NRG Energy, Inc. (NRG) Earnings Call Transcript & Summary
August 4, 2026
What were the key takeaways from NRG Energy, Inc.'s August 4, 2026 earnings call?
In the second quarter of 2026, NRG Energy, Inc. reported solid results with adjusted EBITDA of $1.2 billion, reflecting a 34% increase year-over-year. However, adjusted net income decreased to $315 million from $339 million, and adjusted EPS fell to $1.49 from $1.73. Management reaffirmed its 2026 guidance, indicating confidence in navigating current market challenges while advancing a significant 1.2 gigawatt project in Texas, which is expected to generate $500 million in annual adjusted EBITDA upon completion.
What topics did NRG Energy, Inc. cover?
- 1.2 Gigawatt Project in Texas: NRG is advancing a 1.2 gigawatt project in Texas with a leading global cloud and AI hyperscaler, which is expected to generate $500 million of annual adjusted EBITDA. Management stated, "The project is designed to bring more new generation to Texas than the data center is expected to require," highlighting its strategic importance.
- Reaffirmation of 2026 Guidance: Management reaffirmed its 2026 financial guidance despite tracking below the midpoint due to softer load and power prices in Texas. They emphasized, "Our first half results largely reflect the impacts of weather and market conditions, not a change in the underlying performance of the business."
- Adjusted EBITDA Growth: NRG reported adjusted EBITDA of $1.2 billion, up $308 million or 34% year-over-year, driven primarily by the acquired portfolio from LS Power. Bruce Chung noted, "The year-over-year increase in adjusted EBITDA was driven primarily by the acquired portfolio, higher PJM capacity values and continued growth in Smart Home."
- Challenges in Texas Market: Texas adjusted EBITDA declined by $131 million year-over-year, primarily due to lower load and power prices, with ERCOT prices averaging $33 per megawatt hour, which is 8% lower than last year. Chung stated, "Our fleet had fewer opportunities to run, and our commercial team had fewer opportunities to optimize the portfolio."
- Virginia RGGI Costs: The rejoining of Virginia to the Regional Greenhouse Gas Initiative is expected to create an estimated $70 million of incremental cost in 2026, which was not included in prior underwriting. This was highlighted as a new headwind for the acquired portfolio.
What were NRG Energy, Inc.'s August 4, 2026 results?
- Adjusted EBITDA: $1.2 billion (+34% YoY)
- Adjusted Net Income: $315 million
- Adjusted EPS: $1.49
- Free Cash Flow Before Growth: $1.025 billion
- Texas Adjusted EBITDA Decline: $131 million (null)
- Total Investment for 1.2 GW Project: $3.2 billion (null)
NRG's solid quarterly results and the advancement of the 1.2 gigawatt project position the company favorably for future growth. However, challenges in the Texas market and regulatory changes in Virginia present risks that investors should monitor. The company's commitment to disciplined capital allocation and shareholder returns remains a key focus as they navigate these dynamics.
Earnings Call Speaker Segments
Thanks, and good morning, everyone. Welcome to NRG Energy's second quarter 2026 earnings call. Today's call is going out live by phone and through webcast. You can find the webcast presentation and earnings release in the Investors section of our website, www.nrg.com, under Presentations and Webcasts. I want to note that today's discussion might include forward-looking statements, which rest on assumptions we consider reasonable as of this date. Actual results could differ materially. We encourage everyone to look over the safe harbor in today's presentation, along with the risk factors in our SEC filings. We assume no obligation to update these statements because of future events, unless the law requires it. Also, we'll discuss both GAAP and non-GAAP financial measures. For details on our non-GAAP financial measures and reconciliations to the most directly comparable GAAP measures, please see our earnings release and the non-GAAP reconciliations and supplemental data file in the Investors section of our website. With that, I'll hand the call over to Robert Gaudette, NRG's President and Chief Executive Officer.
