Prepared remarks
Good afternoon, and welcome to Sunrun's Second Quarter Earnings Conference Call. Please note that this call is being recorded and that 1 hour has been allotted for the call, including the Q&A session. I will now turn the call over to Patrick Jobin, Sunrun's Investor Relations Officer. Please go ahead, sir.
Thank you, operator. Before we begin, please note that certain remarks we will make on this call constitute forward-looking statements. Though we believe these statements reflect our best judgment based on factors currently known to us, actual results may differ materially and adversely. Please refer to the company's filings with the SEC for a more inclusive discussion of risks and other factors that may cause our actual results to differ from projections made in any forward-looking statements. Please also note these statements are being made as of today, and we disclaim any obligation to update or revise them. Please note, during this earnings call, we may refer to certain non-GAAP measures, including cash generation and aggregate creation costs, which are not measures prepared in accordance with U.S. GAAP. The non-GAAP measures are being presented because we believe they provide investors with a means of evaluating and understanding how the company's management evaluates the company's operating performance.
Reconciliation of these measures can be found in our earnings press release and other investor materials available on the company's Investor Relations website. These non-GAAP measures should not be considered in isolation from, as substitutes for or superior to financial measures prepared in accordance with U.S. GAAP. On the call today are Mary Powell, Sunrun's CEO; Danny Abajian, Sunrun's CFO; and Paul Dickson, Sunrun's President and Chief Revenue Officer. The presentation is available on Sunrun's Investor Relations website along with supplemental materials. An audio replay of today's call, along with a copy of today's prepared remarks and transcript, including Q&A, will be posted to Sunrun's Investor Relations website shortly after the call. And now let me turn the call over to Mary.
Thank you, Patrick, and thank all of you for joining us today. In the second quarter, we delivered strong financial and operating results while providing a best-in-class customer experience for over 1 million Americans. We generated $1.6 billion in top line aggregate subscriber value, significantly exceeding our guidance and growing 40% year-over-year. Contracted net value creation of $376 million, which was our highest ever, more than doubled from last quarter and was also well above guidance. We generated this record profitability by growing the attachment rate of our storage offerings to an all-time high of 70% of customer additions in the period and by driving significant cost efficiencies and performance improvements across the business. We also reported the highest upfront net subscriber value quarter in the company's history, a 17 percentage point margin improvement compared to the prior year and now representing an 11% margin on contracted subscriber value.
We achieved this result by growing contracted subscriber value while reducing our installation and customer acquisition costs. The trends in our operating performance highlight the cash generation trajectory of the business as our customer origination activities are financed over the coming quarters. We are structurally generating cash. In the quarter, we generated $27 million in cash, our fifth consecutive quarter of positive cash generation. While the quarterly number is lower than our prior guidance, we are on track to meet our cash generation outlook of $200 million to $500 million for the full year. We paid down another $21 million in recourse debt in the quarter and ended with $618 million in unrestricted cash, a $13 million increase from the prior quarter. One of the things that excites me the most about our work over the last couple of years, and that really accelerated this quarter, is our definitive leading position as the nation's largest home-to-grid distributed power plant operator.
We have transformed the business to be a provider of energy resilience for homeowners and a formidable independent power producer, now with more than 3 gigawatt hours of dispatchable energy from our fleet of home batteries and nearly 8 gigawatts of solar generation capacity. Our transition to lead with storage and provide more sophisticated products and services not only differentiates us in the market but also provides tremendous energy resources that are extremely valuable to the grid. We now have nearly 200,000 storage systems installed. Over 71,000 customers have enrolled in home-to-grid programs, representing 300% year-over-year growth. These programs provided 354 megawatts of power capacity to the grid over the last year. Based on current activities and the energy capacity challenges our country faces, we are finding that our prior estimate of $2,000 or more in incremental net present value per participating customer is not only realistic; it is likely conservative.
As we continue to scale storage and provide utility-scale energy resources back to the grid, we expect a rapidly growing cash flow stream over the coming years. We expect to have more than 10 gigawatt hours of dispatchable energy online by 2029. Turning to an update on policy on Slide 8. Paul and I spent a good portion of the quarter actively engaged in Washington, D.C. and in legislative offices around the country to ensure that the work we are doing to build the nation's largest distributed power plants, driving American energy independence and dominance, is well understood. What we do is provide our customers with an opportunity to take control of their own energy future while helping Americans gain vital energy capacity by strengthening the grid. Given our rapid transition over the last couple of years, many stakeholders hadn't realized we are scaling just what the country needs: a massive customer base of Americans who become independent power producers that provide valuable dispatchable energy resources to America's grid.
