Talen Energy Corp Q3 FY2024 Earnings Call

Ladies and gentlemen, thank you for standing by. Welcome to Talen Energy Corporation Third Quarter 2024 Earnings Call. At this time, all participants are in a listen-only mode. After the speaker's presentation, there will be a question-and-answer session. Please follow the operator's instructions. Please be advised that today's conference is being recorded. I would like now to turn the conference over to Ellen Liu, Senior Director, Investor Relations. Please go ahead.

Thanks, Michelle. Welcome to Talen Energy's third quarter 2024 conference call. Participating on today's call are Chief Executive Officer, Mac McFarland; and Chief Financial Officer, Terry Nutt. They are joined by other Talen senior executives to address questions during the second part of today's call as necessary. We issued our earnings release this morning along with the presentation, all of which can be found in the Investor Relations section of Talen's website, www.talenenergy.com. Today, we are making some forward-looking statements based on current expectations and assumptions. Actual results could differ due to risk factors and other considerations described in our financial disclosures and other SEC filings. Today's discussion also includes references to certain non-GAAP financial measures. We have provided information reconciling our non-GAAP measures to the most directly comparable GAAP measures in our earnings release and the appendix of our presentation. With that, I will now turn the call over to Mac.

Well, thank you, Ellen. Good morning, everyone, and thanks for joining us today. Before we get into our quarterly results, I would like to start with a few brief remarks. Market and regulatory events over the last few months have further underscored how critical existing generation is to serving demand growth and supporting grid reliability. Since we signed the Amazon deal in March of this year, the market fundamentals for power in the United States have only become more constructive for independent power producers. The higher PJM capacity auction results in July, the Microsoft Clean Energy announcement and increasing utility load forecasts, mostly driven by data centers, the reshoring of industry and the electrification of our economy, all support this thesis. The U.S. is expected to be the fastest growing market for data centers, growing from 25 gigawatts of demand in 2024 to more than 80 gigawatts by 2030. Meeting this demand will require significantly more generation. Speed to market and access to long-term power have become top priorities for hyperscalers and data center customers. As I said before, serving this massive data center demand will require an all-of-the-above approach. This includes co-location like our arrangement with AWS, hybrid arrangements that co-locate primary power behind the meter while using the grid for backup, and front-of-the-meter connections to utility transmission. While we are disappointed in FERC's first decision to reject the ISA amendment, it does not change the fact that this load growth is coming and it will not stop our progress. First, development of the data center campus will continue under the existing 300 megawatt ISA as we and AWS work together on the path forward. Talen's colocation arrangement with AWS is part of the solution to issues raised at the FERC technical conference on large co-located load. It brings service to customers quickly and without expensive transmission upgrades that would impact a retail consumer's energy bill. That said, we are exploring the whole suite of commercial and legal solutions to facilitate full development of the Susquehanna campus as well as progressing other opportunities across our fleet. This includes filing a motion for FERC rehearing in parallel with AWS contract discussions. We are keeping all of our options open when it comes to colocation, front-of-the-meter or the hybrid solution I discussed earlier. I know the first questions in our Q&A today will be: when are you going to have this solved? What will it look like? And what about your next data deal? The short answer to all of this is, we'll let you know when we're done. Quite frankly, this reminds me of a year ago when we hosted our Susquehanna site day visit and data set at our data center campus and talked about colocation as a novel concept. Many of you asked these same questions then, and you heard me give the same answer. You all know that we don't do commercial negotiations in public and we're not going to comment on them at this time. That said, I believe we can leverage our transaction experience, advantaged grid location and strong stakeholder relationships to utilize all of the options on the table. I am confident that we as an industry can meet the challenges in front of us and seize the opportunity to power the AI economy and bring its substantial economic benefits to Pennsylvania specifically and the U.S. more broadly. So now turning to our key highlights, starting with Slide 3. Talen has had an active third quarter and I'd like to highlight several of our achievements, starting with our solid operational and financial performance. In