120 to $150 a MWh range.
Add battery storage, and I think solar-plus-storage is still cheaper than natural gas.
Keep in mind that natural gas projects have to pay for fuel. Somebody is taking the long-term risk on natural gas prices going up. We can provide a flat price on solar with battery storage. You cannot do that with natural gas.
Returning to the cliff issue that Greg Hutton raised, a 500-MW solar project requires 18 months for construction. You would have to start construction by the middle of 2018. However, the lenders are not going to feel comfortable with a one-month safety margin. They will want six months. That means work on the project site must be underway in earnest by January 1, 2018 for a project that must be completed by the end of 2029. It means power contracts for such projects are going to be signed in 2027. PPAs signed in 2027 are going to start to reflect nervousness about power prices needed to support projects whose completion dates miss the cliffs.
Market Shares
MR. KOENIGSBERG: Declan Flanagan, how quickly do you expect gas realistically to take a larger share of new capacity additions, given the current constraints around turbine manufacturing and delivery?
MR. FLANAGAN: To start, it is not a zero-sum game of renewables against gas. Therefore, I don't see gas taking a market share. The current demand for additional electricity is so great that there is a need for whatever can get online.
I think what you will see in 2027 is a year of the engineer over the equity capital markets and bankers. You will see interesting conversations about torsion on generator shafts and the realities of liquid-metal-cooled fast reactors that the developer is saying will be online by 2030, and it just won’t. The year 2027 will also be a year when people face up to the scheduled realities of new gas.
MR. ARMISTEAD: Is there a prediction market where we can get an over or under on the first nuclear, because I am going long.
MR. SMITH: I am with you.
MR. KOENIGSBERG: Perfect segway. We have seen strong momentum around DOE-backed financing and renewed interest in AP1000 nuclear projects. Is nuclear viable at scale with government support?
MR. SMITH: Without government support, no. With government support, yes. That is the only way you are going to get any kind of substantial nuclear build-out, in my view, whether it is SMRs or whether it is large-scale projects of 1,000 MW or more. You are talking about a $10 billion investment in a single large project. How many people are really going to do that? There is a single point of failure.
You are looking at power pricing of around $200 a MWh around the clock.
The good and bad points of solar with a battery are it is not 24 hours a day. It can meet those peak periods, but you are not paying for it at two in the morning. You have to pay for the nuclear plant to run at $200 a MWh all night, or it will not pay back its debt. How much peak power do we need at 2 a.m.? Who will commit to a power contract at $200 a MWh 24 hours a day? A utility that is able to pass through the costs to ratepayers may. Heavy government support is the only way nuclear is being built around the world. I started my working life as a design engineer on nuclear facilities.
The Vogtle project is interesting. I was at SolarReserve at the same time that Vogtle was going up, and there were comparisons between our molten-salt storage projects versus nuclear. Vogtle was a disaster from a capital cost and schedule standpoint. A decade or more later, people point to it as the last success on the nuclear side. It was not a success. It was a disaster, people got fired, and utilities almost went bankrupt.
We will see a few repowerings of existing nuclear projects. We will see some pet projects. Bill Gates is going to build a 500-MW nuclear facility in the middle of the desert somewhere, but it will be 25 years before we see tens of thousands of megawatts of new nuclear facilities.
MR. KOENIGSBERG: Let’s drop from AP1000s to SMRs. Cassidy DeLine, given the long development timelines and the history of cost overruns, can nuclear realistically address near-to-medium term power demand, and do SMRs meaningfully change that equation?
MS. DELINE: I agree with my colleagues. I do not think they do. As a citizen, I am rooting for nuclear. We need electricity, and we need clean electricity. There is a tremendous amount of hype around SMRs, particularly in Washington. One thing that absolutely will happen is there are going to be policies from both sides of the aisle that are targeted at promoting nuclear.
Electricity is a commodity, and it needs to be built quickly and on a cost-effective basis. Economics 101 suggests scaling things to make them larger is more likely to reach economies of scale rather than making them smaller.
SMRs are already starting at a higher cost basis than where the AP1000 was, which was over a decade delayed and billions of dollars over budget. They have to get to that level and then the cost has to come down to be cost effective.
There has been a lot of hype around some SMR manufacturers achieving criticality. I am not a nuclear engineer, but my understanding is that just means that fission has occurred. It doesn’t mean that it can happen at the scale it needs to happen. It doesn’t mean that it can happen safely. The majority of the design work still needs to be done. You have at least one SMR company under federal investigation for fraud because it claimed it had a design that it did not in fact have. I think we are likely to waste a lot of time on SMRs and then what gets built is something along the line of the AP1000, which is what China is building.
MR. KOENIGSBERG: Greg Hutton, do you think nuclear will appeal to bankers?
MR. HUTTON: Maybe in the future. Given the cost overruns and the attenuated timelines for large-scale nuclear, we have been wondering as a thought exercise what SMRs look like and how they solve some of those problems.