While there were a few twists and turns in the process that led up to the final budget bill, the ultimate legislation is something that will encourage the continued build-out of dispatchable energy. Sunrun is well positioned to continue to generate strong financial returns under the enacted legislation. The investment tax credit for customers who purchase solar outright or finance it with a loan known as 25D will sunset at the end of 2025. Sunrun, however, primarily benefits from the commercial investment tax credit, known as 48E, as 94% of new customer additions are subscribers. The 48E credit ends starting in 2028 for the solar portion of a project but remains in place for storage through 2033. While the sunset of the 25D homeowner tax credit could lead to large declines for a segment of the market in certain geographies, Sunrun is positioned to continue to grow margins and volumes into 2026.
You can see on Slides 9 and 10 that nationally, we represent over 40% of storage installations and more than 1/3 of subscription volumes. While market dynamics will present significant growth and market share opportunities, our focus will remain on running a sustainable business with strong margins, high-quality installations, and delighted customers. We are building a business that can generate value with lower incentives. On Slide 11, we demonstrate one of many achievable paths to generating strong margins in 2028 without the solar portion of the tax credit. With conservative assumptions for pricing increases against utility rate escalation, equipment cost declines, customer acquisition cost reductions, and grid services value, we would more than offset the reduction of the solar tax credit. These items are just a subset of the value we plan to unlock in the years ahead. While we are planning for a step down in the solar portion of the tax credit in 2028, we are, of course, taking actions to lengthen our runway.
Per statute and current treasury guidance, projects that have commenced construction before July 2026 are eligible for the solar portion of the tax credit beyond 2027. In accordance with these rules, Sunrun has already commenced construction on projects or plans to soon in order to retain the full solar portion of the tax credits through 2030. Treasury guidance on the requirement to commence construction may be updated, but new guidance is not expected to be retroactive and must be consistent with legislative statute. I'll now turn the call over to Danny for the financial update and outlook.
Thank you, Mary. Turning first to the unit-level results for the quarter on Slide 13. Subscriber value increased to approximately $54,000, a 22% increase compared to the prior year, as we increased our storage attachment rate by 16 percentage points to 70%, grew our Flex deployments, and benefited from a 43% weighted average ITC level, an increase of 7 percentage points from Q2 of last year. Subscriber value reflects a 7.4% discount rate this period. We meaningfully reduced costs as well with creation costs falling 4% from the prior year. Though installation costs were approximately flat compared to the prior year, we were able to offset a 12% increase in equipment costs, driven by the jump in storage attachment rate, with a 13% improvement in non-equipment costs such as installed labor and other soft costs. We also lowered customer acquisition costs and overhead by 10% on a per-subscriber addition basis.
We accomplished these strong cost reduction outcomes while delivering high quality, maintaining strict safety standards, and embracing product and technological innovation. The higher subscriber value and lower creation costs led to a 182% year-over-year growth in net subscriber value to $17,000, the highest outcome in the company's history. Turning now to aggregate results on Slide 14. These results are the average unit margins multiplied by the number of units. First, on the top line, aggregate subscriber value was $1.6 billion in the second quarter, a 40% increase from the prior year. Aggregate creation costs were $1.1 billion, which includes all CapEx and asset origination OpEx, including overhead expenses. Excluding the expected present value from non-contracted or upside cash flows, our contracted net value creation was $376 million, an increase of $285 million from last year and about $1.64 per share.
This level of value creation reflects a net margin of approximately 26% of contracted subscriber value. Slide 15 breaks down the unit-level economics and aggregate economics on a contracted-only basis, along with the main underlying drivers for the increases. Turning now to Slide 16. Sunrun raises nonrecourse capital against the value systems we originate each period from tax equity, which monetizes the tax credits and the share of cash flows and asset-backed debt, along with receiving cash from subscribers opting for prepaid leases and from governments and utilities under incentive programs. We estimate these upfront sources of cash will be approximately $1.2 billion for subscriber additions in Q2, representing approximately 85% of the aggregate contracted subscriber value or what we call the advance rate. When we deduct our aggregate creation cost of $1.1 billion, we are left with an expected upfront net value creation of approximately $165 million.
