← 返回

2026.10.07 | TB | Power | US Gas-Fired Power: CCGT Capacity & Costs

原标题：US Gas-fired Thermal Power – Capacity Growth Pressures & Capital Cost Outlook for New Combined-cycle Gas Turbine Plants

#power #energy

Third Bridge · 2026-10-07

Analyst-led • 1 Oct 2026

Companies: ENR DE, 8058 JP, NEE US, NRG US, AES US, TLN US, VST US, GEV US, CEG US

Scott Armando (SA) Independent consultant

Anthony DeRuijter (AD) — Third Bridge Sector Analyst

Agenda

Key Insights

Introduction

AD [00:00:04]: Let me start today's Third Bridge's Interview entitled US Gas-fired Thermal Power – Capacity Growth Pressures & Capital Cost Outlook for New Combined-cycle Gas Turbine Plants by introducing myself as the moderator. My name is Anthony DeRuijter. I'm very pleased to be joined today by Mr Scott Armando, Director of Capital Projects at Entergy. Scott, thanks much for joining me today.

Before we start today's Interview, could you please state specifically I agree or I disagree to the following statement, and the statement is: You understand the definition of material non-public information and agree not to disclose any such information, or any other information which is confidential, during this Interview.

SA [00:00:34]: I agree.

AD [00:00:36]: Would you mind briefly introducing yourself?

SA [00:00:41]: Yes. I'm Scott Armando. I am one of the directors for capital projects for Entergy. I have the projects in North Louisiana, which consists of building six combined-cycle gas-power units.

Questions

00:00:59] What is your outlook for new CCGT [combined-cycle gas turbine] capacity growth across the US, given the data centre hype and broader load growth expectations?

SA [00:01:20]: I would say that, obviously, data centres are driving a very restrictive market, and everybody wants to be the next data centre supplier of energy, but that is leading to a restriction in our ability to meet the need over time because, obviously, there's a finite amount of suppliers for everything, not just the turbine island and the main equipment, but every other piece part that we use to build these plants. That all said, I think the future is very bright, but we're going to hit a point where not everybody can go as fast as they want to go just because we've saturated a lot of markets and a lot of suppliers with the orders they currently have. It's going to be harder and harder and with longer lead times as we go forward just because so many people are trying to do this.

00:02:33] Do you think we have hit incentive pricing for new CCGTs? Which regions or ISOs [independent system operators] would make sense for building new CCGTs?

SA [00:02:58]: Obviously, you have to be in a regulatory environment that would allow for gas plant building, which in some areas, that's getting harder and harder and some other areas like Louisiana, Texas, it's a little bit easier to do so. Virginia is not. It's getting more restrictive and more rules. Obviously, California doesn't like building fossil fuel plants. Arizona is building fossil fuel plants, which can supply energy to California. There's a lot of other states around with land, with opportunity to build these power plants and the data centres have shown they're willing to go wherever the power can be had. Whether that's Utah, West Texas, North Louisiana, you name it, they're willing because they understand the need and the environment they need to be able to build a plant to begin with to power their plants.

00:04:20] Apart from where you might have hit incentive pricing levels, or at least have what you need to build a new CCGT, in terms of execution, are you more confident in certain regions or ISOs that a new CCGT plant build will be able to execute vs not? What would inform that opinion?

SA [00:04:49]: I think regardless, most companies that are big enough to do construction of a multibillion-dollar plant are national or global where they can bring in additional resources as necessary. Really, right now, between the materials shortages, but there's also a significant labour shortage to do all this work. Then on top of having skilled workers to construct a power plant, you're competing with the some of the same skills to build a data centre. You're even competing with the customer for the resources you need for construction. Yes, the labour pool is very limited. I know there's a lot of work being done, even the federal government is seeking to help increase the labour force through apprenticeship programmes as part of incentivised funding for certain projects. That's all well and good, but just like the supply chains have stretched out over 4-5 years, so has the ability to resource labour and get it up to speed where it's truly effective, it doesn't happen overnight. If you are in the labour force, you're going to make good money because you are going to stay very, very busy.