the third quarter, we generated $230 million of adjusted EBITDA and $97 million of adjusted free cash flow. Based on our strong performance year-to-date, we are raising and narrowing our guidance for 2024 and we are also affirming the 2025 financial guidance announced at our Investor Day in September. Terry will provide more details on that as well as our 2026 outlook. In October, we acquired TeraWulf’s 25% share in Nautilus, which provides strategic flexibility with the building and its power use. Operational activities at Nautilus have been suspended, which releases 150 megawatts of power to be sold at more profitable levels to the PJM wholesale market and eventually to Amazon. Lastly, we were added to five equity indices over the last few months driving passive fund demand for our stock and continued shareholder rotation. I'm proud of what the team has accomplished this quarter while setting the stage for more long-term value creation. Turning to Slide 4, you've heard me talk about the ISA and path forward. We also participated in the FERC technical conference on November 1st. There was a lot of good discussion and we applaud the commission for taking up this serious matter. We continue to believe the path forward is that all solutions should be on the table as long as the RTO, the generator, the transmission operator and the respective state PUC are on board. Turning to the PJM capacity auction. PJM has requested and FERC has approved the six-month delay of the ’26/’27 auction that was originally scheduled for next month due to complaints filed by the Sierra Club and other NGOs. Subsequent auctions will occur every six months through the ’28/’29 auction in December of 2026. PJM is focused on reevaluating the auction reference technology, which impacts the steepness of the supply curve as well as the treatment of RMR, or reliability must-run units. We are generally supportive of PJM taking another look at the supply curve. However, we think PJM should compress the time between auctions to get them back on track sooner to the original timeframe of three years in advance. Further delays in the capacity market create uncertainty in the very market that needs signals to incentivize new build. We also believe that requiring RMR units to either bid into the capacity market as price takers or be accounted for as phantom supply will distort price signals. RMR units are meant to support transmission reliability, not to serve as a capacity resource. Removing RMR resources from the capacity market is appropriate to send the proper signals that new generation is needed. We encourage PJM to resolve these issues as quickly as possible and look forward to engaging constructively with them on the process. On that note, let's turn to Slide 5 and put some numbers behind the supply-demand situation in PJM. Since 2020, 19 gigawatts of generation assets have retired in PJM and nearly all of those are gas and coal plants. While only 10 gigawatts of new gas plants have come online, along with 13 gigawatts of renewables and batteries—which do not provide the same dispatchability as gas plants and coal plants—from a demand perspective, PJM recently reported significant increases in submitted requests to the 2025 power demand forecast for anticipated large loads like data centers and manufacturing. These requests include 15 gigawatts of demand by 2026 and over 50 gigawatts by 2030. PPL itself forecast that by 2030 large loads could double the peak summer demand curve in its control area and that is where most of our plants are located. These data points highlight how PJM needs more reliable dispatchable generation to meet the power demand growth that is coming. Moving on to Slide 6, let's look at our year-to-date operational and financial results. Our team continued to deliver from an operational perspective. Our fleet ran well, generating over 27 terawatt hours with an equivalent forced outage factor of only 2.4%, which is an improvement from 3.5% in the same period last year. Roughly half of that generation came from our carbon-free Susquehanna nuclear facility. Importantly, our team works safely during the busy summer months. We have a year-to-date TRIR of only 0.3, which is truly remarkable. This is in line with or better than our peers and we continue to emphasize safety as our first priority across the fleet. We leveraged our strong operational foundation and commercial strategy to deliver significant adjusted EBITDA and adjusted free cash flow on a year-to-date basis. We continue to prioritize capital returns and balance sheet discipline during the quarter. Terry will take you through the year-to-date numbers, our leverage and our liquidity later in the presentation. I'd like to stop and take this opportunity to recognize and thank our employees across the company who have worked safely to deliver impressive operational results across the entire portfolio. These team members are key to our overall performance as they operate, maintain and improve our generation fleet and other assets. Without their hard work and commitment to excellence, none of this is possible. I'll now turn the call over to Terry. Terry?