This represents our estimate for the expected net cash to Sunrun from subscriber additions in the period after raising nonrecourse capital and receiving upfront cash from subscribers and incentive programs. This figure excludes any value from our equity position in the assets over time, including potential asset refinancing proceeds and cash flows from other sources such as grid services, repowering or renewals, or upside from Flex electricity consumption above the contracted minimum. Actual realized proceeds in the quarter were $1.3 billion, with $679 million from tax equity, $526 million from nonrecourse debt, and $82 million from customer prepayments and upfront incentives. Aggregate upfront proceeds differ from proceeds realized due to the former being an estimate for subscriber additions in the period and the latter being proceeds received against subscriber additions that may have occurred in a different period.
Cash generation, which reflects realized proceeds as opposed to aggregate upfront proceeds and is after working capital changes and parent interest expense, was $27 million in Q2. Though upfront net value creation is different from cash generation due to working capital and other items, it is a strong indicator of cash generation over time. Cash generation was impacted negatively by working capital timing in Q2. Inventory increased by $77 million from Q1, and taken together with changes to payables and receivables, this represented an investment of $45 million in Q2, as you can see on our cash flow statement. We also are continuing to see tax equity partners spend extra time digesting policy developments and changes with competitors, leading to extended timelines associated with monetizing tax credits. Turning now to Slide 19 for a brief update on our capital markets activities. Sunrun's industry-leading performance as an originator and servicer of residential solar and storage continues to provide deep access to attractively priced capital.
As of today, closed transactions and executed term sheets provide us with expected tax equity capacity to fund over 210 megawatts of projects for subscribers beyond what was deployed through the second quarter. Thus far in 2025, we have added $1.7 billion in tax equity, resulting in this strong runway. We also have $323 million in unused commitments available in our nonrecourse senior revolving warehouse loan to fund over 114 megawatts of projects for subscribers. We are underway with plans to execute multiple term-out transactions in the coming months, including a private transaction with counterparties already identified. Our strong debt capital runway has allowed us to be selective in timing term-out transactions. In July, we priced our third securitization transaction of 2025, where we refinanced a seasoned pool of residential solar systems, the $431 million securitization priced at a yield of 6.37%, in line with the yield of our prior securitization in March.
The weighted average spread of the notes was 240 basis points, which is approximately 15 basis points higher than our securitization in March. The higher spread followed overall market movements in credit spreads for similarly rated credit. Inclusive of this transaction, we have issued approximately $1.4 billion in asset-backed securitizations thus far in 2025. Though some investors are taking extra time to assess transactions, as noted earlier, asset financing markets are open and healthy, and there are an increasing number of investors, especially from private credit, who have done repeat transactions with us. We plan to continue executing both publicly placed transactions and direct placements in the private credit markets and to expand our tax credit buyer universe with more large corporations. On the parent capital side, we continue to pay down recourse debt, paying down another $21 million during the second quarter.
Since March of last year, we have paid down recourse debt by $235 million. We have also increased our unrestricted cash balance by $131 million and grown net earning assets by $2.4 billion over this time period. We expect to pay down our recourse debt by $100 million or more in 2025. Aside from the $5.5 million outstanding of our 2026 convertible notes, we have no recourse debt maturities until March 2027. Over time, we'll explore further capital allocation options to maximize shareholder value based on market conditions and our long-term outlook. Turning now to our outlook on Slide 20. We are either reiterating or raising all of our guidance for 2025. For the full year, we are reiterating our guidance for aggregate subscriber value to be between $5.7 billion and $6 billion, representing 14% growth at the midpoint. We expect contracted net value creation to be in a range of $1 billion to $1.3 billion, an increase from our prior range of $650 million to $850 million, and representing 67% growth at the midpoint.
A strong performance in the second quarter, along with continued cost efficiency improvements and value optimization is leading to improvements in our contracted net value creation outlook for the year. We are reiterating our cash generation guidance for the year of $200 million to $500 million. This reflects the strong operating performance along with the increased working capital investments. For the third quarter, we expect aggregate subscriber value to be approximately $1.5 billion to $1.6 billion, representing 8% growth at the midpoint, and contracted net value creation to be between $275 million and $375 million, representing 58% growth at the midpoint. We expect cash generation to be between $50 million and $100 million.
Questions and answers
The first question comes from Moses Sutton with BNP Paribas.
Congrats on quite an impressive update on really everything. On Slide 11, where you detailed the start of construction, if 3 years will have started construction by mid-August, is that like saying extending through all of '28 and '29 because you don't have the safe harbor for '26 and '27? And if so, why on that same slide would you show the bridge above the $6,500 bridge, you have all that done by 2028? It seems to me like you're properly protecting the solar ITC through year-end '29 and that you'll do 2030 in 1H '26, notwithstanding the new rules?