Good morning, and thanks for being with us today. Right from the start, our focus has been on meeting the coming surge in power needs in the correct manner. When it comes to the biggest new loads, fresh demand ought to be paired with fresh generation, with the customer backing that investment. That is the way expansion on this scale ought to function. It shields current ratepayers, bolsters the grid and generates lasting worth for the communities we serve as well as for our shareholders. What has happened in Texas during the past 24 hours underscores why this method is so important. States are eager for the economic expansion that data centers deliver, yet they likewise anticipate that added demand will come with added supply, sustain the infrastructure needed and reinforce — rather than burden — the power networks and the communities that enable it all. The landscape has shifted. Our approach has not; as a matter of fact, policy is heading in the direction of the very model we have been constructing since day one; we possess the commercial framework, the machinery and the expertise to execute it at scale. Later today, we will take you through the commercial structure we are chasing, the 1.2 gigawatt project moving forward under it and the wider chance that lies ahead of us. We have reached alignment on the key commercial terms with a top-tier global cloud and AI hyperscaler, which includes their capital pledge to back 1.2 gigawatts of fresh generation in Texas, with room to grow to 2.4 gigawatts. This is anticipated to be our inaugural Bring Your Own Power project, and it mirrors our playbook for large load expansion. We think it ought to become the industry norm, fueling economic growth, satisfying our customers' growing power requirements and safeguarding families and small businesses. The commitment is going to be long term. Credit quality is robust, the economics underpin both the investment and our targeted return. What we're seeing here is growth that's both disciplined and substantial, anchored by a major investment-grade client with a clear path toward expanding the relationship. On top of that, we delivered solid results for the second quarter and are holding steady on our 2026 financial guidance. Bruce is going to walk everyone through the details of the quarter. As I mentioned earlier, we're advancing a 1.2 gigawatt project in Texas together with a leading global cloud and AI hyperscaler. We've come to agreement on the core commercial terms, and both the negotiations and the remaining land issues are progressing at the same time. The customer has made a financial commitment to help move this project forward. The key point is that this project is designed to supply Texas with more new generation than the data center is expected to require, and in our view that design leaves us well positioned to help meet the state's goals for power and reliability. Before we reach any final investment decision, the standard conditions will need to be satisfied, and that includes getting the internal approvals we need. These deals are highly complex, and there is still work to be done, but we feel good about the way we have put the structure together and about what we plan to deliver together with our partner. The new combined cycle gas plant is going to be developed, owned and operated by NRG. The facility is set to serve a 1-gigawatt data center load, with further Texas development opportunities that might grow the relationship to as much as 2.4 gigawatts. The turbine and EPC capacity we locked in via GE Vernova and Kiewit back this project. This investment likewise needs to benefit the community around it. We anticipate over 1,400 well-paying construction positions, 30 permanent jobs at the facility, and meaningful new tax dollars for nearby governments and schools. For decades, NRG has run power plants in Texas, and our workers reside in these same communities; we recognize water is important, and both we and our customers are dedicated to responsible water stewardship, and we're collaborating closely with local stakeholders as development moves forward; we also grasp the wider worries about data center expansion. What communities want is for that growth to be responsible, to honor local resources, and to generate genuine, lasting benefits. That's the way we're pursuing this opportunity. The project's initial term runs no less than 15 years starting from commercial operation, with room for extensions. Given the present development timeline, commercial operations is aimed at late 2029, with full run rate earnings from then on. When it's up and running at full capacity, we expect to see $500 million in adjusted EBITDA each year, along with $375 million in annual free cash flow before growth. Those figures apply to a 1.2 gigawatt project and don't include any possible expansion. This represents high-quality, long-duration earnings, and they're backed by an outstanding investment-grade counterparty. On a stand-alone basis, the project ought to generate attractive returns that clear our required investment hurdles, and on a risk-adjusted basis they look even more compelling. It also works out to a build multiple that's below where NRG currently trades. The facility being contemplated is expected to require $3.2 billion of investment. Bruce will cover the specifics of what the capital requirements look like and our approach to funding this project, but I want to state plainly that our commitment to giving back a minimum of $1 billion each year to shareholders through buybacks has not changed. We have the financial capacity to support this project as it moves forward, stay on track toward our target leverage, and continue carrying out our capital allocation framework. The economics are attractive, and it's our commercial structure that makes us confident those economics will hold up. Let me now take you through the details. Turning to Slide 6, there are two pieces to the commercial framework: the Capacity Payment is structured to recover the capital we put in and provide the return we need. There's also a distinct Operating Payment that handles the cost of natural gas along with expenses tied to running the plant. Basically, the way we earn is based on the megawatts we construct and keep ready to serve — it doesn't depend on the extent to which the data center is actually used. That difference matters a great deal. Within this commercial framework, Capacity Payments underpin 95% of the project's free cash flow throughout the term, no matter how much the data center gets utilized. Costs for fuel and operations are recouped through a separate channel, and a guarantee from an investment-grade parent company will stand behind what the customers have committed to. What that produces is cash flow that's both steady and easy to see. The return we earn is locked in from the start and isn't tied to merchant power prices or the price of natural gas. And even more significant — this kind of structure isn't limited to just one project. We don't have to reinvent the model on each occasion. Yes, the customer, the location, and the scale of the project may vary, but the fundamentals remain the same. Within the commercial framework, NRG backs the investment, handling development, ownership, and operation of the generation, and the financial terms are settled before construction begins. What distinguishes NRG is our ability to assemble the whole solution; we provide a unified path to power, from bridge solutions through to permanent combined-cycle generation, with the capability to operate in island mode, grid-connected, or to alternate between the two. Pairing a load with generation can also reduce how much additional transmission buildout is required to meet that demand. That's a further major benefit of the BYOP approach. Beyond that, we possess the internal expertise to develop, engineer, interconnect, commission and run these assets across their full life cycle. So the customer ends up with one experienced partner who remains responsible starting from the earliest design through many decades of operation; fewer handoffs result, and that helps lower execution risk in a power development that is highly complex. We built those capabilities across many decades, and we are proving them today. The 1.5 gigawatt portfolio we have through the Texas Energy Fund remains on schedule, and that includes T.H. Wharton, a project we finished on time without going over budget.