Thank you, Mac, and good morning, everyone. Now turning to financial results. For the third quarter of 2024, Talen reported adjusted EBITDA of $230 million and adjusted free cash flow of $97 million. Compared to the same period last year, expanded spark spreads and higher power demand drove increased generation margin across our fleet. Generation margin, along with the combined impacts of our hedging strategy and the PTC, more than offset the absence of earnings from our ERCOT generation portfolio, which was sold in March of 2024. Q3 2024 adjusted free cash flow included the impact of a $40 million higher pension plan contribution, reflecting our continued commitment to our workforce and retirees. Additionally, we accelerated some raw uranium purchases during the quarter to take advantage of pricing opportunities. These collectively resulted in lower adjusted free cash flow compared to the third quarter of 2023. For the year-to-date period, adjusted EBITDA was $606 million and adjusted free cash flow was $262 million. Moving now to guidance on Slide 8. With three quarters of performance behind us, we are raising and narrowing our 2024 adjusted EBITDA and adjusted free cash flow ranges. Our new adjusted EBITDA range is $750 million to $780 million. And our new adjusted free cash flow range is $265 million to $285 million. Looking ahead to 2025, we are reaffirming the guidance ranges we announced at our Investor Day in September. Our adjusted EBITDA range remains at $925 million to $1.175 billion and our adjusted free cash flow range is still $395 million to $595 million. Additionally, our 2026 outlook also remains unchanged from what we disclosed at our Investor Day. These ranges continue to demonstrate Talen's significant earnings and cash flow growth profile, which includes nearly tripling adjusted free cash flow per share by 2026. Turning to Slide 9, we remain committed to maintaining net leverage below our target of 3.5x, along with maintaining ample liquidity. As of November 8, our forecasted net debt to 2024 EBITDA ratio was only 2.1x, well below our target. In addition, we have nearly $1.3 billion of liquidity, including over $550 million of unrestricted cash on the balance sheet. We continue to engage with the rating agencies, two of which have responded to our balance sheet discipline by upgrading our credit ratings. In September, our S&P corporate credit rating was upgraded to BB- and in October, our Moody's rating was upgraded to Ba3. We remain focused on unlocking value and returning capital to shareholders. In September, we announced another upsizing of our share repurchase program and have over $1.2 billion of capacity remaining through year-end 2026. We have returned approximately $950 million to shareholders year-to-date by repurchasing roughly 8.3 million shares. Due to the timing of our Investor Day and subsequent non-deal roadshow, there was limited time to repurchase shares at the end of this quarter. That said, we continue to see share repurchases as the best use of our capital and continue targeting a return of 70% of adjusted free cash flow to shareholders. Turning to the next slide. After uplisting to the NASDAQ, Talen has become eligible to join several equity indices, which has driven substantial institutional and passive stock demand. Talen has been added to five indices resulting in passive investment funds acquiring over 6 million shares on September 11. Earlier this month, Talen was added to the MSCI USA Small Cap Index, which will be effective on November 25th, and we anticipate additional passive fund demand from that inclusion. Further, Talen could qualify for additional inclusion for sector-specific indices, further enhancing stock demand and accelerating the natural shareholder rotation. With that, I'll hand the discussion back to Mac.

Great. Thanks, Terry. As all of you have heard me say before, Talen remains an IPP that's focused on being an IPP. And that's at a time when reliable and flexible generation assets are more valuable than they've been in many years. We appreciate your interest in Talen for joining us on today’s call. We will now open the line for questions and I’ll turn it back to the operator. Michelle?

Please follow the operator's instructions. And the first question will come from Shar Pourreza with Guggenheim.

Morning. Mac, just by your prepared remarks, it seems like there's not a lot of appetite to answer my AWS strategy question. So let me just shift gears towards resource adequacy, which is obviously very topical at EEI. Are you involved, I guess, in state level conversations in Pennsylvania at this point? And if the PAPUC would eventually run an RFP and look for peaking PPAs or the state passes a test-like mechanism, would Talen be involved in new development? I guess, how do you see this sort of unfolding the session overall? Thanks.