Right. I think, yes, two different things. On the safe harbor, we are articulating that there's a Section 48 of a few months. There's Section 48 safe harbor of a few months, and then there's 3 years that you noted from the slide there. And you're correctly noting that, that is extending the runway by a few years beyond 2028. And on the top portion of that slide, we're just doing a very simple walk of the loss in margin from the loss of the solar portion of the ITC. You could think of that as 2028 or even a different year, but really to show the buildup back up to full recovery, if not a little bit more than full recovery of that, about $6,000 per system loss in ITC value. And we've shown a few of the factors. Not in here is other cost reduction in the business. So operating cost efficiencies, servicing cost efficiencies, everything else we'll do on top of this. This was just to show one possible path, just to illustrate that we think it's achievable to recover that solar portion. And of course, we're complementing that prudently with a safe harbor strategy as well.
Got it. That's what I thought. Very helpful. And then on the cash generation being reiterated, the guidance for the year, is that net of the safe harbor spend that you've done a little bit already and then finishing through August? Or will you be like netting that out when you do the cash generation math at the end of the year? Will it be excluding what you spent on safe harbor? How should we think of that? I don't know but that's what I'm thinking.
Yes. We've reflected the working capital effect we noted in Q2. We've also reflected expectations for the balance of the year.
The next question comes from Brian Lee with Goldman Sachs.
Kudos on the nice quarter and execution. I guess first question kind of related to Moses' question. You had the big uptick in net value creation. So one, I was wondering if you could maybe quantify the different buckets to what's driving that significant increase in the view for the year? And then kind of related to that, how come that's not really translating to anything on the cash generation outlook for the year, maybe some of the puts and takes between what drove the uptick in net value creation, but why that's not translating at least this year into cash generation?
Certainly. I'll elaborate on that. The growth in volume played a significant role. Looking at a year-over-year perspective helps explain the 17 points of unit level margin expansion. Approximately 7 to 8 points of that margin expansion came from the increase in the weighted average ITC due to more additions in the business. The battery attachment rate also contributed to the overall results, and the robust year-over-year growth in subscriber additions was beneficial as well. Overall, the volume has shifted towards a higher value mix. Additionally, while operating cost efficiencies show a 4% year-on-year reduction at first glance, when we dig deeper, we see that the increased consumption of materials was absorbed by reductions in operations and sales and marketing costs. This ability to maintain stable creation costs year-over-year while boosting revenues led to significant margin expansion.
In Q2, we also observed some changes in our inventory balance and working capital, along with a slight increase in capital market activity as people adjust to the budget bill. The summer period may have also played a role. As we continue tax planning for the remainder of the year, we expect that activity to ramp up. Thus, when we combine the sequential growth in volume with installation activities and the usual cash conversion cycle involving tax equity and turnout, we anticipate stronger cash generation in the latter half of the year.
Okay. Fair enough. That's helpful color. And then just on the safe harboring, presumably, you're focused on meeting the 5% threshold to satisfy the commenced construction criteria. Can you just give us a sense of what the kind of working capital financing needs will be between now and sort of, I guess, July 2026, kind of quantify that for us? And then where you are with respect to discussions with lenders on securing what percent have you secured, what percent do you have a line of sight to secure?
I'll address the first part. I may need to clarify the question about lenders, but we were quite active at the end of last year, which accounts for several months' worth of activity. We also discussed additional activity in the last quarter. The impacts are reflected in our inventory balance and some changes in working capital. We are committed to a capital-light strategy for safe harbor initiatives and will continue in that direction. The outcome of the executive order will affect the timing of further safe harbor projects. We anticipate that any changes will apply going forward rather than retroactively, and we've already made substantial progress in recent years. Regarding your question about financing, I'm not sure I completely understand, but we've managed to pursue this in a very capital-efficient way.
The next question comes from Praneeth Satish with Wells Fargo.
Going back to the safe harboring question, in terms of quantifying how much cash you've spent so far in Q2 and into Q3, we might consider the inventory change as a proxy. I believe that was around $70 million to $80 million. You mentioned that by mid-August, you'll have enough safe harbor to cover the current volume through 2030. Is that where you plan to stop, or will you continue to seek safe harbor to account for expected volume growth and create a buffer?
