Talen Energy Corp (TLN) Q4 2025 Earnings Call Transcript

Ladies and gentlemen, thank you for standing by. Welcome to the Talen Energy Corporation Fourth Quarter 2025 Earnings Call. Please be advised that today's conference is being recorded. I would now like to turn the conference over to Sergio Castro, Vice President and Treasurer. Please go ahead.

Thank you, Michelle, and welcome to Talen Energy's Fourth Quarter 2025 Conference Call. Speaking today are our Chief Executive Officer, Mac McFarland; President, Terry Nutt; and Chief Financial Officer, Cole Muller. 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 afternoon, along with the presentation, all of which can be found in the Investor Relations section of Talen's website, 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. And with that, I will now turn the call over to Mac.

Great. Thanks, Sergio, and welcome, everyone, to today's call. As always, we appreciate your ongoing interest in Talen and participation in our calls. We closed out the full year 2025 with strong results in Q4, adding the Freedom and Guernsey assets and operating well during the early winter in December. And 2026 is starting off the same with overall strong performance by the fleet and the commercial teams during the cold winter months. And I'd mentioned that PJM and other operators also performed well, maintaining grid reliability during some of the highest day-after-day loads we have seen. We saw a fair amount of elevated prices and volatility. And all in all, 2026 is off to a good start, and we are reaffirming our 2026 guidance range. Just recall that that range does not include the recently announced Cornerstone acquisition that we anticipate closing this summer. As I reflect back on 2025, we said it was going to be an exciting year, and it was across the IPP space and at Talen we accomplished a lot. We signed the reliability-must-run agreements. We signed a revamped and doubled front-of-the-meter PPA with Amazon at Susquehanna. We signed and closed Freedom and Guernsey and we delivered on the basics of being an IPP, which is safely, reliably and profitably delivering megawatts to the grid thanks to all the Talen employees that make this possible. Looking forward to 2026, we are optimistic about the continued long arc of the powering AI thesis and Talen's position in it. As I've been saying, 2025 was a year of option development, and 2026 will be the year of rationalization. In 2025, as everyone in the space was racing to develop options for data center development, and the associated power, a bow wave of expectations built across the industry for deals and more deals, whether they were virtual purchase power agreements or behind-the-meter developments. Investors were anticipating the next big thing and the next big announcement. To some, 2025 fell short; for others that grasp this long arc, they believe things will rationalize themselves out. Some projects will simply not make it, others will, some will be delayed and will need to be rationalized in 2027. But overall, we believe the long arc remains unchanged. As the CEO of Anthropic wrote in his recent essay, and I quote, "every few months public sentiment either becomes convinced that AI is hitting a wall or becomes excited about some new breakthrough that will fundamentally change the game. But the truth is that behind the volatility in public speculation, there has been a smooth unyielding increase in AI's cognitive capabilities." Again, that's a quote from the CEO of Anthropic. From my perspective, you could replace AI in that quote with IPPs or replace AI with Talen itself and the quote would keep its same meaning. And this is what I mean when I say a long arc: our capabilities to power data centers and AI have had a smooth unyielding increase. That said, there has been a lot of near-term noise that can be conflated with the rational long arc view: reliability backstop auction, overbuild, resource adequacy, regulated new build, behind-the-meter, front-of-the-meter, local zoning. They are each relevant in their own sense, interrelated in some sense. And when taken all together, they culminate in a vastness of noise, noise that can be misunderstood or worse yet turned into something that it is not. But when taken in reality, they don't change the long arc, and we remain committed to our Talen flywheel strategy. For investors, please know that we manage this long arc and seek to maximize long-term value creation and not to overreact to short-term events. We do not over rotate; nothing has changed our fundamental view that data centers are coming, coming at a rapid pace. We have the ability to contract with these entities across our fleet. We are building a further diversified fleet to support those contracts and we are building capabilities to contribute to the addition of new build. With respect to Montour, what is our plan B? That is and remains the question asked by many. I see this situation analogous to the ISA denial and the questions after the FERC decision about our initial plans at Susquehanna. But what did we do? We stayed flexible, we retooled and ultimately pivoted to a better commercial solution. We remain confident that we can do the same in this instance, too. Short-term hurdles do not define long-term success; how you respond to them does. And so therefore, we press on. Of course, Montour is just one opportunity we have in our pipeline, albeit the most well-known, and that is likely my fault for talking about it too much. We have numerous other organic and inorganic sites we are developing across the PJM footprint to further implement the Talen flywheel. This includes both powered land opportunities as well as new build opportunities. And I know many of you will want to dig into this pipeline of opportunities. But before you ask about them, let me say this: we will not discuss them at any level of detail, and we no longer plan to discuss development in the public forum and repeat the frenzied speculation that ensued around one decision by Montour County commissioners. But you can rest assured knowing that we are working the pipeline every day and have options at our disposal. On the regulatory front, we are engaging with policymakers at both the state, federal and RTO level to bring about the reliability backstop procurement, or RBP, formerly the RBA, in PJM that provides for a one-time solution to resource adequacy, which will minimize the cost on the system and allow time for real capacity market reform. And that is what our broad-based coalition of generators, hyperscalers and utilities recently proposed at a PJM workshop. We look forward to continuing the dialogue on this critical policy development. And in the meantime, we support the extension of the current floor and cap of the base residual auction in order to provide time to make these longer-term reforms. Before I turn the call over to Terry, let me conclude with this. Our strategy, and therefore our investment thesis, is based on real assets on the ground today that can support data center buildout. In doing so, we are creating infrastructure assets out of what were previously merchant generation assets subject to commodity prices, and that, in turn, is driving lower capital costs and higher returns for our investors. While we have room to run on this current portfolio, we are also set up for the future. In the future, we can augment our current assets with contracted new build and future inorganic powered land sites, something that we started last year, by the way, creating a pipeline of opportunities, as I previously described. And we have dedicated part of our management team to go after this opportunity with the recent management changes announced last December. This is a durable and tangible model built on today's reality, but with an eye towards future growth. We look forward to your questions. And with that, I'll turn the call over to Terry.