00:06:31] How vulnerable would you say active builds, primarily CCGTs, are to any lower-than-expected load growth down the line? Could you outline the development timeline and how much capital gets tied up in the initial 12-24-month stage? That could be potentially stranded capital if, 1-2 years after that gets dispersed or put down, we see material downgrades in load forecast or anything similar.

SA [00:07:10]: That one is a little bit trickier because if there's going to be a balance where regardless of whether, let's say, data centres stop being a thing, there's still a need for industrialisation, the need for electrification of industrial processes, whether that be refining or metal working and stuff like that. There is a significant opportunity for load growth beyond or outside of the AI data centres. It might be a bump in the road short term, but as we see, as the country, this is my opinion, as we're reindustrialising certain things in our country, they all take power. When we had the flat load growth over many years was due to the globalisation, overseeing jobs, sending them overseas. Now as we're looking at trying to revitalise our national manufacturing to some degree. Obviously, it's not going to be humongous, because that will take even longer time. There will be a need for power growth, CCT growth.

Will it be as hot as it is now? That remains to be seen. Even upfront, it is a significant investment.

00:09:57] My understanding is that the upfront CAPEX for CCGTs today might be roughly USD 2m-2.5m-plus per megawatt. How much of that would you expect to be in the initial 1-2 years of an active or new-build CCGT?

SA [00:10:42]: It depends on how mature the organisation is and how robust their supply chain is in that if you have all your ducks lined up, it can be significant in the first year just because you're paying out money for long lead items, you're paying out money for the engineering, you're paying out money to set all the contracts and such. It can be a significant investment right upfront in the first year out of the total price tag of the project because obviously, your big numbers are tied up in the power island equipment, and you've got to start paying for that, and it can be expensive. Particularly in a very competitive environment, the equipment suppliers are in the catbird seat in that they can demand pretty high upfront commitments to ensure that you get their equipment when you want it. It can be a significant portion.

AD [00:12:07]: What would be significant, in your experience? Are we talking 50% or more of the project price tag might be needed in the first 12 months, if you have your ducks lined up?

SA [00:12:19]: It probably wouldn't be quite that high, 25-30%, and it really comes down to how the contracts are all written. In a highly competitive environment, there might be incentivisation to pay more to jump ahead in the line to expedite all those things, which happens in a very tight market. Your costs will go up to stay ahead of everybody else. That usually means that the overall cost is going to go up as well. The percentage would probably stay roughly the 25-30%.

AD [00:13:04]: What about if it's the initial two years – does it then expand to over half the total project CAPEX, at that point, dispersed? If so, as an operator or developer, what's the reality of that? Does that mean the structures are all built? Does that mean all the long-lead items are secured? Does it mean all the permitting is done?

SA [00:13:30]: Yes. By the end of year two, you've got all your permits, you've got all your long lead equipment ordered and/or starting to deliver. You also have your underground work, your foundation work done, you're going vertical on the equipment. You're pretty deep into construction and you're about at the 50% mark, maybe a little bit more just because all of that other stuff that you laid out in the beginning now is starting to arrive and you're paying the full price for it, including taxes, which can be significant.

00:14:19] For the utilities and IPPs [independent power producers] that are taking on the new-build risk with CCGTs, would you say there's the ability to force contractors and EPC [engineering, procurement and construction] companies to bear the brunt of any cost overrun or delay impacts, especially if it comes down to issues with bearing more capital risk in the timeframe where there might be more uncertainty of whether the load growth to support the build is coming? To what extent can they push their contractors, if at all?

SA [00:14:58]: Actually, it's the very opposite. The contractors are in such high demand, if they don't like the contract, they'll pass on that one and move on to somebody else that has more favourable risk profile and/or payment profile, how much G&A they're going to get. Shifting risk and saying, "Oh, well, you're responsible for it." Once you agree and you make it come to terms for scope, schedule and cost, there will be some liquidated damages if they don't deliver on time because you have ready customers that have demands for the time. The risk is not going to… Your question says, "Oh, it's going to shift to the contractor not delivering." If it becomes too risky for them, they'll pass on your project and move on to the next one. It is a balancing act. Risk vs reward, if the risk is high, the reward has to be even higher for them because they can just move on and find another client and work with either less risk or for more money.