I think we did note that there's about 3 years' worth of activity. There's a 1-year period from the date of bill passage to supplement with more safe harbor activity, and we would plan to do more. The details of that will very much depend on the treasury guidance that comes out following the executive order. So we still need some more time to firm up thoughts on how much more, but we can comment by saying we do intend to do more. It's just kind of a matter of the details with what qualifies and what that strategy looks like.
I would like to revisit the grid services. I understand you're estimating a minimum of $2,000 net present value per customer. Can you clarify how much you are currently generating on a recurring revenue basis? I believe we are estimating around $20 million per year from your enrolled customers. Is that accurate? Additionally, at what threshold do you plan to securitize that revenue stream? Would $50 million be sufficient, or is a higher amount necessary? Based on your guidance of reaching 10 gigawatt hours by 2029, it seems you could achieve $50 million or $60 million in recurring revenue by then. I'm trying to understand the potential for securitizations in that scenario.
I will start by mentioning that we currently have around 3.2 gigawatts of storage devices, which translates to roughly 200,000 units in the market. About 35% of these are enrolled in programs, which means over 70,000 of our battery devices are part of active and paying programs. Our strategy going forward is focused on deploying battery devices and increasing our presence in areas that have favorable home-to-grid programs. The current enrollment rate is 35%, and we expect it to rise significantly. We have set a goal of achieving 10 gigawatts, which we believe is very attainable. A $20 million target seems conservative but reasonable. We're currently exceeding the $2,000 per subscriber benchmark and aim to keep doing so. As we expand our capacity and enhance our programs, we anticipate greater stability that will facilitate financing, and I will let Danny address that in more detail.
Yes. And then, latching on to the $2,000 per customer, a vision and a realistic path to tripling the deployed capacity over the next few years. We start to push in the several hundred thousand customer range for participating customers with their batteries. So on a present value basis, that starts to look very meaningful in terms of scale, several hundred million dollars of total present value. As far as the financing outlook for that, I don't think we've concluded yet. I think we're happy with the way the scale is building. It's a new bank financing activity, certainly subscale. But as it scales, those options will open up and become very obvious to us.
The next question comes from Joe Osha with Guggenheim.
I have two questions for you. First, as we consider the market after 25D ends, is there any part of the cash and loan market that could transition into leases or PPAs, or should we just expect it to disappear? I'm trying to figure out how to approach this. Then I have a follow-up.
Yes, great question. Kind of independent research consensus suggests half the market is that cash flow today and about half of that goes away. So a 25% pullback in the market overall. And we think that is quite reasonable. So if 25% of that market largely is areas where third-party owned isn't a viable solution. So most of that market, I think, goes away. The sales and fulfillment partners that operate in third-party owned markets that are today selling under 25D, I do think most of them make a reasonably successful transition over into a third-party owned model with the solutions available to us or to them. I think for us, there maintains a really heavy focus on quality, margin control. And I think most of the players who are a bit more focused on the superior consumer offering that third party has been for quite some time are focused around and understanding the value of dispatchable controllable load have already migrated into a third-party model and are working with us.
And so while I do see some migration of the 25D volume flowing into Sunrun, I think there's a lot of development and maturity for many of the partners that would need to take place, even simple things like we deploy domestically manufactured equipment and are focused on that. So having these organizations build out the supply chain and the sophistication to track and report on that is necessary for us. I think there's some development that would need to take place. But I think overall, a 25% contraction in the overall market, some of that flowing to Sunrun will be natural.
Yes, I agree. I'm curious if you have made any efforts to reach out to the existing installed base to see if you can enroll nonparticipants, especially since I heard that about one-third of your current storage fleet is signed up for grid services.
Are you talking about signing them up to install storage with them or going back to...
That's because of the programs and the geographies, Joe. It's not like all of our customers are technically capable of participating in programs. Again, it's more about the geographies where we are already seeing a lot of grid constraints and challenges, and we expect to see that grow. And then...
Yes, I wasn't trying to be negative. I'm just wondering whether there's some opportunity to go back and try and remarket that existing fleet of storage. That was my question.
100%. Yes, absolutely. As is we also see tremendous opportunities with our existing customer base, and we've started to do that successfully, which is scale in a super low CAC way, storage attachment for the 800,000 customers that we have that don't currently have storage so then we can enroll them in these programs.
The next question comes from David Arcaro with Morgan Stanley.
Could you elaborate on how actively you are pursuing cost savings and efficiencies at this time? What is the status of your efforts to reduce customer acquisition costs? There has been good progress on creation costs for the quarter, but I'm curious about how much of that is due to your aggressive approach in this area. Is this aspect part of your near-term strategy?