Sure. Well, obviously, that's a broad policy question because I understand that there are a number of people who are talking about either running an RFP or rate-basing generation. I think that there's a couple of things that are going on. One, first of all, for the first time in seven years, we had a capacity clear that was higher than $100 a megawatt day. And I think we have a fair amount of reactive voices going on here with respect to how to solve the resource adequacy problem. I think first and foremost, the solution to that problem is, as I said in my remarks, to get the capacity market on track so that we're providing pricing those three years out. With respect to the TEF that has been floated, we have been in conversations in Harrisburg. Look, we think that something like a TEF that provides low-cost loans to generation can help the resource adequacy issue that we see on the horizon. We're supportive of it, but generally it depends on how those loans are structured, what strings are attached to them, etc. So when you ask that question, Shar, there's not a lot of specifics out there right now. There are people talking about should we rate-base generation, should we do something like the TEF—a Pennsylvania Energy Fund? Should we think about other ways to solve resource adequacy? I think what's getting lost in all of this with respect to Pennsylvania is that Pennsylvania has excess reserve margins in excess of 30%. It's other states that have lower reserve margins that are somewhat of the issue. That said, Pennsylvania has an opportunity given that it has local gas, plentiful gas, and a pro-business stance regardless of political persuasion. So they have the ability to export energy. I think that they can continue to do that and I think capacity markets are the way to do that. But I'll open it to Terry, so if you have any further comments, Terry.

Yeah. Thanks, Mac. A couple of things to add to that response, Shar. First of all, the Pennsylvania PUC is holding a conference on resource adequacy here right before Thanksgiving and we plan on participating in that. So obviously, being a large generator in the PJM market, we definitely are engaged in those discussions and we'll continue to be engaged. Back to the general policy question, the other thing I would add to Mac's comments is I think a Pennsylvania Energy Fund is an interesting concept. We've seen that done in Texas. I think whether it's a Pennsylvania Energy Fund or whether it's some of these discussions around RFPs, the devil is always in the details. What does that asset look like if you put in a new asset, how does it participate in a capacity market? What economics are borne by the asset itself and what can it do and what can it not do from a participation standpoint. So I think they're constructive discussions, we'll be engaged and look forward to helping solve the resource adequacy issue.

And Shar, just one more comment on that real quick, which is as we talk about our sites which are in the PPL zone as advantaged sites, we have a lot of sites that have access to gas and are in the right point on the transmission system to interconnect and we're looking to see how we can leverage those sites as redevelopment opportunities. Again, it's going to have the right kind of economics and the right kind of returns, but we are exploring it there. And if you want to ask your AWS question, feel free.

Sorry, I could pass that to someone else. But just my only question is as legislation starts to form, are there discussions you're having with the wires companies or is the bid-ask kind of too wide right now to even come to a discussion table?

I think that the things that are being discussed right now are so preliminary. It's not like there's a bid-ask that's wide or tight. It's just that everything is so preliminary. That's why I said that these have been somewhat reactive to this capacity clear again. Quite frankly, I think that we're not even close to what it's going to take in the capacity clears to incentivize new generation, and people say we should rate-base it or go through an RFP process. Let's be clear. The reason why PJM is the most effective deregulated wholesale market in the country is because people didn't want to pay for stranded assets. They wanted to move to a deregulated market in PJM because they felt it would drive the lowest cost of supply possible. We continue to believe that and we will participate in that.

Got it. Perfect. I'll touch base with you guys in a little bit. I appreciate it. I'll pass it to someone else. Thanks, Mac.

Please follow the operator's instructions. And our next question comes from Jeremy Tonet with JPMorgan.

Just wanted to come back to, I guess, the technical conference here, if I could. Looking at the proceeding there as a read-through for the amended ISA, how do you think about that or just really any thoughts on data center development more broadly coming out of the conference here? Just wondering what your take was on the technical conference?