We acted quickly to lock in both turbine and EPC capacity via GE Vernova and Kiewit, which gives us the equipment plus the execution capability needed to keep building at scale. Not many companies are able to pull all of those pieces together. I'm proud that NRG is one of them. That's the reason this opportunity landed with us, and the reason we're set up to repeat it. Turning to the next slide, the market backdrop keeps getting more attractive. In both ERCOT and PJM, the demand growth being projected is well ahead of the supply that's presently anticipated to come online. Not every forecasted project has to come to fruition for substantial new generation to be needed in both markets. That gap is reshaping the market. Customers require power solutions they can actually execute. Policymakers are either driving toward growth or steering growth toward supply backed by customers, and value is shifting to companies holding genuine development positions along with the capability to deliver. Our BYOP framework addresses the reliability and affordability worries of elected officials and regulators. Being able to design, build, own and operate a power plant across decades sets our solutions apart. We've worked in and lived in the community for a long time. We operate responsibly and act as a responsible community member. In the world we're in today, that counts, and that's exactly where NRG stands right now. Let me now frame how large this opportunity really is. The 1.2 gigawatt project we're discussing today is the initial step toward realizing the full potential. It accounts for the first 1.2 gigawatts out of the 5.4 gigawatts of turbine and EPC capacity we locked in through 2032, with line of sight to the critical labor needed to carry out that build-out. Our wider development pipeline exceeds twice the 5.4 gigawatts of capacity we've secured, and each turbine slot is tied to an active customer discussion. Customers see the value and the scarcity of the development position we've put together, along with our technical expertise and capabilities. And as you would expect, engagement across that pipeline keeps growing. Potential capital partners likewise see the value in what we've assembled, opening additional pathways to move the broader opportunity forward through capital-efficient structures while keeping balance sheet flexibility and maintaining our disciplined, consistent return of capital to shareholders. On top of that, we hold roughly 2 gigawatts of uprate opportunities throughout our PJM fleet. Combined, that leaves us a substantial runway to apply the model we just walked through. Let me be clear on how we'll chase that opportunity. We won't swap discipline for scale. Every project has to stand on its own, clear our risk-adjusted return thresholds and carry the commercial and credit protections suited to the capital we put to work. Pairing the established base with a 1.2 gigawatt BYOP project produces an illustrative 23 — pardon me, 2030 contracted free cash flow opportunity of $1.2 billion. For this illustration, we keep current capacity auction prices flat through 2033. That's an assumption, not a forecast of what future auctions will produce. Assuming we get this project across the finish line — and I have a lot of confidence that we will — then when you combine that with signing up the rest of the new build possibilities and completing the uprates, the free cash flow backed by long-term agreements and capacity revenues could hit 95% of the midpoint of the 2026 free cash flow guidance for the whole company by 2033. And that would represent just a single piece of NRG. The remaining portions of the business would keep producing cash flow and building value right alongside it. Just to remind everyone, prior to any data center opportunities, we expect our business to deliver adjusted EPS CAGR of 14-plus percent through 2030; that is the chance we have to meaningfully grow NRG while fundamentally raising the quality of its cash flow. Our goal is to assist in constructing the power infrastructure that sits behind America's digital economy, while safeguarding communities and customers, whether large or small, and all the while building a bigger, stronger and higher quality NRG along the way. This marks a significant step. We want it to be the first of many. Bruce, the floor is yours.
Thanks, Rob. Now moving to Slide 10. NRG posted a strong quarter, with adjusted EBITDA reaching $1.2 billion — a gain of $308 million, or 34%, versus the same period a year earlier. Adjusted net income came in at $315 million, compared with $339 million last year, while adjusted EPS was $1.49 versus $1.73. Free cash flow before growth totaled $1.025 billion, an increase of $111 million year-over-year. This marked our first complete quarter including the portfolio we purchased from LS Power, and we couldn't be happier with the asset quality and what these assets are adding to the business. The rise in adjusted EBITDA from a year ago stemmed mainly from the acquired portfolio, stronger PJM capacity values, and ongoing expansion in Smart Home. Adjusted net income and adjusted EPS came in somewhat lower, since interest expense tied to the acquisition and D&A more than offset the larger EBITDA contribution. That's the net income and EPS pattern we anticipated near term while de-leveraging. As debt and the related interest costs come down, a greater share of the portfolio's earnings will reach EPS. Now on to segment results. Texas adjusted EBITDA fell $131 million from a year earlier, largely because of weaker load and power prices. During the quarter, around-the-clock power prices in ERCOT Houston came in at an average of $33 per megawatt hour. That's 8% lower than a year ago, and it falls well short of the $52 figure we'd built into our 2026 planning assumption. With prices this low and volatility this subdued, our fleet didn't get as many opportunities to run, and our commercial team didn't get as many opportunities to optimize the portfolio. East adjusted EBITDA was up $370 million versus the prior year, driven largely by what the portfolio we bought from LS Power contributed. Energy margins on those assets didn't fully capture