AD [00:16:33]: Interesting. Basically, in this environment, let's say you're an IPP and you're building a new gas-fired asset, potentially baseload, you're looking to get it contracted, so on and so forth. You've got a forecasted CAPEX per unit. Even with contingency, you blow past it. You're not then going to Kiewit or whoever else you might be working with and saying, "Hey, you need to take this on." Because of this environment and what you just described, you actually have to take that on yourself and work that into the overall budget?

SA [00:17:12]: Yes. Because if you tell, let's use Kiewit or Zachry or anybody and just say, "Well, you're responsible for all the cost overrun and your company." They're not going to eat USD 1bn. They're just not. They can't afford to nor would they need to or want to. You have to be very careful and the companies that are doing IPP, that's a significant risk. The profile has changed over the years. Before, if you didn't make it on time, there could be some sharing and there could be some pushback because there are not as many jobs. When across the industry, you're building hundreds of gas plants, the next job is just down the road and you can get good money regardless. It is going to be a challenge for whomever is going to be building the gas plant. It doesn't matter who it is, because supply lines are tight, construction labour force is tight. There's a significant amount of constraints, which results in risk, schedule risk, cost risk that not necessarily is the construction contractor going to take that on. You're going to have to do some serious calculations on how much that you're willing to bear in overruns and such.

00:18:58] There seem to be lots of risks with new builds, given what you described. How do we think about the economics of uprates and repowers by comparison, whether expanding an existing CCGT or other gas-fired asset, replacing turbines with newer models that may be more efficient or have better economics or doing full-on technology conversions, such as taking an old coal plant due for retirement and converting it to a gas unit?

SA [00:19:33]: I think everybody is doing all of the above to meet the demand, but obviously, if you have a turbine, that's great, or a generator because those are the really hard things to get in a quick order. If you already have a turbine generator and you're just changing the motive force from coal to gas, the conversion isn't easy, but it can be done, but you're not buying the big rotating equipment that a new plant needs. You're not necessarily competing in that particular part of the market, you're in a very specialised market on the uprates and changing from coal to gas, that kind of thing. It is lucrative and people are doing it because it's hard to get the big pieces of rotating equipment quickly, whereas most other things, uprates, depending on how it is and what it is, you can do much quicker usually and start reaping the return on that because electricity doesn't care how it's generated and nor does the market care dollars per kilowatt.

Uprates are a good way to go and conversions are a good way to go. Again, it also comes down to availability of capital, availability of labour, availability of parts because no matter what you're doing, anything to do with big electrical work and industrial work like that is constrained by the suppliers now because so many people are drawing on pipe manufacturers, valve manufacturers, electrical enclosure manufacturers, wire, wire can also, even think copper wire is becoming a scarce commodity because everybody is using it.

00:21:58] Focusing on coal-to-gas conversions, do you have any visibility into where those opportunities are in the US and how feasible they are? I've been told that coal-to-gas may face difficulties due to the associated capital equipment intensity and efficiency issues.

SA [00:22:28]: In discussions with folks that have done it and it's not easy, turning something that was designed to burn coal, the boiler system for that and converting it to gas is not a simple task because that's not how the system was designed to operate. It is technologically and equipment intensive to do so. With all the other regulations and other issues, people have and probably will be willing to shift from coal to gas just because of the regulatory environment and it comes with regards to coal. It's a very tough environment. Having done some coal ash management in my past, it was a lot of money to spend just to handle the residue from the coal, whereas gas, you're just handling the gases that come from burning it, a lot different. There is additional savings by conversion, but it isn't easy. People should not be fooled that, "Oh, yes, you just put in gas burners and everything is fine." That's not the case. It's very complex work.

00:24:17] Do you think the newer CCGTs that are expected to come in from now to the 2030s can compete better than the existing gas-fired assets, whether on a heat rate basis or something else? Why or why not?