So I think in general—and Cole is on the line who was a witness during the third panel and he can jump in here—there was a lot of discussion that was beneficial to thinking about how to solve the growing demand from data centers. I do believe that ISAs got conflated with resource adequacy and that the distinction between so-called front-of-the-meter and behind-the-meter, collocated, etc., sort of worked its way into that conversation, and those issues should be parsed separately. We believe the outcome should be—as I said in the opening remarks—that if the RTO, in this case PJM, the transmission operator, in this case PPL as it relates to our specific ISA, and the generator are all in agreement, as is the state-level PUC in this instance, this is a state-level issue and if those four parties are in agreement and they don't see incremental costs—which is exactly what was found in the case of the ISA and it created reliability—we think that's a model that should be approved. Whatever fits and where those four parties agree. Now, we may have colleagues that have a different view of that, but that's what should come out of this as we go back. There are PJM guidelines and other things that need to be addressed. You started the question with amending the ISA. We're looking at that right now. I think the current path on a regulatory front—put aside the commercial, which is where we're focused for a moment—on the regulatory path is first to file a motion for rehearing and that's the step that we're taking on that path. Cole, do you want to talk about what we're doing on the current AWS contract for a bit?

Yeah, sure. Hi, everyone. I think first on the question, as Mac said, co-location got kind of lost between the forest and the trees during that conference and resource adequacy conflated with co-location. I think it's going to get sorted out. It's going to take a little bit of time here, but we're optimistic that FERC will start to set a path forward that gives clarity on co-location in particular. On our commercial paths forward for our existing agreement with Amazon, I think it's important to reiterate what Mac said in the opening remarks that we have an existing ISA that's been approved for 300 megawatts, which enables us to have runway here to optimize getting to the 960 megawatts for this deal over time. We are evaluating all of our options jointly with AWS on how to accelerate that. Folks probably saw that AWS recently publicly reiterated a commitment to the site. The fast-forward options are a wide range as Mac also outlined in the opening remarks from kind of the status quo configuration all the way to a full grid connection. There's a bunch of shades of gray in between. We're looking at each of those options and weighing those carefully. Some of those configurations may have technical and engineering adjustments that are going to require some review and analysis. So we're in process to get to an optimal answer here jointly between all the parties, and we'll provide an update when we have one.

I'll just add on to it. When we signed the original 960-megawatt deal with AWS earlier this year, it's not as though we then started resting on our laurels. We've been looking at a number of different commercial arrangements and that is what allows us to focus on how we might pivot or look at a different way to get to the 960. We're actively doing that. We're also going to preserve our optionality with respect to colocation because we think colocation is one of the ways to speed the market and to power the AI economy and should not be lost. So we're going to continue on that front, but we're focused on the commercial aspects of things.

Got it. Very helpful there. Thank you. And then just pivoting here, data centers are looking for firm sources of power and there's only so much nuclear out there. So just wondering if you could provide any updated thoughts on the appetite for gas to service this demand, be it behind the meter or what have you. Any updated thoughts on that side?

Look, I think there's a couple things at play here. Obviously, many of the hyperscalers and those that use cloud services have aspirations for carbon neutrality at some point in the future. Weighing that, people are looking at the carbon-free aspect of nuclear because it's baseload and it fits the load side of the equation—in this case, data centers, which are effectively 24/7 load. The use of renewables is also part of their strategy. But in the near term, over the next decade, until there are SMRs that can be put in place in meaningful volume in the mid-2030s, that gap, if you believe the load is coming, will have to be served by gas units. If you look at PJM, PJM has a real opportunity to serve that load because of where it's geographically located relative to data centers. We think we're in an advantaged spot because Pennsylvania has abundant gas, a pro-business stance, and the ability to support additional electrons on the grid via new gas units. I think the appetite will develop because the need is there. I am very pro-nuclear and think that for energy independence and decarbonization, more nuclear plus gas as a transition fuel is desirable. But it's going to take time to add nuclear capacity, and so to fill that gap you'll likely see gas units. We're looking at what opportunities we have to play to help fill that gap.