Yes. I would say we have had a laser focus on running incredibly efficiently and effectively for our customers for some time, and it's really gaining a lot of traction, particularly as we've talked about onboarding a really strong AI team that has helped to accelerate that throughout the business. So we still see opportunity for years to come in the context of continuing to drive down cost in the business and drive down the cost of customer acquisition. And so yes, we not only have achieved a lot of that but we see a lot of continued opportunity as a business.
Yes. The only additional call-out I'll make is on post-installation servicing costs. We've seen those come down quite a bit over the last year as well. And that's been a key focus on top of installation and customer acquisition cost efficiency. On that side, we are also seeing as we kind of scale up volume, we're seeing operating cost leverage across our fixed cost base in a very, very meaningful and nice way.
The next question comes from Jon Windham with UBS.
First of all, I'm not usually one to dole out compliments on quarterly calls, but over $5,700 on the upfront net subscriber value is a massive number. So good work by the team. I wanted to ask a question that maybe has been an area where people aren't looking these days. We've obviously had so much focus on the federal level. Could I give you the opportunity maybe to walk around the room a little bit and talk about anything you have an eye on in terms of state-level subsidies and policy programs? Historically, in this industry, when the federal government wants to go sort of one way under one administration, the opposite states tend to go in another. Meaning like the last time we saw Trump elected saw some of the more democratic states sort of double down on their commitment to renewables. Just if you could help us on anything we should be keeping an eye on, investors should be keeping an eye on at the state level over the next 6 or 12 months, would really appreciate just your high-level thoughts.
I think like an overall perspective, you articulated a dynamic we've seen play out and to be true multiple times and continue to be the right way to think about it. Overall, there are and have been long-standing durable rebate programs that we continue to see as strong stable elements, like the SREC programs that exist for example, or other state programs that have dedicated funding and continued stability. And we've seen, I think, in some areas, some enhancements there and some further commitments there. And then I would say early conversations are kind of budding in several new markets that create interesting and exciting opportunities for us.
Yes. The only thing I would add to that is that as we all know from reading the news every day, the grid is running into challenges. And now with also the AI challenge and the AI race with China, you're seeing a lot more focus on energy capacity. So not only do I expect to see some of the same phenomenon we've seen in the past, but I think someone is in a particularly athletic position because I think you're going to see a real focus on storage as a part of that. So as states are looking at that exact topic of how do we make sure we're accelerating it in our jurisdiction, I think there is also a very strong focus on dispatchable resources, which again plays very well into our vision and our work in the energy space.
The next question comes from Dylan Nassano with Wolfe Research.
I just want to go back to the Slide 11, where you show what a post-ITC world could look like. So my question is, how can the industry kind of drive down those customer acquisition costs without significantly lowering volumes? It just seems like there's a view in the industry that sales commission payments are kind of like the lifeblood of volume growth. So to me, it seems like a little bit of a chicken and egg problem. I'm curious how you think you could kind of achieve both.
I think we've programmatically been focused on innovation and creating a differentiated consumer offering. And as we've done that, CAC as a percent of proceeds or value created has been programmatically falling. And we're in, I would say, the early stages of that innovation cycle and building out that differentiated offering. So our sales reps are focused more on selling consumers features, functionality, and benefit. Layer on top of that revenue that we were talking about with home-to-grid capabilities around virtual power plants and that becoming a more substantial thing as the grid feels the impacts of the electron shortage, the price you'll be able to fetch to be sitting on currently a little over 3 gigawatts of dispatchable capabilities soon to be 10 gigawatts over the next few years becomes very meaningful and a real source of incremental top line revenue with no real associated CAC to speak of burdening it. And so as those kind of dynamics play out, we have been seeing and expect that trend to continue to walk us into CAC reductions in line with what's shown on Slide 11.
Got it. And then just as a follow-up. So I guess just given that residential solar is generally shorter cycle than utility scale solar, can you talk a little bit about how you plan to balance safe harbor-related equipment acquisitions with a little bit more limited forward demand visibility? Just trying to think through some potential headwinds like stuff like equipment degradation. And then separately, kind of on that point, are you looking to further diversify your supply chain mix kind of maybe more towards some historically less represented suppliers?
I think the safe harbor activity and the available capacity to engage in it with our current vendor mix have proven to be quite manageable. I wouldn't say that safe harbor necessitates a significant change in our vendor selection. It may affect the distribution among our vendors, but it won't lead to a drastic shift. Regarding the question about the multi-year runway, it's really a combination of activities accumulating over several years. We want to see the executive order fully implemented and complete our strategy before diving into specifics.