SA [00:24:40]: A, they're more efficient because the equipment manufacturers have not just said, "Oh, I've designed this, and I'm never going to…" They've been working to optimise it for fuel use, optimise it for to reduce the negative outputs, the nitrosenol (ph 25.05) and CO2, which the newer gas turbines are about half of the discharges that you would have seen, even 10-15 years ago. They are more efficient. They burn cleaner. They produce power at similar rates, but usually a smaller package and with much better efficiency and air controls are a little easier to do, and that means less cost because you're not buying as many chemicals or water treatment, spending that time and money on that side of the house. The newer ones are better. There's no doubt about it. They're more efficient and produce less greenhouse gases.

00:26:06] What would you expect CAPEX on a per-megawatt basis to look like for a new CCGT plant, assuming that a project starts development today? How might that break out into key equipment or capital cost categories?

SA [00:26:32]: When you look at it, the metric is normally in USD 3,000 per kilowatt. That's what we're seeing in the market today. There's above and below that depending on contracts and stuff, but that's where it's going right now. That's a lot, and that can obviously lead to a significant amount of expense. A good chunk of that, obviously, is in the power island equipment and all the associated equipment that goes in. The labour cost is obviously going up because more profit for the construction companies, more profit for the equipment manufacturers. There is and has been and will be a constant arms race on cost just because any time you have a high demand market and you're a high-demand resource, you can demand higher prices. That's what we're seeing right now. It went up for about USD 2,000, and it's already up to about in the high-USD 2000, closer to USD 3,000. That's significant, a 50% increase over the last 5-6 years. Is that likely to slow down? I don't think so. How high will it go? It will go as high as people let it go, as long as there's demand like anything else. Once it gets too painful to build a combined cycle plant, you're talking it costs as much as a nuclear power plant. People may want to start looking at, "Wow, nuclear power, I don't pay as much for fuel. I don't pay as much for environmental controls and all this other stuff."

It may allow nuclear power, whether it be the AP1000, like Vogtle 3 and 4 or some other newer item may price itself into the market just because the gas prices are going to price themselves out of the market. That all said, that's a little bit of a nuclear pipe dream to a degree because if the labour market is that expensive, the price per kilowatt is going to go up on a new power plant as well because the labour is not going to be cheap on a more complex system than it is on a combined cycle plant. I don't see the numbers coming down anytime soon, but at some point, it's going to taper off because people's appetite to spend billions is going to taper at some point. I just don't know when, probably predictive between five and 10 years from now just because a lot can happen in 5-10 years.

00:29:57] Which components of the benchmark USD 3,000 per kilowatt for CCGTs today do you think are at the most risk of blowing out materially – perhaps by 10-20% – because of factors within or beyond one's control?

SA [00:30:42]: I think just based on what we saw in the change from USD 2,000-3,000 is everything, every component is going up 10%, 15%, 20% because of the overall demand. If you start adding all the pieces together, I don't think it's necessarily one thing because, obviously, the turbine manufacturers, they're going to look at, "Hey, how do I maximise my efficiency to sell two or three more turbines per year, of course, at prices that people want to pay or can afford to pay before it becomes prohibited." Same thing for the labour force. They're going to go, "Okay, I'm going to get as much as I can now while it's hot, and I can pick and choose the most profitable jobs." That raises the prices, labour goes up 20%. I can't foresee any one thing being a problem. I see everything being a problem, whether it's balance of plant, valves, pipes, everybody needs valves and pipes. Everybody needs wire, everybody needs a PDC. All that is in great demand.

If you're a PDC manufacturer and you raise your prices because you're sold out, you have no more capacity or you need to make enough profit so that you can build extra capacity, that's fine. Here's the thing. You're not going to go out of business if you let a customer go because you're already full of capacity. Same thing for the guy next to you that does PDCs. Everybody is busy. They are trying to add capacity. They are trying to add, but it's not the lower prices, it is to sell more at the higher price. I don't see any one thing getting out of control causing a bust. I see death by a thousand cuts because everything is going to be more expensive.

AD [00:33:06]: If you had to pick one thing to be the worst of those various feeders into that USD 3,000 per kilowatt, which would it be in terms of potential blowout risk? Why would you pick that one thing?