Got it. Very helpful there. And if I could just quickly pivot to the PJM auction, any expectations for changes there or any reforms that you would be supportive of?

I think as I said in the opening remarks, we don't want very binary outcomes or extreme volatility in capacity prices. That volatility can be politically challenging and can reduce confidence in the market. We are supportive of PJM examining and adjusting the supply curve given the delay, but the devil is in the details. We have been clear that RMR units should not distort the capacity and energy markets. They're needed for transmission reliability but should not be treated as a normal capacity resource. Those are two very important issues as we go into this auction. Terry?

Maybe to add to Mac's comments. The other variable we're paying attention to is the demand forecast. PJM runs various subcommittees and their load analysis subcommittee met a few weeks ago. Some of the demand forecasts coming out of that committee are interesting. In our slides, we alluded to these forecasts; if you take a look, there is tightness in the market, especially as we move through the next few years. So that is probably the biggest variable we'll have to see how it comes out. On RMR units, as Mac said, putting an RMR unit in the supply stack or netting it out of demand can distort signals. If an RMR unit retires at the end of its period, you just see that drop off. We're obviously opposed to that distortion. We'll stay engaged and see where PJM comes out on the parameters and then look forward to the auction next year.

Please follow the operator's instructions. And the next question comes from Angie Storozynski with Seaport.

Thank you. So just on the timing of the resolution for Susquehanna, I know that I'm one of the very first to say that I want an answer now, but I'm actually having some second thoughts here having spoken to your peers. They're making interesting points. They're basically saying co-locations are still the most viable option from a speed-to-power perspective. We have a new administration that is very supportive of economic growth in AI in the U.S. We may have some meaningful changes at FERC come next year, and you have 300 megawatts to deploy and seemingly all options available given, as you said, the transmission availability. So again, why rush on any decisions here with changes to the existing contract? Why not just take some time to actually assess the backdrop that is about to change or potentially changing for AI in this country?

Angie, I appreciate the question. When we say we're looking at commercial arrangements with AWS, we're not taking anything off the table. There are many possible options, and colocation is one of them; we're continuing to keep that option open while we file the motion for rehearing. I agree with some of the points you laid out, but the time element is meaningful. As we go through our options and discussions with our counterparty, we consider a lot of factors including economics. We will take our time and be reasoned about the amount of time we take, just as I hope we're viewed as reasonable in our commercial activities. There's no easy answer, but we are being deliberate. Cole?

Yes, I would reiterate that we're not rushing, Angie, and we are being methodical in analyzing all of our options. We do want to be responsive to our customer who does need clarity at some point to accelerate and make large investments. That is part of the analysis. But as Mac said, we're going to look at the economics and other pros and cons of each option, and we'll let folks know when we have an answer.

Okay. And then the other thing is there are already questions about your gas assets. You have former coal plants converted to gas that used to be baseload assets that ran 80% to 85% of the time. One could argue that if these assets were to stay on the grid, you can't necessarily ramp the output from these assets until market prices rise. We're having all of these discussions about additionality and how, for example, extensions of operating licenses of nuclear plants relate to additionality, but how about the fact that you could increase the output from some of your gas plants? For example, if there were to be supported by an above-market contract, you could increase output significantly. So can you comment on the potential to increase dispatch from these gas plants or repower opportunities?