Thank you, Mac, and good afternoon, everyone. Moving to Slide 3 for a quick review of our strategic activity in 2025. During the year, we introduced Talen's flywheel, a repeatable value creation strategy that leverages our reliable, scalable generation assets and commercial capabilities to deliver durable free cash flow per share growth for our shareholders. As part of this strategy, we executed on the contracting component of the flywheel through the Amazon 2.0 PPA that was executed in June, which moved the transaction to a front-of-the-meter arrangement and upsized the volumes to 1.9 gigawatts in total. Transactions such as this will provide cash flows to support other parts of our overall strategy. In July, we executed on the acquisition component of our strategy by announcing the purchase of the Freedom and Guernsey plants, adding approximately 2.8 gigawatts of efficient CCGTs, including a significant foundational position in Ohio and subsequently brought those assets into the portfolio in late November. Throughout the year, we focused on continuing our balance sheet discipline with the ability to reduce our net leverage to below 3.5x by the end of 2026, while also maintaining a clear focus on our shareholders by increasing our share repurchase program to $2 billion through 2028. In 2026, we will continue this path of maximizing value with a focus on creating the most adjusted free cash flow per share while selectively exploring inorganic and organic opportunities that support the Talen flywheel. Turning to Slide 4. Let's talk about how we are continuing to grow the Talen fleet through acquisitions. As previously mentioned, we expanded our presence in Pennsylvania through the acquisition of Freedom and expanded our footprint into Western PJM with the acquisition of Guernsey. Shortly after closing those transactions, we entered into an agreement to acquire the three Cornerstone generation assets located in Ohio and Indiana. Western PJM has significant data center tailwinds and accessibility to reliable, low-cost natural gas from the Marcellus and Utica shales. And Ohio is an active data center hub that continues to grow. Additionally, these acquisitions diversify Talen's generation portfolio by adding high capacity factor assets that have high free cash flow conversion rates. The assets will also enhance Talen's large load contracting opportunities. As a reminder, we underwrite acquisitions on a merchant basis using current forward market energy and capacity prices, combined with more normalized views in the out years. Executing offtake agreements on these assets creates additional upside potential. Turning to Slide 5. The driving factors behind large load growth and overall power demand fundamentals continue to remain constructive. One of the largest driving forces is the significant amount of capital that is being deployed by the most well-capitalized technology firms in the world. Recent CapEx forecasts from the largest hyperscalers show significant increases in spending in 2026 and beyond, including over $650 billion of estimated spend in this year alone. We see the resulting growth of data center capacity from that spend showing up in several states where we have or plan to have solid generation positions, including Pennsylvania, Ohio and Indiana. A deeper dive into the fundamentals through the most recent PJM peak load forecast for the primary regions that we operate in provides a view of the expected load growth over the next several years. This forecast, even after the recent modifications to put more rigor around proposed large loads, shows PPL zone increasing peak load by over 70% in the next five years, while AEP zone increases by over 30% in the same period. Earlier this month, AEP reported contracted load growth of 4 gigawatts in PJM in 2026, largely driven by load growth in Ohio. Approximately 90% of AEP's reported 15 gigawatts of incremental load growth through 2030 is supported by executed take-or-pay electric service agreements. Meanwhile, PPL also reported significant growth in its territory and expects to have 10 gigawatts of signed agreements by the end of the first quarter of 2026. So what does this all mean for Talen? Two primary results: First, demand growth means higher run times for our existing generation fleet, especially our intermediate dispatch and peaking units; second, increased demand will drive more attractive economics for spark spreads and potential offtake agreements. Moving to Slide 6 and to follow up on what Mac mentioned earlier. Nothing has changed in the outlook for basic market fundamentals. Talen's underlying value proposition remains the same and still points up and to the right. The PJM capacity markets have been reflective of these tightening fundamentals as well, with the last two base residual capacity auctions clearing up the price gap. This trend is expected to continue and PJM, with the support of the governors and other stakeholders, has indicated it intends to seek an extension of the price collar for two additional base residual auctions. In relation to energy and spark spreads, we have seen appreciation in the forward curves for 2026 to 2028 from the end of July to the end of the year, with growth in spark spreads in PJM increasing over 15% during that period. Turning to Slide 7. I'd like to provide a brief update on our hedging activity this past quarter. As a reminder, we have a pragmatic, not programmatic hedging strategy. Our strategy is focused on maintaining appropriate risk tolerances and financial discipline to support cash flow stability while also leaving room to capture upside when opportunities arise. This gives our team the flexibility to add hedges during higher pricing periods as detailed on the right-hand side. As you can see from the graph in the table on the slide, spark spreads for the PJM market for 2026 to 2028 experienced upward movements during the fourth quarter, which allowed our commercial team to layer in additional hedges for 2026 and 2027 as the opportunities presented themselves. I'll now turn the call over to Cole to discuss our financial and operating performance.