SA [00:33:24]: Labour because we've already seen it. Labour costs are going up faster than any other part of the cost just because it's a tight labour market, and it's hard. A guy can say, "Well, I don't have to travel anywhere because they're building one in my backyard, and I can get extra money. Even though I live here, they'll pay me per diem, they'll pay me living costs, just so I'm happy to work here." That's expensive. It has raised. That is the one area. Labour is the one area that we've seen over the last 2-3 years that have driven the price up significantly. Power island contracts are locked in. That's fairly easy. That hasn't gone through the roof five years from now when contracts aren't locked or that kind of thing, they could demand more. Right now, labour is a today thing because you need the worker out there turning the wrench, welding the pipe today, and it's a very good market to be a welder, a very good market to be an electrician, and they're getting paid and their wages are going up. That would be the one area that I would say labour has the biggest potential effect on the price because we've seen it already.

00:35:06] You mentioned electricians and welders. Are any specific roles the most in demand or shortest in supply? Are any of those mission-critical to a CCGT build – if you don't have enough of them, progress stops and you add X months to the build?

SA [00:35:25]: The two we just talked about, the welders, obviously, there is so much pipe that needs to be welded and it needs to be successfully welded with high quality because obviously, we're talking high-pressure, high-power systems. Having successful welds and having enough qualified welders to make those welds is vitally important. People say, "Well, robot welders." That takes welders to set them up to make sure they do what they're supposed to do. It can improve, but you're still going to need welders to set them up, but then there are things that robots can't do. They like long and straight. That's great, but not everything is long and straight. Some things are in tight areas where you can't mount a robot welder.

The other one is electricians. The reason why electricians are so important is our biggest competitor for electricians is the people next door doing the data centre build. They're putting in a whole bunch of electronics and electrical equipment, fans and cooling and HVAC, that's another area that's of concern. I would say welders and electricians get to write their own ticket these days. If they can't make money, then they're doing something wrong just because they are so in demand. Qualified good welders are hard to find and electricians, there's such a high demand there probably five jobs for every available electrician today. Just like I said, you can find a job anytime you want where you want pretty much if you're an electrician and/or a welder.

00:37:35] In terms of the new build environment, I'd assume labour is capitalised into the EPC component of the CAPEX. Is that the case? Ultimately, how much would labour and EPC-related components contribute to the USD 3,000 per kilowatt, on average?

SA [00:38:05]: It is the biggest driver of that increase to USD 3,000. Obviously, everything contributes to it. When it comes down to labour, EPC has got to pay their labour force, keep them happy to keep them on the job because if they don't pay an extra, let's say, USD 10 an hour and just down the street, somebody is paying… These are made up numbers. If that electrician is making USD 30 an hour working on a CCT, but he can make USD 40 an hour working on the data centre, guess where he's going to go. What does that mean? We're going to have a 25% increase in that guy's wage to keep them building the power plant that supplies the data centre. There are other factors, obviously, but one of the things that we do look at is how do you attract and maintain your labour force and how you go about that. There are things like per diems, other incentives.

Some people pay extra on top of per diem. Some people will pay if you stay here for so long, you get an incentive payment. There's a lot but that all costs money. That goes to the bottom line of the price because, obviously, labour is a big part of it and overheads, benefits, better benefits, things that people as workers deserve and need. They're not dumb. They're going to go where the money is, and when there's plenty of jobs, that means you got to put more money against it to get more workers. That's very simple economics.

AD [00:40:10]: I get that labour is the biggest risk factor or driver of any increase of that benchmark CAPEX per kilowatt past the industry average of USD 3,000 that you highlighted. What I'm curious about is, how much labour, assuming it's part of that USD 3,000, typically makes up. Is it 10%, 20% or 30%? Is it capitalised at all? How do we think about that?

SA [00:40:36]: It is capitalised, obviously, because the EPCs or the constructors put that into their pricing. Their prices are going up. Part of it is to cover the labour. What we're seeing is besides all the 20-25% increase in materials, you're seeing that and more in labour costs. It's probably neck and neck with materials. It obviously depends on the area. If there's a lot of work in an area and there's a lot of people doing things, then the prices for labour is going to go up and the percentage of how much the project cost goes up is directly dictated by bringing labour to the area. For some places where labour is scarce, it can be 50% of an increase. If labour is plentiful and everybody is paying the going rate and so prices aren't inflating as rapidly, then it might only be 25%. It's still going to go up, but it's not going to be as much. You're not incentivising them to come, not paying them extra bonuses or that kind of thing to incentivise them. On the labour side, that's still a significant number.