They have a better profile in some respects. They are modular. Parts can be fabricated offsite. Installation is faster. Those features solve some of the problems with large-scale nuclear projects. However, no one has put a confidential information memorandum on my desk yet for an SMR financing.
The first-of-a-kind technology risk makes them tough to finance. There are multiple competing technologies. These are demonstration projects at best. The projects are so nascent that we have not thought through the permitting risks. If the theory is you can put them closer to load, then there will be a lot more scrutiny in the permitting process and of security concerns, if they will be closer to large population centers.
All of this seems a long, long way off. I know the DOE has a timeframe of the late 2020s to early 2030s, but I don't think that is realistic.
MR. ARMISTEAD: I agree with Cassidy. We should all be cheering for it because, if we could have reasonably cost-effective zero-carbon energy, that would be a wonderful thing.
We would love to develop them because we love developing hard things, but we don’t like developing hard things with technology risk. Right now, I don’t think it is an investable thesis for people like us.
It is an investable thesis for the government to prove that the technology works. Fission has been in use for decades. It works, but is it economic? I think there is more chance with the AP1000, but at prices in the range that Kevin Smith said.
MR. SMITH: Don't get me wrong. I worked for five years on nuclear design for projects in Illinois in the early 1980s when all kinds of nuclear projects were being built. I am a big fan of our nuclear fleet. I think it is safe. I think it is reliable. What killed the nuclear industry was not safety issues, notwithstanding Three Mile Island. It was economics, and economics continue to be the problem.
Retribution
MR. KOENIGSBERG: Hunter Armistead, the Federal Aviation Administration has issued very few no-hazard determinations for new wind projects over the past year. Are wind projects still being developed and financed? Or is the market shifting away from wind?
MR. ARMISTEAD: We are closing one early next week. Wind projects are still being built, and very large ones, too. Data centers need the electricity. The slowdown at the FAA is the Achilles heel. We are still getting no-hazard determination certificates. We got some within the past week.
MR. KOENIGSBERG: Can you share how you are able to do that?
MR. ARMISTEAD: I think they are historical stuff. Most people in the government want to do their jobs; they really do. They think that they have a task, they’re proud of doing it, and given the opportunity to do their jobs, they get them done. The ones we are getting were already on the verge of being released before last July when the FAA started to slow down approvals. There is no interaction with the Department of Defense. They had already been cleared.
I don't understand why the government would actually reduce the supply of the one thing -- electricity -- that is holding our country back.
MS. DELINE: There are a lot of Republicans who are saying exactly that. It is our job as an industry to encourage them to be louder. This administration is right that we have a national energy emergency. Everyone in this room is acutely aware that all the solar projects generate electricity during the same daily window period. Wind is needed. We are also seeing FAA determinations trickle out.
MR. ARMISTEAD: Governors need to deliver that message to Washington.
MS. DELINE: When talking to Republican governors, two things come up. One is the Keystone pipeline, and the other is the LNG permitting pause that the Biden administration imposed during its last year in office. This is 100% retribution. We have to stop fighting with other forms of electricity generation. We are not in it to replace gas. We do not even have a solution to replace gas.
MR. ARMISTEAD: Some people in our industry started it. Let’s be clear.
MS. DELINE: We absolutely started it.
MR. ARMISTEAD: It was the most ridiculous thing that gas will be phased out by 2030 or even 2040.
MR. FLANAGAN: This is a key point about Keystone. There has been a lot of bad energy policy in the last 12 to 18 months, but it all goes back to ripping up the existing permits on Keystone. That was the direct predicate to what we are experiencing now.
Back to your question, wind projects are being financed. We have had a few large wind projects reach notice to proceed with construction this year. People are doing wind with their heads down and their mouths shut.
No-hazard determinations are coming out of the FAA, as other people said. In 2027, I think you will start to see the reality set in that other projects are being delayed, and the government will start to recognize the insanity of throttling back the cheapest and quickest generation. Wind and solar development are being artificially slowed by the national government, but projects are being financed.
Tariffs
MR. KOENIGSBERG: How is tariff risk affecting your business? Are you able to allocate most of the risk to equipment suppliers?
MR. FLANAGAN: Inflation in capital costs generally is a big problem. Just because it is perhaps a bigger problem for other competing sources should not make renewables developers complacent. The capital cost to build new wind projects today is the highest I have ever seen. We risk a lot of political pushback in terms of the ability of end consumers to absorb the price. The burden is hidden in the pricing dynamics in the PPA market, but you have to look at where power markets are clearing on a day-to-day basis and what is reflected in customer tariffs. Add in the $100 billion in costs for additional transmission capacity. That needs to be paid for as well. Then add tariffs to that.
For projects with good schedules, you can get suppliers and offtakers to work with you. The issue is more the uncertainty of what is the next shoe to drop and what might happen a month from now. Tariffs are adding a lot of volatility to an already volatile situation. It is a problem, but it has been manageable so far.