Thanks, Terry, and good afternoon, everyone. As Mac mentioned earlier, for the year ended 2025, we are reporting $1.035 billion of adjusted EBITDA and $524 million of adjusted free cash flow. These results exceed the high end of our revised guidance ranges issued last quarter, primarily due to the closing of Freedom and Guernsey acquisitions in November 2025. We have more than $2 billion of liquidity available including $1.2 billion of cash and full availability of our $900 million revolving credit facility. Given that our net debt includes Freedom and Guernsey financing, but only five weeks of EBITDA contribution, our net leverage ratio using actual 2025 EBITDA is like comparing apples and oranges and therefore is not a meaningful metric for 2025. Turning to our operational metrics. Safety remains our top priority across the fleet and our team worked safely during a busy year. Our recordable incident rate was 0.55, which continues to be below the industry average. Our fleet ran well with a 4.7% equivalent forced outage factor, and we generated approximately 40 terawatt hours, about 10% more than in 2024. This was driven by a significant increase in dispatch opportunities across our fossil fleet driving higher generation and energy margin. Turning to Slide 9. Our full year 2025 financial results were significantly higher than 2024 due to a number of factors: higher capacity prices and RMR revenues that began in June 2025, the continued ramp of AWS revenues as the campus continues to progress, five weeks of Freedom and Guernsey operations as well as higher power prices net of hedges. Our results were partially offset by the impacts from the Susquehanna Unit 2 extended outage last spring and Susquehanna also not receiving the PTC in 2025. During the fourth quarter, we generated adjusted EBITDA of $382 million and adjusted free cash flow of $292 million. Note that our adjusted free cash flow in Q4 2025 alone was higher than all of 2024, demonstrating the free cash flow growth of the business, growth that we can expect will continue as we move forward into 2026 and beyond. Speaking of 2026 on Slide 10, we are reaffirming the previously announced 2026 guidance ranges. Our adjusted EBITDA range is $1.75 billion to $2.05 billion and our adjusted free cash flow range is $980 million to $1.18 billion. All of this remains consistent with our Investor Day guidance and does not include any contribution from the pending Cornerstone acquisition. As Mac mentioned earlier, while our fleet ran well during the recent winter weather, it's still early in the year, and it's not our practice to make any adjustments halfway through the first quarter. Slide 11 may look familiar to those who listened last month when we announced the Cornerstone transaction. We project continued free cash flow per share growth with our 2026 forecast more than double our 2025 actual results. Further, we anticipate the Cornerstone acquisition to create more than $4 in incremental annual impact on adjusted free cash flow per share upon closing. While we illustrate this impact beginning in 2027, there's room for upside in 2026 as we anticipate closing the transaction as soon as this summer. And our base free cash flow per share continues to move higher, supported by increasingly contracted cash flows from our long-term AWS PPA ramp. We continue to see additional upside through the four growth levers we outlined at our Investor Day last September, with the uplift potential to further build on our increasing free cash flow per share. We illustrate this impact on the slide, noting that we are already executing on these levers as demonstrated through the Cornerstone acquisition. We are focused on building our track record of delivering on opportunities to create additional growth in the coming quarters and years. We remain committed to returning capital to our shareholders through our previously announced $2 billion share repurchase program and further data center contracting opportunities, including support for the AWS ramp and potential acceleration opportunities established in the existing PPA, and we're always evaluating accretive M&A opportunities. We will continue to maintain capital discipline and focus on the most accretive levers that meaningfully increase free cash flow per share available to investors while seeking compelling growth opportunities through the Talen flywheel. Now Slide 12. Our balance sheet strength is a strategic asset that gives us the flexibility to execute the flywheel and grow our free cash flow per share. We remain committed to maintaining sufficient liquidity and keeping our long-term net leverage ratio below our stated target of 3.5x. As of February 20, our net leverage ratio using our current net debt level and 2026 EBITDA guidance midpoint is 3.0x. Upon closing the Cornerstone transaction, we expect to maintain the ability to achieve below 3.5x net leverage on a go-forward basis by year-end 2026. I'll turn it back to Mac.