the upside from stronger PJM power prices, because some hedges that predated the deal were still in place when the transaction closed. The results also took in elevated supply costs across our Retail businesses. There's another East-region development worth noting: Virginia has come back into the Regional Greenhouse Gas Initiative, which people call RGGI. Once we closed on the portfolio we bought from LS Power, lawmakers in Virginia enacted a law that puts the state back in the program starting July 1. Because of that shift, the 1.2 gigawatts of Virginia assets inside the purchased portfolio will carry roughly $70 million in added cost during 2026 — something our underwriting had left out. Turning to the West, adjusted EBITDA was up $27 million compared with a year earlier, largely driven by lower operating expenses once a facility lease ran out during the prior year. At Smart Home, adjusted EBITDA climbed $42 million, helped along by ongoing gains in customers and a bigger recurring service margin for each customer. We finished the quarter with 2.45 million customers, which is up 8% compared to the same period last year, and that growth continues to outpace what our long-term outlook assumes; given strong results in the second quarter, our 2026 guidance ranges remain unchanged. During the first six months of 2026, weaker load and power prices in Texas, combined with higher regional power supply costs from Winter Storm Fern, mean we're running under the midpoint of those ranges. Although PJM power prices have risen, the hedges already in place on the portfolio we bought, together with higher RGGI expenses, have limited the near-term upside. What we saw in the first half mostly comes down to weather and market forces rather than any shift in how the business is fundamentally performing. When we set our guidance ranges, we anticipate outcomes of this kind, and we actively manage the portfolio so that expected supply lines up with committed customer load, which keeps us on track to land inside those ranges. So as we progress through the rest of the year, our unhedged exposure is limited, and a meaningful rebound in commodity prices isn't something our outlook depends on — that's why we feel confident about delivering within our guidance ranges. Turning to Slide 11. Our 2026 capital allocation plan has been updated to account for the first tranche of spending on the 1.2 gigawatt Texas data center new-build project Rob mentioned. As the chart indicates, aside from moving some planned liability management dollars over to new build investments, all remaining elements of our 2026 capital allocation are unchanged. The key point is that this spending does not affect our previously stated commitment to repurchase a minimum of $1 billion in shares annually. What's different is the addition of a data center new build investment category, representing $721 million in anticipated project investment for 2026. Of that amount, $40 million was previously classified under plant and other investments and has now been moved, meaning the full project investment is now shown in one line. The remaining $681 million represents the added change to the plan, and it will be covered through reduced liability management — which translates to a smaller decrease in net debt for 2026 than we'd originally intended. It's worth pointing out that the vast bulk of what we expect to spend in 2026 relates to procuring equipment. That spending isn't only critical for the project we're currently evaluating — it's also critical for keeping alive the option value of that equipment, which keeps growing, since new generation is going to play such a major role in the data center build out. Since most of that money goes toward equipment, we have the option to shift it into other projects that make sense — so it isn't money we can never recover. The way we're helping enable the data center build-out, plus the set of potential deals we're working on, makes us feel good that we're making wise investments that should generate the right level of return. Just to remind everyone, back in April we accelerated our plan to bring down debt following the acquisition by completing several refinancing deals. We paid off nearly all of the $1.5 billion in Lightning senior secured notes that came with the acquisition, and we also paid back part of what we had drawn on the revolver to finance the deal. Those moves pushed out our average maturity dates, brought down the amount of secured debt, and should produce annual interest savings north of $10 million. Our 3x long-term leverage goal stays exactly where it was. On top of that, we're making progress on the 2026 plan for returning capital. During the first six months, we finished $921 million in share buybacks and distributed $202 million of common dividends. For the full year, nothing has changed in our outlook: we expect to buy back $1 billion in shares and pay $407 million in common dividends. Let's turn to Slide 12. Rob has already covered the proposed commercial framework. What I want to focus on is what it means financially and how we intend to fund this project. Thanks to the way the new build project's commercial framework is set up, the return we underwrite is safeguarded via a capacity payment tied to availability, with fuel and operating costs recovered separately and only limited exposure to commodities. The counterparty carries an investment-grade rating, and suitable credit support will stand behind its obligations. Once fully operational, this initial 1.2 gigawatt project should produce no less than $500 million in yearly adjusted EBITDA and roughly $375 million in yearly free cash flow before growth; given $3.2 billion in total investment, we anticipate the project will yield a pretax un-levered IRR falling inside our 12% to 15% target range. If you apply that anticipated run rate EBITDA, you get a build multiple of roughly 6x. Keep in mind that these earnings fall outside the long-term framework we laid out earlier this year. That framework—which contains our forecast of adjusted EPS CAGR above 14% through 2030—rests on the base business by itself. This project adds substantial incremental earnings power.