Yes, Angie, good point. Whether it's behind the meter or front-of-the-meter, the load is coming. These assets have been used as peaking assets for the most part over the last several years. Now they're receiving higher capacity revenues, but the energy margin on them is relatively flat year-over-year. We do have opportunities to run them more. For example, at Montour, a 105 heat-rate unit, we have ample gas supply and we ran a gas lateral there. We've made the conversion. From an energy standpoint, it can provide a lot more energy. It's just there's no additionality at the peak, but there's a lot in the other hours that it doesn't run. When it's running 10% to 15% of the time and we start to see uptick, we saw strong loads in the second quarter and Montour is starting to run more. So you will get increased dispatch even if these don't have a long-term contract and increased spark spreads should manifest themselves over time as demand shows up. There's also opportunity at many of our sites because of interconnection locations for repowering. Converting Montour to gas was a repowering opportunity. We're looking at all opportunities to increase megawatts or get incremental megawatts out of assets. When people ask about resource adequacy and what we are doing to solve it, the first thing is how do we run these units more, get more megawatts out of them. That will be a first part that contributes to resource adequacy. It's not a 1,000-megawatt CCGT announcement, but when you do this across our fleet and others, you start to fill some of that need over time, and then you'll get to new builds. So those gas assets are in a prime position for increased energy margin or to be a solution to power the AI economy as well.

Angie, to add to Mac's comments: in Q2 and Q3 this year our gas fleet did run appreciably more. During periods of tightness and warm weather in PJM, those assets dispatched more this year than in prior years. On a weather-adjusted basis, PJM load was higher this year than the prior year. Usually, the first reactions to resource adequacy are to dispatch the existing fleet more; the second is to consider assets that had notices of suspension or retirement to see if they can be unretired; and the third is conversions of assets previously one technology to another, similar to what we did at Montour. Those are the quicker, more economical responses. A new build, like a utility-scale combined cycle plant, takes five to six years. So you'll need these incremental steps in the interim. Given the additional dispatch we've seen, we've added incremental maintenance cost into our numbers that were included at Investor Day to ensure those assets are hardened and ready to go over the next several years.

Please follow the operator's instructions. And the next question will come from Nicholas Campanella with Barclays.

So you just brought it up at the end of your remarks there, but you brought the hedges down. I just want to confirm that's not a change in view of freeing up additional open capacity or energy for longer-term growth. Can you confirm?

Maybe just to clarify, that was not an action by us to take hedges off. It's just the denominator in that calculation is expected generation and expected generation rose. And because of slight increases in spark spreads, particularly in periods when the gas units are more likely to run, around summer and winter, when you increase expected generation, if you keep the numerator the same, it just drives the percentage down. So we didn't take hedges off as a strategic action. Given we're moving to more contracted revenues with the PTC as downside protection, we feel more comfortable being a little more open, particularly as we look at demand coming. But all that means is we haven't layered on additional hedges and there has been increased generation forecasted.

That's great context. And then just on the current RMR processes at FERC, I know you're in settlement discussions, but can you reconfirm your timing there? Do you still expect that to be solved by year-end? And is there any impact from the wider RMR inclusion-in-capacity-auction discussion that might delay your ability to get to the settlement? Timing is of the essence on that.

I'm going to ask John Wander, our General Counsel, to provide context around the RMR. But let me start with this: even with the delay of the capacity auction and the overlay of the RMR discussions, regardless of the intersection there, we've got plants with employees, with fuel contracts, with maintenance needs and outages and things that you cannot make a decision on one day before a scheduled shutdown. We have to make many decisions that take time to impact fuel contracts and staffing. We have employed all constituents to come to the table and to find a solution to RMR this year. That was our stated goal and time frame because we need time to prepare. John, do you want to provide an update on the process?

Sure. Nick, thanks for your question. What we're focusing on in the immediate moment is the settlement negotiations. There are settlement conferences set both at the end of November and early December, and we've made it very clear throughout the process that our objective is to be done by the end of December with the knowledge of what we can reach an agreement on. Whether that's an agreement that has a FERC order behind it or not, we won't know that probably by the end of December, but we'll have an idea by then what the parameters of a settlement will look like. We made that very clear to everybody.

Please follow the operator's instructions. And our next question will come from Michael Sullivan with Wolfe.

I'm going to pick up where Nick left off and drill down a little further. How can you settle in front of what FERC or PJM may come back with in early December that could impact how these units are treated? Is there enough time to react and turn it around by year-end?