All right. Thanks, Cole. With that, Michelle, why don't we open the line for questions.

And our first question will come from David Arcaro with Morgan Stanley.

Maybe if I could ask for a little bit more color on how you're thinking about the backstop auction. Just generally with some of the policy uncertainty in PJM, how are your contract negotiations and discussions progressing? Is there still interest? And is it still possible to successfully reach contracts while some of this uncertainty is going on in PJM?

David, it's Mac. I'll start and then anybody else can jump in. Look, first of all, with respect to the RBA, which is now being couched as the RBP, it is a procurement more than an auction, at least in our view, because we think it should be done as pay-as-bid, and that's the distinction that's being made there. I think that's how it's starting to be referred to even at PJM. When you implement the backstop, the concept is to use the existing tariff to procure and to fill a resource adequacy need for the out years, which would therefore relieve some of the tightness in the market. But again, if it's done at the same levels of whatever the load projections are, it wouldn't change the outcome. That's our view on that. So it's there to provide the supply to maintain the reserve margin. You know the last auction would have cleared over $500 if it had not been for the $330 cap that was imposed on that. But that backstop procurement, in our mind, actually provides a relief valve and therefore allows for contracts to continue to go forward. It was one of the tenets of the National Energy Dominance Council's statement that came out; in that group when they put that forth, they said that in addition to it being a one-time and limited procurement, there should be longer-term capacity reforms and then a return to the outcome, but it should allow for continuation of existing contracts. I think that by gaining more certainty as we work our way through the process that obviously continues to support that. And so that is, as a practical matter, we view that as a relief valve to doing existing contracts. As far as existing contracts and the discussions that are ongoing across the fleet and across our pipeline of opportunities, those have not slowed down. I think that the regulatory uncertainty — who pays for this and how it gets paid for and how it gets allocated and how it gets procured — will work its way through. But data centers are coming and they're not slowing down. Any time you talk to or hear any of the analyst calls, whether it be from the chip manufacturers to the hyperscalers themselves, they continue to talk about the race for creating data centers on the ground, powering them today, powering them in '28 and then soon '29 will become the new '28. As we progress through that, it just further aids in the ability to continue those discussions. So we don't see any slowdown to it. And we think that the RBP will ultimately increase the level of those discussions.