Our intention is to finance this project using cash generated from operations along with capacity on our balance sheet, which includes reducing liability management activity. This means net debt will come down by a smaller amount than we had originally anticipated across the build timeline. Our dedication to a long-term net leverage target of 3x is unchanged, and we view that level as fitting with investment-grade credit metrics. From a credit and ratings standpoint, we see the anticipated cash flows and the credit quality of our counterparties as supportive. Because we expect to keep carrying out share buybacks of no less than $1 billion each year throughout construction, this funding approach maintains the capital allocation promises we have already announced. Finally, we anticipate the project will be eligible for bonus depreciation at COD. That COD pushes our cash tax runway out even further. Let's turn to the following slide. For this 1.2 gigawatt project, the total investment is projected at $3.2 billion, equivalent to $2,700 per KW. Capital will be spent across 4 years, with the biggest expenditures coming after important development and construction milestones are reached. By the close of 2026, cumulative investment should reach $0.8 billion, which counts reservation payments already made. After that, our expectation is to put in $1 billion during 2027, $1.1 billion during 2028, and the final $0.3 billion during 2029, prior to the anticipated late 2029 COD. EPC accounts for 60% of the investment, while turbine equipment and other project costs make up the rest. This investment schedule is intentionally staggered. The money goes out as the project advances, so the bigger outlays come once key milestones are hit. We keep significant flexibility all the way through development and construction. As noted earlier, a large portion of the 2026 spending is tied to equipment that, if needed, might be shifted to other viable projects. So we regard this investment as less tied to one specific project and more as a bet on NRG's distinctive ability to provide solutions that suit customers. Under our present plan, NRG provides the funding and holds ownership of the project. As development moves forward, we'll look at ways to boost capital efficiency, such as bringing in financial partners, while keeping the economics and strategic worth of the investments intact. To wrap up, we posted strong second quarter results and reiterated our 2026 guidance. This new-build data center project brings a significant additional source of contracted earnings on top of our current framework, with returns safeguarded by a strong commercial structure and a funding approach that maintains the commitments we've made to shareholders. With that, I'll turn it back over to Rob.
Appreciate that, Bruce. Let me wrap up by describing exactly where things sit. We turned in strong results for the second quarter, stood behind our 2026 guidance, and moved the needle considerably on our large load strategy via the 1.2 gigawatt BYOP opportunity we walked through today. Back when the year began, we stated our aim was to secure no less than 1 gigawatt in large load agreements during 2026. We're pushing forward on a prospect that would get us to that goal, with the main commercial terms settled and talks along with the outstanding land items moving ahead. Should a final investment decision come, it will hinge on customary conditions, among them the internal approvals we require. As I noted at the top, the environment looks different now. What hasn't changed is our strategy. Texas has sent a clear signal that the way large load growth gets served is something that matters. Any new demand has to carry with it the power infrastructure needed to back it, make the system stronger, and prevent the investment burden from landing on families and small businesses. That stance fits neatly with the model we've established. This project is built to supply more generation than the data center is projected to need, cut down on the call for additional transmission, and put the investment burden on the customer. That's the reason we see this project as well positioned in Texas, and the reason NRG is well positioned to take the lead. Work remains in front of us; we'll keep our attention on moving the project forward, delivering across the rest of the business, and holding to the discipline that got us here. Relative to what we set out to accomplish, we've made real progress; we intend to keep our heads down and see the work through. Operator, we're ready now to open up the line for questions.
Nice work, everyone, on crossing that finish line. Well played. Rob and Gang. Yeah, for sure; so here's the question I always bring up. Great job on this, and I'm really interested in how this site might grow. What I mean is, it appears certain sites you have could scale all the way to that 2.4 gigawatt figure. What's your thinking on the schedule for getting that done? Just to be precise about the slides, there appears to be as much as 18 months separating the first COD and the second one. So how do you frame expectations for the pace of those additional 1.2 gigawatt blocks, whether at this location or another? And separately, if you could, address the returns. Is this roughly a build multiple that becomes the standard for how you view future projects? Or might they come in a bit weaker since this is the initial one and possibly the least expensive?
Yeah. There's quite a bit packed in there, Julian; thank you. I'll do my best to cover all of it. Let's begin with returns; the returns presented today for this specific project we're advancing reflect our expectation. That's the commitment we've made to shareholders. And in our discussions with customers, that's the number. Essentially, this is what it will be. Everyone will shift somewhat one way or another; but broadly, that's the return we anticipate delivering to shareholders for the capital they put to work. Regarding how to view timing and where projects land, what governs delivery on these projects are the construction CODs and the turbine deliveries themselves, correct? It hinges on what the customer prefers, where the sites will be located and when we can get the turbine on the ground, and that sets the pace we move at; however, what we've previously laid out matches what we observe throughout our pipeline since it's shaped by what comes out of our GEV agreement. So we hold those discussions with customers. And then the final point, just reacting to what you said. One thing worth keeping in mind is the 1.2 gigawatts at a site expanding to 2.4 gigawatt, that doesn't preclude doing 2.4 gigawatt at another location, that doesn't preclude 4.8 gigawatt at another location, right? When we speak with customers and consider these turbines, we have several customers seeking several turbines. The project introduced today, the one with the greatest alignment, sits at a site, right? But don't get hung up attempting to figure out where or how, because that isn't the key part. The point we want to make clear is the commercial framework we've proposed, so all of you can understand how it operates. That's also the discussion we're having with each customer about the way we set up these deals, since that's the proper approach. We're putting a lot of effort into this. It isn't finished yet, but we feel this is a key piece of information for every one of you to have.