MR. SMITH: I agree with that. Uncertainty around tariffs is a big issue. It is one of a number of issues that affect pricing. Tariffs are leading to higher electricity prices across the board.
Project returns have had to go up to reflect the volatility and risk. Five years ago, developers were chasing single-digit returns. We are a long way today from single-digit returns. That is reflected in a higher cost of equity.
Some of the risk is built into capital costs.
Equipment suppliers are not stepping up and taking big tariff risk, nor are the developers, nor are the electricity offtakers. Maybe they take a little bit, but you put the risk in your returns and you put it in your overall capital cost forecasts.
MR. KOENIGSBERG: Greg Hutton, then it comes down to lenders. Are lenders consistently requiring sponsor guarantees or other credit support to cover tariff risk?
MR. HUTTON: Not consistently across all deals. Of all the walls in our industry such as the wall of money and the wall of demand, the tariff wall is my least favorite. It causes a lot of uncertainty.
Understandably, this is key diligence item today. A colleague produced a tariff matrix that was a deliverable in one of our deals. Whenever anything is called a “matrix”, you know it is not a simple thing. I was impressed by the level of detail, with analysis down to the sub-sub-component level, country of origin, the existing tariffs, the potential for future tariffs, and quantification of the risk and how the total capital cost of the project could be affected.
If the potential effect on capital costs is in the 1% to 2% range, we look to the construction contingency account to absorb the higher costs. Beyond that, depending on the magnitude, we look to other instruments like sponsor support, letter-of-credit backstops and contingent equity to mitigate the risk. The analysis is done on a case-by-case basis.
FEOC
MR. KOENIGSBERG: Cassidy DeLine, how are the new FEOC restrictions on use of Chinese equipment and intellectual property constraining project development, and how are you managing FEOC risk in practice?
MS. DELINE: We get a lot of help from Norton Rose. FEOC is uncertain but navigable. There obviously needs to be further guidance.
Another restriction that is very much top of mind is potential new limitations around cyber and remote connections to any equipment that comes in contact with the grid.
MR. KOENIGSBERG: Declan Flanagan, do you agree?
MR. FLANAGAN: We have been mostly focused on wind in the last 12 months, so the situation around FEOC is more straightforward. It is just another variable. Add together the FAA stuff, tariff stuff and FEOC stuff. We should not be making it this hard to generate more electricity.
To Hunter’s point earlier, we like doing complicated stuff. It drives up our margins. That is partly why returns and margins are now higher. But as a public policy, all of these new policies are just slowing down projects and adding friction to the system at the time when the system needs it the least. These are a growing list of self-inflicted wounds at a time when we can afford none.
MR. ARMISTEAD: This is unnecessary and expensive complexity for no reason. Balancing the grid, finding the right product at the right place, finding the right resources, figuring out how to interconnect, dealing with complicated supply chains and raising capital are complicated enough. The honest truth is the government does not want to see any more wind and solar.
The biggest FEOC issue to me right now is the fear that US banks could be viewed as prohibited foreign entities. If US banks are prohibited foreign entities, then there is something seriously wrong with the statute.
Lightning Round
MR. KOENIGSBERG: We have to wind up. I have three statements. Say agree or disagree. One-word answers.
State and local resistance to data center power consumption will meaningfully slow new development of data centers even as AI demand continues to surge. Agree or disagree?
MR. SMITH: I disagree because I think they are just going to locate someplace else. You get somebody objecting, you move down the road or you move to the next state.
MR. FLANAGAN: I disagree because a lot of the projects generating opposition are not real to begin with. I would love to go to some of these community meetings and explain the 10 reasons why this data center will never happen so they do not need to worry about it.
MR. HUTTON: Disagree, it gets redistributed.
MS. DELINE: I agree.
MR. ARMISTEAD: Agree.
MR. KOENIGSBERG: The phase-down of renewable tax credits will strengthen the renewables industry by forcing discipline on costs and returns and should not be reversed by future policy.
MR. ARMISTEAD: I can’t put enough exclamation points on agree.
MS. DELINE: Agree.
MR. HUTTON: Agree.
MR. FLANAGAN: Agree.
MR. SMITH: I am going to take a few words. I agree, but I think there is so much competition in the industry. We have enough discipline. We can compete without tax credits right now. The issue is affordability. And I think the Democrats will start pushing tax credits as an affordability issue, and they may have a point.
MR. KOENIGSBERG: Current FEOC restrictions and tariffs will result in a permanent build-out of US domestic manufacturing for renewable energy, which will not change even if the FEOC restrictions and tariffs are rolled back under a future administration.
MR. SMITH: I disagree. I think we still need something to support US manufacturing.
MR. FLANAGAN: Disagree. We are seeing a transient build-out.
MR. HUTTON: Disagree.
MS. DELINE: I agree with my colleagues. I disagree with you.
MR. ARMISTEAD: Disagree. 