The next question comes from Hannah Velasquez with Jefferies.
This is Hannah Velasquez on for Julien. Like everyone else has said, congrats on the quarter. So I just wanted to open up with a clarifying question. It sounds like of the 3-year safe harbor program that you all are pursuing, you safe harbored about 1 year's worth post the One Big Bill to date. And then you're waiting, I assume, until treasury issues guidance for the remaining 2 years. So a clarifying question there. Is that correct? And the follow-up there is, what gives you confidence that no portion of guidance will be retroactive? I'm sure you're hearing all the whisperings that it could be retroactive back to 7/4 or 7/7 or even retroactive back to January 1?
Yes, I can address the retroactive comment first. This is Mary. From what we've gathered during my time in Washington, the chances of retroactivity are extremely low. We've seen a lot of Republican Senate leadership indicating a desire for minimal changes, particularly not retroactive ones. Everything we are hearing suggests that there will likely be some changes, but they will be prospective. Danny, why don’t you take it from here?
Yes. The end of year activity was about 12 months of solar equipment, 6 months of battery. So just to bridge that to the bullet point you see at the bottom of Slide 11, 2 months of Section 48, that implies the equipment that is available from that safe harbor activity. There's 6 months that have commenced construction in the calendar year before the passage of the bill. So that's the 6 months. And then there's 3 years expecting to have commenced by this August. So that's fully incremental to the first two. And then what we're talking about further to that is doing more than the 3 years by July 2026, which is the 1-year deadline from the data bill passage, and that's still in flight. Again, that's the part that's more TBD based on the treasury guidance.
Okay. And then just as a follow-up on the battery side. When you talk about potentially getting to 10 gigawatt hours by 2029, what macro and market level assumptions are you considering to underwrite that number? Does it assume other states adopting maybe VPP programs across utilities or expanding across the current base of states where batteries are particularly strong? And then maybe you could also just identify which of the key states currently you feel will drive a bulk of this ramp to 10 gigawatt hours through 2029?
Yes. So the state participation in grid service programs would just be the enrollment rate and the economics we would drive off of that 10 gigawatts. But the batteries absent grid service programs are highly economical and attractive for us to be selling and create a great grid resource and good customer functionality. So building the 10 gigawatt base is attractive and independent of program enrollment. As we said, we're at 3.2 gigs right now, and I think a kind of steady-state mix or consistent state mix of what you've been seeing. We're at about 70% attach rate today. So that puts us at about 70,000 new batteries installed on an annual basis, that puts us not far off with little acceleration needed to be walking right into 10 gigawatts through 2028.
The next question comes from Maheep Mandloi with Mizuho.
Congratulations on the quarter here. I just want to clarify one thing on the safe harbor through '28 or even beyond that. Like what growth rates are we assuming here for that? Is this $5.76 billion per year kind of run rate or slightly higher than that for those years?
Just to clarify, the growth rate, is that the growth rate we're assuming for our business when we sign the safe harbor activity?
Yes, I believe only part of that appears in inventory. The rest is represented as...
Modestly growing, I would say, is the expectation. Obviously, we've talked about the industry dynamics with 25D and the opportunity we have there. I think also we're being very disciplined on the volume growth. I think we're just anchored, as we've said for many quarters, on the margin growth. I think that has us with a view of modest growth and the resulting multiple year runway comment when we're talking about the magnitude of the safe harbor activity is underpinned by that assumption.
The next question comes from Colin Rusch with Oppenheimer.
Given some of the distress that's out there in the market, are you seeing any opportunities to pick up portfolios of assets that you could optimize and leverage with kind of add-on sales, especially given what you're doing with the balance sheet right now?
That's definitely an opportunity that exists out in the market. It's not something that we're pursuing. I think there's an opportunity for us to do more with battery deployments and growing with and optimizing our own fleet, and we're really focused on that for the time being.
That's super helpful. On the battery side, with recent innovations in battery chemistry and longer lifetimes, are you observing significant advancements in new chemistries that you could begin to implement, potentially enhancing pricing and margins through these incremental chemistry improvements?
Yes, one of the benefits of being the largest deployer of residential batteries in the country is you get the first pick and you get insights into what anybody is doing. So we do look at a lot of technologies. There's some incredibly impressive innovators out there, and we're excited about the prospects of continuing to develop around cheaper batteries, more functionality, better installability. There's a lot of innovation taking place in that space. And we think we're seeing it both from our existing partners as well as new emerging opportunities.