Got it. I appreciate that color. And then maybe when it comes to the procurement, do you have upgrades or new builds that you think might be opportunities to bid into the procurement?

We are working on a set of opportunities in the new build front. We do think that upgrades should count if you're asking that. Most of the upgrades that we had at Susquehanna were put in about ten years ago, so there's not a lot there that you may hear from other producers. But there are opportunities we've been working on thinking across the spectrum of the form of generation, whether it be batteries, CTs or CCGTs in developing those opportunities. With a 15-year contract at the right price, you can make the math work. So obviously, we're gearing that up. And once the rules are more defined, we would look to see how we can participate.

And our next question will come from Angie Storozynski with Seaport.

So I'm just trying to link the comments that we're hearing from PPL and AEP to your generation contracting. For example, the comment that you quoted in the slides that PPL expects 10 gigawatts of load under ESAs by the end of the first quarter, which sounds like one more month. How does that relate to you being the largest generation company in the PPL zone and signing generation contracts to back this 10 gigawatts of load?

Well, first, we don't have that list; PPL does. So what's in that list specifically, you'd have to talk to them. But I do think that is a very supporting point to the question that was just asked that this is not slowing down and that PPL is signing up the ESAs. Now signing up ESAs, you don't necessarily need to procure your energy and capacity immediately, but what you're doing is making a commitment to pay for the network upgrades and the rest of it. That, to me, is a highly positive sign that supports that nothing is slowing down. Now we don't have what the list is. But obviously, we're working a pipeline of opportunities ourselves to participate going forward.

So look, Angie, the ESA point, that's the first step. Without an ESA, data centers aren't going to contract for a PPA. So I think that's just a good leading indicator of PPAs coming. And just to be really clear, we obviously have announced roughly two gigawatts of tangible PPA in that zone. So leave it to PPL to break down their count, but that's two of the ten right there.

Okay. Okay. I mean we're waiting as you are aware. So maybe one thing... (inaudible)

Angie, I would add that we have a model that we're doing it with hyperscalers, but there are other developers that are co-locators — your typical data center people — who build and connect data centers and then lease those out. In those models, energy and capacity typically can be a pass-through. So you get the ESA first even in our model and then get the energy and capacity. In some models, data centers just take energy from the grid as a pass-through. Hyperscalers and the like and the development model we have is powered land: get the ESA, get the energy and capacity and then put all that together like we did at Susquehanna.

Okay. So on Slide 11, and I know it's the same slide that you had in your Investor Day presentation, two things. One is the upside potential to the free cash flow per share you're showing for 2028. Is there no potential upside to, say, 2027? And number two, as you show the new one gigawatt of data center PPAs and then accelerated Susquehanna contract by 480, is that just a measurement or is that what you would expect to happen as potential upside? In other words, is that the cap of additional data center PPAs and additional ramp under the Susquehanna contract by '28?

Let me provide some context on the slide. When we did this slide at Investor Day, we ended with the 2028 outlook and showed levers that could increase free cash flow per share. The timing of the levers was tied to 2028 as the illustrative terminal year. The timing can vary: for example, we pulled Cornerstone forward into 2027, and we could pull some of it into 2026 with the expected close. So there's upside to the guidance. One of those levers was the Susquehanna acceleration and the new gigawatt PPA. The slide was an illustration of potential upside by 2028. The timing can be pulled forward for some items, but others — like a new large data center PPA — are more likely to deliver capacity starting in 2028 due to build-out timelines. So while there is flexibility in timing, the slide was meant to show a representative view of upside by 2028 rather than a strict scheduling of each lever into 2027.