Michael, I'm not sure I fully understood the timing in your question. We're targeting by the end of this period to have a settlement. Early next year we'll prepare for implementation. We stated that the plant was scheduled to shut down on May 31, so we need to have decisions well ahead of that to prepare. We're asking everyone to come to the table and resolve this now. But John, if you want to add on timing with regard to PJM filings, go ahead.

I can add: I think the two issues are relatively distinct. The inclusion in the capacity market of RMR plants is about the capacity of those plants, not about what we get paid under an RMR agreement. Whether PJM includes RMR capacity in the supply side and offsets payments, we'll see how that shakes out. From our perspective, whether we can get to an RMR agreement that satisfies us on what's required to run those plants in the next several years is distinct from whether PJM includes the capacity for those plants in the supply side in the auction.

Okay, that's very clear. And then when you say all options are on the table working with Amazon, to clarify: when you say the 960, could you be discussing something potentially even larger or involving other plants in your fleet?

I'm not going to talk about specific commercial negotiations with AWS. When we say all options are on the table, that means a variety of commercial arrangements including grid-as-backup, grid-as-primary, co-location, hybrid solutions, and potentially leveraging other parts of our fleet. We have been exploring different commercial arrangements that could allow us to achieve scale in different ways. But we won't discuss specific ongoing negotiations with a counterparty.

Yes, that's right.

And the next question will come from Anton with Jefferies.

Congrats on the quarter. Can we talk a little about speed to market, since that's a main factor for hyperscalers? Can you give a sense of how long it would take for, say, a Susquehanna-related deal to go through the process? If you agree on terms today, how long until the contracts start contributing, going through PPL and PJM? Is there anything you can recycle from the 960-megawatt ISA and how would that change if the deal were larger than the combined 960?

That's getting too specific about Susquehanna and the AWS path, so I'm not going to provide timing for that. But to pan out a bit and discuss front-of-the-meter solutions: front-of-the-meter typically refers to a retail contract or a virtual PPA or PPA, similar to what was done with Microsoft and that development. Regarding speed to market, our view is that PPL's zone has attributes that can enable faster connections, whether it's colocated with grid backup or front-of-the-meter, because PPL has transmission capability and has identified upgrades to serve anticipated large projects. Because we have sites and assets in the PPL zone, we believe we are advantaged. That said, 'front-of-the-meter' has not been universally defined and different parties may see it differently. For Susquehanna specifically, we have runway with the existing 300-megawatt ISA. Cole?

I would reiterate the point about PPL's timing: you can look at some of the filings they have made with PJM and subcommittees that indicate timing for large projects. We think that will not delay potential opportunities across our fleet. For Susquehanna, we have plenty of runway with the 300 megawatts under the existing ISA.

On the new-build side, is there any appetite for you to build a new CCGT today on a merchant basis in PJM or elsewhere?

It depends how you define merchant. A full merchant plant, meaning all revenues from capacity and energy, depends heavily on what the capacity market rules look like and the expectations for energy margins. Over the years many fully merchant CCGT projects have faced challenges. We think a more likely model is a hybrid where a portion of a new plant is tied to long-term offtake or supported by a customer balance sheet or data center developer. That combination is more likely to underpin new builds tied to large loads.

If I can just, real quick, can you characterize the opportunity for uprates at Susquehanna? There have been a bunch over the years. Any opportunity there?

We're looking at uprate opportunities. Several upgrades were done across the nuclear fleet in the early 2000s, and further uprates can be more expensive and require careful consideration of safety margins and reliability at a nuclear facility. We're evaluating potential uprates, but I wouldn't say it's a large, immediate opportunity compared to other options. So great—thanks, everyone, for joining us today, and thank you for your continued interest and support of Talen. Everyone, have a good day, and we look forward to connecting with you in the future. Take care.

This does conclude today's conference call. Thank you for participating. You may now disconnect.