Julian, regarding that COD item, let me just jump in. The initial one lands in late 2029. As we've mentioned before, given how the GEV and Kiewit arrangement is set up, you can figure that a further 1.2 gigawatt block will come online sequentially each year following that first 2029 COD.
So a cadence of 12 months. That's good. And now, only a few small details. To start, regarding how long the contract lasts, is that usually the — does that length serve as the standard now to match the return? And also, what's your take on what the governor of Texas announced yesterday? Again, I realize it isn't necessarily particular and relevant to this project as such, but how does that affect the schedule from your perspective?
Okay. So about the contract's duration, we've said to you folks 15 to 20 years. The structure here specifically is 15 years. We won't go below that, or I wouldn't anticipate doing so, since that would significantly alter the customer's price. Regarding the Texas governor's matters, look, I get where the politicians and regulators stand in Texas. And I attempted to convey that point in this in our — what I mentioned before in my prepared comments, our project addresses those issues, correct? It's the appropriate project to satisfy the worries of the communities and the elected officials since it doesn't burden the grid and since it can also lower the requirement for <<inaudible>> transmission in that region. When it comes to timing, Texas is a state that gets things done. I anticipate they'll sort through things to reach a better — a better comprehension of the projects to be placed down as time passes. And the final thing I'd mention is, keep in mind, this is a COD in 2029. So I believe we're fine.
Rob, could you perhaps take it up a level and walk us through the real headway being made, plus what gave you the confidence to put out the principal terms now? So I suppose, which final approvals might still be pending? And what's the expected timeframe for those?
Alright; so we'll <<inaudible>> the reason we spoke up at all. Our shareholders need to grasp the makeup of what we're going after, the approach behind how we're delivering on our GEV and Kiewit turbines. And we felt we'd reached a material point where a discussion about progress made sense, so that all of you could see where we stand. Regarding what we've said today, right, we're commercially aligned, which means that we have — they've reviewed the structure, they signed off on the structure. We're near, and in daily close talks about particular timing or a parcel of land or whatever those items happen to be. What we're still bound by is that negotiation still applies. Progress isn't finished, and we'll keep pressing until it is, and we'll keep talking with multiple customers until it is. And then naturally there are the necessary internal approvals and everything tied to that. One last point on timing. The next occasion we return to you, Shar, we're going to let you know when we have another material piece of information to discuss. I won't box myself or the negotiation team into a deadline to beat. But my feeling is very strong that we'll keep driving ahead, and I believe we'll hit our objectives for both the near term and the long term at this company.
Understood. That's great. And finally, considering a bit of the chatter we're hearing regarding collateral requirements and such. Could you perhaps go into a bit more detail about the counterparties. Clearly they're investment grade, but are they BBBs? Are they single A? Could you maybe expand a bit on how strong the credit is on the counterparty side?
As my predecessor would say, "No".
Right now I can just picture Larry shaking his head and telling you not to respond to that, Rob. Just don't respond.
That answer makes him extremely pleased. Investment grade, Shar — that's all you're getting.
We're grateful for the information regarding the BYOP transaction. I have one more question about the specifics of the contract. In the past, when you were laying out for investors how they should approach this targeted pricing, you would speak in certain terms, correct? I recall you mentioning a figure like $80-plus per megawatt hour. Considering the returns shown on that slide and the fact that CapEx costs seem somewhat higher following your latest update, has the PPA equivalent moved above $90 now, or above $100? Do you have any thoughts on that? I understand it functions somewhat as a fixed capacity charge pass-through, but what's your perspective?
Yes. Because of how it's set up, the dollars-per-megawatt-hour figure is essentially irrelevant. If someone ran the numbers and assumed a capacity — pardon me, an expected usage level for the data center, the result would land somewhere in the $85 to $90 plus range. That said, it isn't really our focus, correct? Not concentrating on it benefits the customer, and not concentrating on it benefits us too. What this structure accomplishes is giving our investors the certainty and returns they require, while also offering the customer flexibility and options for managing their own risk based on their particular outlook. So if they wished, we could hedge that variable portion, and the collateral would be their responsibility. We've incorporated as much flexibility as possible here, since from the very start we asked what our shareholders need. How do we best serve our customers, and how do we best serve the surrounding communities? That's the lens through which we've examined data centers.
Perhaps we could shift quickly over to PJM. During your walk on contracted cash flow visibility, you mentioned the possibility of doing something involving the 2 <<inaudible>> of uprates in PJM. So could you maybe give an update on how you are viewing the bilateral process or the procurement, and how should we think about that?
It is a multipronged approach, correct? There is the long-term auction opportunity, into which we will bid — and we are in bilateral discussions at this very moment as well. Anybody who wants to interconnect within PJM and avoid being subject to curtailments is going to find the extra capacity from those upgrades valuable. So we recognize that value, and we keep monitoring, and we will bid in through any process. The way to think about it is this: for this company, I am going to put capital to work in a place where long-term durable cash flows can be achieved. 15-year auction proceeds, that makes sense. And a bilateral conversation of the like term would as well.