The next question comes from Philip Shen with ROTH Capital.
First one is a follow-up on retroactivity. I know, Mary, you talked about it being an extremely low likelihood. We're hearing similar, that they're not leaning that way. That said, there is still a probability out there that's non-zero. And so what happens if there is retroactivity? What's the path? And then in terms of what you guys have safe harbored thus far, how have you guys done that? Is it through inverters, batteries, modules? What's the dominant category of equipment?
Yes. It's great to speak with you. The key point to remember regarding Sunrun is that it pertains specifically to claiming the ITC for solar from '28 to '30. As shown in our slide, we believe Sunrun can achieve attractive returns even without the solar ITC. This is a crucial aspect to understand from a foundational perspective. Additionally, given the very low likelihood of retroactivity that we've all been hearing, we feel we are in a strong position.
Vendor equipment type mix, kind of all of the above. I think we'll reserve comment on exact specifics of our strategy.
Okay. And then shifting to my second question here on tax credit buyers and buying some of our checks suggest a substantial percentage of the tax credit buyers out there are not interested in resi solar as a category in general. So then you have just less demand, right? And there's a fair amount of supply, especially with the new entrants, notwithstanding. Pricing appears to be challenged with some of these resi players. We're hearing kind of in the, call it, mid- to high $0.80 on the dollar of credit. I got to imagine the pricing for you guys is better. But to what degree has that overall kind of challenged sentiment for resi solar impacted your pricing for tax credits? Have you seen a meaningful change that could impact margins despite your strong margins in the quarter, but could it impact margins ahead?
Yes. I believe it's a significant market. We have raised $1.7 billion year-to-date. We noted during the call that buyers slowed down a bit due to the various versions of the bill playing out, particularly in the summer. Some of this slowdown was driven by the residential sector, as buyers wanted to understand the differentiation among sponsors in the space. This situation is similar to what we observe in the debt capital markets during our securitization transactions. In a time like this, where we're presenting our results, we encourage capital providers across all markets to carefully assess the unit margins from the originators. We take pride in showcasing how robust those margins are, reflecting the results of our efforts to ensure this. This is an important credit factor for anyone evaluating an asset deal while trying to understand a sponsor’s unique aspects compared to other options they may encounter.
In contrasting residential with utility scale, there are distinct considerations for utility scale, such as a very long and uncertain deployment window and the overall project completion rate, which can also be unpredictable. Our business model stands out because there are not many with a consistently reliable flow rate, which we observe in the residential sector. That's a compelling aspect. There are pricing discrepancies between residential and utility sectors, but we believe they will converge over time. Currently, there might be some discussions around specific price points, but such fluctuations seem more short-term and do not indicate any significant long-term change in the expected economics.
The next question comes from Chris Dendrinos with RBC.
I wanted to follow up on the 25D conversation and the opportunity to capture some of that, I guess, call it, non-TPO market that might come to TPO. You mentioned you're not looking to employ a long tail or a roll-up strategy. What kind of is the strategy here? Do you do anything different, I guess, from this point forward in terms of onboarding new installers? Are you looking to add new installers? Or how are you thinking about making sure that you're benefiting from some of this opportunity here?
Yes. We have really successful and great relationships with our affiliate partner business and do see some opportunity to onboard people in the 25D space to join that affiliate partner network. Just calling out that there are some limitations given our strict underwriting, quality, safety controls, and things like that. So there are strict parameters that we think limit out much of that market. That being said, there are some fantastic players in that space that we see could be and likely will become good Sunrun partners. Beyond that, we do see some companies seeing this as an opportunity to kind of go to safer harbors, and we've seen the onboarding and actually organizations come into Sunrun and become Sunrun employees joining our sales force or our installation labor force. And so we see that taking place as well. Those will be the two mechanisms and interaction points between Sunrun and the 25D market.
Got it. And then I guess maybe separately on a follow-up. Tariff impacts last quarter, you had highlighted, I want to say it was maybe $1,000 or $1,300 per customer from tariffs. How has that changed, I guess, today, just given fluctuations in tariff rates?
Yes. The short answer is that we've moderated and ended up at the low end of the range. We've reflected all the associated impacts in our forecast and discussed them in the guidance we've provided.
Thank you. This does conclude the allotted time for today's teleconference. We thank you for your participation. You may disconnect your lines at this time, and have a great day.