Yes, the 480 and the one gigawatt were just representative numbers so folks can make their own assumptions and scale. By 2028, the contract, as we've disclosed before, gets up to the first 480 megawatts. So we just put out what it would take to get to 960. We could have gone all the way to 1,920, but we didn't think that was necessarily helpful. We wanted to show the impact of every 480. And then on the new data center PPA, one gigawatt was a standard illustrative number. It could be more, and you can scale from there.

If you think about the ramp that's going on at Susquehanna, any new data center PPA is more than likely post-2028 because building data centers takes time. That's why we showed it in 2028. Even if a contract were signed today for a virtual PPA or across our pipeline of opportunities, the delivery of those megawatts is not going to be 2026 and will likely ramp starting in 2028. So the timing of signing is often irrelevant to when the megawatts flow under that type of arrangement. That's why there was a lot of short-term discussion around the Montour County commission vote, but when you look at the delivery of megawatts, we're still on the long arc I described earlier. The long arc hasn't changed. It's a short-term hurdle. Would we prefer a different commission outcome? Absolutely. But we are commercial, and we'll figure it out. We have other opportunities in the pipeline to do the same thing. That's what we're looking at: adding another gigawatt data center contract, but the delivery won't start until 2028, so the signing date in 2026 has limited impact on 2028 delivery.

Can I ask just one follow-up on that? Why is it at all linked to that Montour site? AWS has other sites in the PPL zone. You have existing assets in the PPL zone, your Susquehanna 2.0 contract supply, and other sites being developed. So why couldn't a PPA serve some of those other sites and thus the impact on 2028 EBITDA would be more likely?

Excellent point, and you're making our point for us: it is largely virtual PPAs. When we moved Susquehanna to front-of-the-meter, one of the attributes of that transaction was that we're obligated to deliver anywhere in Pennsylvania. So on the 480 acceleration, other data centers can take under that contract earlier, but if they build more than the contracted amount, they'd need additional megawatts from the back end. Does the contract need to tie to Montour specifically? Not necessarily. It's about the delivery point. But if you're a data center developer, you need sites with line of sight to construct the data center and direct the megawatts. So while they are interrelated, they're not necessarily discretely intertwined; they can be, but they don't have to be.

And our next question will come from Michael Sullivan with Wolfe.

I wanted to maybe just unpack a little more of some of the cross currents within Pennsylvania, in light of latest commentary from Governor Shapiro and then with all this PPL load coming forward. Where does existing generation versus new generation fit in? Can it all be served with the existing transmission capacity? And when do we need to start thinking about new build and how that ties to the political rhetoric?

I think the political rhetoric is focused on affordability and resource adequacy. We think the RBP is the way to solve that and it does provide a relief valve. In any of the proposals, it contemplates that there is a carve-out for existing contracts with respect to cost allocation. That's one of the things that's been discussed: how do you allocate any procurement cost? It would not be allocated to new loads if they had an existing contract. That's somewhat standard across the different coalitions. So again, it allows us to continue contracting. It's yet to be seen how that allocation and the rest of the RBP will pan out, but in the meantime, nothing is stopping activity. People are lining things up and trying to figure out where they go from here with respect to the RBP.

All I'd say is everyone would agree that data centers are coming and the loads will continue to increase and ramp up in 2027, 2028 and 2029 and continue from there. I don't see too much new generation that can actually serve that load immediately. So we continue to focus on conversations around existing generation. At some point new generation needs to come online. Those decisions need to be made soon. The RBP is one angle, bilateral contracts are another. Over time, we think it will shift to hybrid models where existing generation powers the first three- to five-year build-out of data centers across Pennsylvania, Ohio, Indiana and so forth. Then eventually it's backed by a second upscaling of a PPA or a second PPA that enables new generation to fill the gap.

New generation can come either via the RBP award or via 'bring your own power' arrangements, and it's a question of transitioning from existing to new. The RBP or bilateral contracts will determine how that transition occurs.