First off, perhaps a brief follow-up to what Nick asked. When you consider the uprate possibilities, could you tell us what portion of the 2 gigawatts is, say, economic at the $555 per megawatt cap, simply to gauge the size of the opportunity on the central procurement side?
So if you — it hinges on how you read that $555 cap, that is, whether they can procure above it or not. The way I see it, Carly, is that likely under half of those 2,000 megawatts would move through that auction in that manner; but we keep having bilateral talks regarding all 2,000 megawatts.
Okay, that's really useful. Then perhaps regarding how we're deploying capital — like you mentioned in the prepared remarks, at this stage it's mostly happening through the buyback program for the year. How are you weighing the possibility of directing more capital there, especially when you consider the valuation at which the equity is currently trading?
Carly, the way I see it right now, whether we can make the program bigger will depend to some degree on where we end up on cash flow for the year. If we're carrying out this project and putting in the capital we laid out, that's where our money would be going, since ultimately we view this project as highly valuable. That said, if there's a chance to expand the program using additional cash flow, we will certainly pursue it.
What I wanted to know is whether you could go into a bit more detail regarding the funding aspect. I believe you mentioned possible capital partners. Not long ago, Williams did something along those lines, right? And what would that mean for you in terms of balance sheet flexibility and credit metrics?
Sure. So Sully, as things stand, our assumption is that we simply cover the entire thing on balance sheet, and the only real effect is that we'd said before we could reach our 3x leverage ratio in 2028. Should we take on this project and pay for it on balance sheet, that simply pushes it to 2029; yet deleveraging would still occur across that stretch, no question, even as we put money into the project. Naturally, if we go after something with a partner involved, be it the Williams-type structure or some other structure, and that generates additional capacity, then in all honesty, that likely opens the door to boosting the annual buyback program more than anything else.
And when it comes to reaching that conclusion, is the issue simply that you have to enter into additional arrangements of this type? Or does that not matter at all — it is merely discussion.
What I would say is that arriving at that conclusion really depends only on holding focused talks with prospective partners and crafting a framework that, from an economic standpoint, strikes us as highly sensible. So we absolutely plan to pursue that. Obviously, possessing the contract matters, since the partners must grasp what, in theory, they would be putting money into. And I believe we are unquestionably approaching a stage at which those talks can genuinely begin in earnest.
That's really useful. Rob, could you share your most current perspective on how pricing is behaving in the ERCOT market? My sense is folks are watching record highs, with volatility staying fairly contained this year. Yet simultaneously, plenty of people are tracking this patch process, and it looks like a substantial amount of load is on the way, though forward isn't truly responding. I'm curious what you believe is behind that price movement?
Sure. Sully, what you're getting at is that nothing is being valued by the ERCOT market at this moment. Pricing sits at low levels. It's low across the curve. There was an expectation that certain announcements regarding batch might push those curves higher; however, what markets have shown us across the past 20 years is that until something becomes real, it isn't. So factors such as worries about delays, factors such as the timing of when things actually get built, those are affecting, roughly speaking, the '27, '28 period in today's curves. Ultimately, Texas remains a market that keeps expanding. And it has a certain amount of battery plus solar development that must be absorbed through, approximately, '26 '27 and perhaps somewhat into '28. Should the data center market — or rather, 8 data center development — lose momentum, then that inflection point shifts, or is deferred. Yet the underlying principle stays the same. Over the medium term, ERCOT requires generation, since we cannot return to where we stood years back. And with the tax consequences or whatever subsidies exist for batteries and solar disappearing in, roughly, '27, that construction will shrink sharply as time passes and the market gets tighter; and keep in mind, all 500 gigawatts aren't necessary. Not even 1/3 of that is needed to genuinely tighten this market to a point where everyone will be thankful they possess generation to serve their customer loads.
I'd like to spend a bit more time on how the growth is being funded and how that ultimately shows up in your free cash flow. So essentially, looking at where we are right now, my assumption is that of the $3.2 billion in total CapEx, $1.5 billion is covered through that financing — is that correct? And that presumes 3x net debt to EBITDA for the project, right? Then, in terms of the free cash flow you'll be reporting, how does that interest actually come through? My understanding is that it's pre-growth — I'm just asking about the mechanics of how that interest gets accounted for?
Angie, when it comes to interest expense — since it's IDC, it gets capitalized. So at the end of the day, that wouldn't affect our free cash flow before growth.
So the idea is that this will essentially be a holdco with debt that amortizes, correct? When I attempted to figure out what the fully loaded return from these assets is, I don't amortize that debt — I simply factor in the interest expense.
Yeah, that seems reasonable. Just take it that a permanent capital structure tied to the project exists at 3x.
Alright. So regarding that 25% you're presenting as a subtraction from EBITDA to account for maintenance CapEx and tax — the way I understand it, the presumption holds that even by 2029 or 2030, you still wouldn't be paying cash taxes, correct?
Exactly. What we gave is essentially the steady-state, longer-term figure, which isn't meant to imply that this represents the actual cash flow figure during those initial years, when we're benefiting from the different tax shields.