00:42:15] If the labour cost inflation is first borne by the EPC, how quickly can the contractor pass that cost to the IPP or utility that it's doing the project for? Is it constant? Is it in real time? Is there a lag? Is there some pushback?

SA [00:42:35]: It all depends on how the contract is written. If it's a fixed price contract, obviously, they baked it in. It's in there because they're not going to accept that risk for increased cost without having it factor into their fixed price cost. If it's time and materials, obviously, it's going to show up pretty quickly in the pricing to whomever is paying. It is a significant risk for whomever is going to pay it. They have to plan for it, whether it's IPP or EPC constructor, EPCM, whatever, you have to think ahead and say, "Hey, my project is going to be in the field for three years. What's the current escalation rate? Is it 5% or 10% per year?" If two million hours, let's use round numbers, that's a lot of money very quickly times 5% or 10% per year, that's a lot on those two million hours. That is why people are… Especially you have to really focus on what the labour cost is going to be because escalation can be quite significant and quite a significant portion of the cost.

00:44:13] How stressed could supply chains get for new CCGT builds – not only because we have all these new builds, but also because for existing plants, there's a chance that we might see more uptime pressure, whether from load growth or data centres that have specific uptime and reliability requirements?

SA [00:44:55]: There's significant pressure on any electrical supplier because of the fact that, obviously, data centres 99.9% uptime because if you drop them, it's not very good for them to have that happen. You have to have redundancy, you have to have ability to bring power from elsewhere to help support that. That's more of the grid stabilisation issue that if you're a utility, you really have to think about it, obviously. If you're an IPP is like, "Okay, what happens? Am I going to get penalised because my plant doesn't goes offline," because there could be penalties. If they were specifically designed to support a data centre or something like that, there may be damages that they will come after them for or not pay them for in their pricing. There is a lot of strain on the grid, on electrical suppliers because of all this sudden demand everywhere. If you look across the country, just about everywhere is getting a data centre of some size and is putting a strain on the available resources because if you have excess capacity, the Googles, the Metas, the AWSs are always going to try to find where that excess capacity is and try to tap into it there, why wouldn't they?

Then from there, they can build more capacity, but they have a ready source to meet immediate needs whereas having to wait for three years for a power plant to be built. It's putting a stress everywhere. It's on existing, on the future. It is a challenging time right now, but it's also very exciting. I don't know anybody thought 10 years ago that the power industry would have growth in near double digits. No one thought that back when it was less than 1% of load growth due to so many other things. That's pretty significant, but it is putting a strain on the overall grid because it has to grow as well, not just the generation side.

00:48:00] When you build a new CCGT, how much pressure or incremental component demand are you adding to the market? I appreciate this depends on its size and so on. How do we think about the volume of key components that you now need, such as compressors, transformers, replacement blades for the turbine and electrical equipment?

SA [00:48:51]: I can't qualify that because obviously, we're buying a lot of stuff and the market is very big, depending on whether you're talking the local market, national market, international market, most of the turbines are coming in from other places. The world is doing the same thing. One CCT is probably a small percentage of the overall global market, but once you start stacking up 100 or 200 of them in the global market, it's a strength, and that's where we're at. Okay, 200 CCTs in any given year across the globe, there you go. One would be a percentage of that. That's how you would have to do it because you'd have to look at the world market because most things are sourced globally, whether it be steel or electrical wire or pumps, motors, breakers, you name it, they're globally sourced. You'd have to look at the global demand and then say, "Okay, how many power plants are being built in any given year," and that would be your percentage of the global market that one plant would do. I don't know the overall global numbers.

AD [00:48:51]: It was more so just anecdotally trying to understand how much component a single new CCGT plant would add in onetime or recurring procurement needs and then how that would compound on what we already have with the existing CCGT fleet. It's very complex.

00:50:43] What are the most stressed factors across key equipment categories and any other important resources, apart from labour? In which categories are there lead time or cost inflation issues?