Michael, to add to those comments, the Governor's team is engaged in the PJM discussions around the RBP. They're an active participant. They've heard proposals from different coalition groups and remain engaged. The concept is going to be to procure additional generation as it moves forward, and the state is active in that process.

Okay. Great. And then on your pipeline — the organic opportunities, the inorganic powered land — can you provide more color on how you weigh those economics and speed? Presumably, you have a lot of land already, but what's the value proposition of the inorganic powered land angle?

It's a great question, and in a perfect world we'd be excited to talk about it. But every time we do, we're running a commercial trade in our face or creating an expectation about a certain outcome. That's why I said we won't discuss specific pipeline opportunities. It's commercially sensitive and it creates expectations. We created those expectations around Montour — our fault — and we'll figure that out; there's a Plan B. But we have other opportunities in our pipeline that give us more options. If Montour had gone and a deal announced, everyone would ask 'what's next?' Well, we're not getting one deal done and then moving on. We're working multiple fronts all the time. What we've realized is there was an undue concentration on one outcome, which isn't going to define the long arc I described. We're working on existing sites and other sites that others have and want to work with us on. We will steer away from public speculation going forward because it impedes our ability to develop other things. I know that doesn't satisfy the desire for more detail, but that's the path we're taking.

And our next question is going to come from Jeremy Tonet with JPMorgan Securities.

Just want to build on some comments. As the hyperscalers head to D.C. next week, what do you see could be possible coming out of those discussions? When they discuss bringing their own generation, what does that mean in Talen's view and how could this impact market architecture?

Jeremy, we don't know what they'll commit to. There's been speculation. Publicly, all of them have committed to pay their fair share; the definition of 'fair share' is the key. PJM is an RTO based on broad allocation and not simply the next incremental megawatt paying for the next incremental generation. There are states that are generation-deficit that have paid for transmission and use it; there are LSEs that are short. We happen to be in an LSE that's long in Pennsylvania with transmission that can absorb these loads. It's going to come down to what they commit to as fair share — I don't know what they'll commit to — but it will influence market architecture and allocation discussions.

Got it. And on gas contracting — how have discussions evolved over time? Any comments around hyperscaler appetite to absorb gas risk? Could there be fixed capacity plus heat rate type of arrangements? How do you see those conversations developing?

Jeremy, it depends on the counterparty. Some hyperscalers may have more appetite to take on gas cost variability and some less. We've explored many different structures internally and with counterparties. There are a number of avenues to ultimately contract and protect ourselves in any structure. We have a commercial desk that can manage positions, and if we contracted in a manner where the buyer took gas exposure, we would obviously have a different premium structure in the PPA to accommodate that. There's a variety of different structures, and when we have a deal announced, we'll discuss it more.

Cole is exactly right. If I were advising someone buying this power, I'd say you want the party who manages the commodity risk to take the commodity risk and to be paid to do that unless you're warehousing that risk yourself. We are set up to manage commodity risk, provide credit support, manage physical gas delivery, and manage the financial risk associated with gas. That's the full suite of services we're trying to provide. Some counterparties will pick and choose across those services.

And our next question is going to come from Nicholas Campanella with Barclays.

I wanted to follow up on how you frame what's going on around the RBA/RBP and the ratepayer protection pledge. Do you still feel you have the ability to sign gas with incumbent generation in a front-of-meter framework? Are you saying this would only now come with additionality and new build commitment, or can existing assets still sign in the front-of-meter framework?

Let me be very clear: yes. We think you can continue to contract with existing assets.

Okay. In the Montour hearing specifically, Amazon said this would not be an additionality deal and discussed the state of the supply chain. Some peers have talked about new build gas using turbines or bridge power that others have procured. Could you talk about your existing EPC relationships or your ability to do something internal via partnership or inorganically to secure the supply chain further to deliver on that?

Happy to. Our view — and the reason we've built up and invested in existing assets — is that there's still capability to use existing assets to contract. Many data centers are looking at whether they contract for a longer period with existing assets. When it comes to new build, new build will require either winning in the RBP and having a 15-year contract that takes merchant capacity risk off the table, thereby allowing financing, or it will require a bilateral contract. If you have either a long-term contract or an RBP procurement award, you can attract turbines, EPC providers and financing; it becomes a live project. We have relationships with EPC providers and have invested in capabilities: we put Dale in the Chief Asset Development Officer role specifically to focus on technology, costing, and EPC work. But ultimately it is the offtake agreement that enables financing and construction, not the turbine order alone.