All right. So approximately how does the calculation work out for maintenance CapEx — is that considered a type of asset? Would it be, for instance, $50 million annually? What kind of range are we talking about?
At this moment, we aren't going to give you that, Angie. That's something we'll share down the road.
Okay, that works. Then moving to the second point — I'm examining how the cost splits between the turbines and the EPC contract. Could you help us understand this? For instance, the EPC portion appears fairly substantial. Is it the case that — regarding the benefit from building new, does that fall mainly on the turbine side, in the sense that when you reveal more projects, there's some kind of market-driven tweak applied to the EPC part? How should we approach that?
Angie, this is Matt — here's how I'd frame it: the EPC consists of two components. One is the labor portion. It also contributes a balance of plant equipment portion. So consider the turbines as the OEM, while a significant amount of the balance of equipment is supplied through the EPC. That explains why it might appear somewhat elevated compared to what you'd expect if only the OEM were furnishing everything.
Great. Then I have one final item. So I’m grateful that you flagged Virginia coming back into RGGI as a headwind. Are there any additional headwinds, such as those below-market hedges tied to LS Power, past 2026 — say, when I’m thinking about 2027 or 2028?
Angie, the portfolio did arrive with certain hedges that ran past 2026 — not almost as many as existed in 2026, but yes, there were some hedges in 2027 that came along with the portfolio. And clearly, because of when they were put in place, they were set at a somewhat below-market level compared with where things stand now.
So you're not going to disclose which percentage? Or, like, the size of the drag?
When we release 2027 guidance on our next earnings call, we'll definitely share that information.
Rob and the whole team, congratulations on this transaction. To make what Nick asked a bit clearer, perhaps with better wording: is it reasonable to look at the return profile as $1,150 per megawatt day, which brings you to $500 million of EBITDA, and that P&L items such as O&M and fuel, plus how much you're utilizing it, are passed through and grossed up into revenue? And since projects coming later would carry more expensive turbines and EPC, is it fair to assume then that for cost of new entries, you might assume on CCGT it sits well above $1,200 a megawatt day?
Would you like to handle that?
Moses, when it comes to the particular provisions in that contract, we're simply not going to offer any specific remarks. Clearly, you've already worked through the numbers, and based on whichever way you'd like to frame the outcome, what we're looking at here is definitely a project sitting inside that 12% to 15% return range. So whatever figure your calculations produce, that's the one you ought to go with, though on that point we won't be giving any specific commentary.
Let me address that question directly, and honestly this applies across the board. Should construction costs rise as time goes on, our projected returns on your cash flow — or on what you've invested — remain the same, correct? So we're going to — and we do have transparent discussions with clients regarding this. Any agreement we finalize will sit within our 12% to 15% thresholds, without exception.
That's quite useful. Now regarding the yearly capacity fee, does it just begin right when COD happens? Or does it phase in over several years while the data center builds up its own facility along with data center usage?
It began right at COD.
Rob, Bruce, I was hoping to understand, Bruce, what gives us assurance regarding the 2026 guidance. How might things settle out, considering that prices in ERCOT are clearly softer, plus the upside you might capture in PJM.
Right. Nick, as we mentioned during the prepared remarks, with how the first half turned out, our own projection would likely put us beneath the midpoint of that guidance range. Even so, we remain confident we'll stay inside the guidance range; much of that assurance stems from how the existing fleet is hedged for the rest of the year, and it's essentially heavily hedged for the remainder of the year. Then there's how that stacks against the way we've aligned our supply with committed load, and nearly all the committed load we'd anticipated for the year has basically been obtained or arranged; that gives us the visibility we have into earnings and margins for the rest of the year.
Turning to the uprate chance in PJM — those 2 gigawatts — I recall you mentioned approximately half could be considered within the RFP. But regarding that sort of CT-to-CCGT conversion, where would — where — what sort of guideline applies if we're looking at it from a build cost standpoint versus the $2,700 figure on the greenfield side?
Cheaper and quicker.
Thanks. And I appreciate all of you being here today. We feel good about the quarter and about the progress we shared. Being careful with how your capital gets allocated will always be our approach. This update makes clear why we think BYOP is the correct approach for handling growth in large load, adding new supply together with new demand, reinforcing the grid, and shielding current customers. There's still work to do, but execution is going well, and we stay confident about the opportunity and about our capacity to build long-term value for shareholders. Thanks once more for your time and for your interest in NRG.
We appreciate everyone joining us on this call. That brings the session to a close. You're free to hang up at this point.
Read the full transcript via the API
You're viewing the first half of this call. Get the complete NRG Energy, Inc. transcript - plus 132,000+ transcripts from 7,000+ companies, speaker segments and full-text search - through the EarningsAPI REST API or hosted MCP server.
Get an API key View API docs →For developers and AI pipelines
Programmatic access to NRG Energy, Inc. earnings transcripts and 132,000+ others is available through the
EarningsAPI REST API and the hosted MCP server.
Quarterly plans from $162 - full transcripts, speaker segments, full-text search,
and the /api/v1/transcripts/recent polling endpoint for ETL pipelines.