SA [00:51:35]: Like I said, everything is under breakers, power distribution centres, piping, steel, just wire, you name something that we put in valves, control systems, microchips, everything has an effect because today's day and age, we use all those things that most people use in any industrial process, a valve is a valve is a valve. Now we're ordering hundreds of them to support building all the power plants. While other people are ordering them to build a new steel plant or some other manufacturing or LNG plant. All these commodities people don't think much about it, but now for us, when you look around, it's like, "Well, how can I get this pipe faster?" You can't. They're already at capacity. Every piece of pipe that comes out is immediately spoken for and is going to be fabricated into somebody's something. You have to plan ahead and lock up runs when you can and be ready to spend money even if you don't necessarily need it yet because you don't want to pass up the opportunity to get steel or get pipe or get breakers, you name it. You're doing some strategic purchasing to get out ahead of the market because you know it's going to take four years to get some of the stuff. What used to take a couple of weeks to get now it takes a year.

You're going to have to plan ahead and look at, and if one of your supplier comes to you and said, "Hey, I've got some space or I got some availability," you're probably going to jump on it because it's not that big a risk because even if you end up not using it, you could probably sell it to another user very easily. That's the best way I can put it is you name it, if it's a component, a commodity, that everybody is doing it. Everybody needs it and you better plan ahead and get out ahead of it because otherwise, you could incur delays, as we talked about earlier, because you're waiting on a spool of pipe, or a PDC, power distribution centre or a HRSG or any component thereof. You can have the whole big HRSG, but you're missing a spool piece so you can't run the plant. That's a sad place to be. You want to plan ahead for sure.

00:55:14] You indicated pre-Interview that you've worked primarily with Mitsubishi Heavy turbines. The GE Vernova and Siemens large-frames make the headlines for lengthy backlogs, ballooning costs, etc. How would you compare your experience with the Mitsubishi Heavy products in terms of costs and lead times?

SA [00:55:41]: We are unique, and that's all I can say about that. Unlike others who are getting in the market and trying to get on a waiting list for the next one, we are on the list. All we have to do is say, I want the next one. That's a big difference. That's about all I can say on that.

00:56:24] A lot of utilities and IPPs have announced some partnerships or consortiums with OEMs such as GE Vernova, Siemens and Mitsubishi Heavy or E&C [engineering and construction] companies such as Kiewit. Do you think there is an inherent advantage with these supplier relationships, and how does that materialise, whether through lead time minimisation or preferential discounts for key capital equipment?

SA [00:57:03]: Consortium, yes.

AD [00:58:35]: Would you say that's applicable more broadly? When an IPP or utility has some such a partnership or consortium, would you expect that they at least have their slots fixed and will get products within the average wait times, or at least within whatever the OEM has advised?

SA [00:59:02]: I can't speak for other utilities or other construction companies. If others have it now, they're coming in in a much higher demand situation. It's probably not as calm or as easy as we have it because if Siemens or GE has 10 customers and saying, "I want 10." They'll say, "Well, this year, I am going to produce five." You know what I mean? Then it becomes a very competitive market for who gets those five slots because there's 10-15 people lined up for those five slots and then the next five slots and the next five slots. If you're coming later to the party, all the slots for the next three years are already taken. You might not be able to get one until year four from now or year five. That is not a comfortable position you want to be in because you can't… Unless you got that thing locked down, it's really hard to build a power plant without the power island. That's a real risk.

If you're like, "Well, I'm semi promised, what happens if something happens and it shifts or someone comes in with more money ahead of you?" No one's getting ahead of us. You want that relationship with the supplier and to build that relationship with the supplier, but it's going to be harder now to say, "Hey, I want 25 generators over the next 10 years or next five years," when they have 10 people asking for the same thing because they know their capacity is only X. That's going to be a tough situation for anybody coming late to the game.

AD [01:01:52]: That's very helpful. With that, I will close this out for today. Thanks so much for joining me and sharing your insights into the gas-fired power space.

SA [01:02:03]: Hopefully, I answered the question so people can understand. I appreciate the time. I really do.

AD [01:02:11]: Absolutely. Thanks so much. Have a great rest of your day, and thanks for your insights.

Transcription ends at 01:02:21 of the recorded material