And our next question will come from Nick Amicucci with Evercore.

On the hedge book looking through 2027, with a lot of upside optionality, how should we expect hedging to continue to increase over time? Are you comfortable leaving that open given that forward curves still aren't fully reflective of the tightening?

Nick, I'll get our Chief Commercial Officer to jump in. We saw an opportunity in December when prices went up to take some of '27 off the table or lock it in through hedges. We do this pragmatically, not programmatically. We don't have fixed deadlines; we add hedges when the market presents compelling opportunities. It's hard to predict exact timing but we've been opportunistic and will continue to be.

We've intentionally been longer in the outer periods, waiting for instances of volatility that provide opportunities to lay off hedges. As we've seen this winter, tightening supply and demand will create real-time opportunities for us to continue layering hedges. We have stated targets and ranges, and we've been at the lower end intentionally; we will continue to add to those as compelling market opportunities present themselves.

When we came out of prior periods, we laid out guideline targets like 60% to 80% for prompt periods and 40% to 60% for prompt-plus-one. Those are guidelines; Chris manages the position with input from Terry and Cole. With the growth of our fleet, including Cornerstone, Freedom and Guernsey, we're managing a much larger book and need to take opportunities when they arise, but it's not a forced approach — we can decide how to tilt the book based on our fundamental view.

As we get more contracted margin in the overall portfolio, there's less need to lock in hedges. We take a view on both contracted cash flows to support the business and the opportunistic upside. As AWS and other contracted margins increase, hedging becomes more opportunistic, which is a benefit we like having.

Quick cleanup question on Cornerstone: obviously it'll depend on timing, but is it fair to take the $500 million in EBITDA and allocate that over the months from the closing date across 2026? Or is there growth embedded in there for 2027?

Nick, I think that's a good run-rate number. Pick your assumption on a close date and twelve months forward from there, it's a good round number.

And if you want to be a little more precise, there is more value in the winter months — July, August into December — across that. But $500 million is a reasonable round figure to model.

And the next question is going to come from Craig Shere with Tuohy Brothers.

A year ago, new build was not really part of the discussion at Talen. Now it sounds plausible that you might have something by year-end, especially through the auction. How are you thinking about capital and balance sheet management decisions over the next two to three quarters given the potential for chunky new build? Would you be less aggressive with the balance sheet, or would you continue M&A and buybacks as before?

Craig, we've always said if we have the right certainty — whether through an offtake agreement or a clear underwriting case with an off-taker — we would be happy to do new build. The RBP potentially gives you that clarity because a 15-year commitment is what you need to underwrite and finance new build. The challenge has been financing in PJM. Regarding capital allocation, we're always balancing highest and best use of capital: buybacks, M&A, new build. We look for high returns; we've done significant share repurchases over the last 2.5 years, buying back over $2 billion worth of stock at an average price of $149 a share. The Freedom and Guernsey transaction was greater than 40% accretive free cash flow per share. We will continue to direct capital to the highest returns and stay disciplined.

(Inaudible) ... what about the balance sheet given the need for potential credit support on long-term gas-fired PPAs? There's more to think about now than a year ago.

Craig, a year ago, we said we'd consider new build; now we're closer to the timeline and the market is evolving. It depends on the structure. If you win an RBP award and receive a contractual capacity payment, it could be financed in a project-finance structure that may require less corporate balance sheet. If we do bilateral PPAs that require credit support, we work through letters of credit, first-lien arrangements, and other options. There's not an easy one-size-fits-all answer. We toggle among share repurchases, balance sheet adjustments and M&A based on returns and opportunity. We'll apply the same discipline we've used historically to any new build decisions.

Yes, we are past time.

I think we're going to end there. I apologize we didn't get to everyone in the queue, and we're happy to take follow-up questions to Sergio and the rest of us here. Appreciate everybody's interest in Talen and have a good evening.

This concludes today's conference call. Thank you for participating, and you may now disconnect.