Brookfield Infrastructure Partners L.P. Stock

Good day, and thank you for standing by. Welcome to the Brookfield Infrastructure Partners L.P. Second Quarter 2026 Results Conference Call and webcast. [Operator Instructions] Please be advised that today's conference is being recorded. I would now like to hand the conference over to your speaker today, David Krant, Chief Financial Officer. Please go ahead.

Thank you, Kristal, and good morning, everyone. Welcome to Brookfield Infrastructure Partners Second Quarter 2026 Earnings Conference Call. As introduced, my name is David Krant, and I'm the Chief Financial Officer of Brookfield Infrastructure. I'm joined today by our Chief Executive Officer, Sam Pollock; our Chief Operating Officer, Ben Vaughan; as well as Dave Joynt, a Managing Partner on our investments team; and Lief Williams, the Managing Director focused on AI infrastructure investments.

I'll begin the call today with a discussion of our second quarter 2026 financial and operating results, followed by an update on our asset sale initiatives. I'll then turn the call over to Sam, who will discuss our new investments and provide an outlook for the business. At this time, I would like to remind you that in our remarks today, we may make forward-looking statements. These statements are subject to known and unknown risks, and future results may differ materially. For further information on known risk factors, I would encourage you all to review our latest annual report on Form 20-F, which is available on our website.

We're pleased to report that in addition to Brookfield Infrastructure delivering strong financial results this quarter, we have also made meaningful progress across our strategic initiatives. Beginning with our financial and operating results. In the second quarter, we generated FFO of $702 million or $0.89 per unit. This represents a 10% increase compared to the prior year on both a quarterly and year-to-date basis, which is in line with our long-term growth target.

The increase reflects organic growth within our 6% to 9% target range, driven by inflation-linked rate increases in our utility segment, strong activity levels across our transport and midstream businesses and the commissioning of new capital projects in our data segment. Results also benefited from the strong cash contribution from new investments, which are generating returns meaningfully above the yield on our completed asset sales.

I'll now go through our results by segment in more detail. Starting with our Utilities segment, we generated FFO of $196 million, an increase of 5% versus the prior year. The increase was driven by inflation indexation, the contribution from capital commissioned into our rate base and the acquisition of a South Korean industrial gas business completed last year. This growth was partially offset by foregone earnings from asset sales, including the largest of 4 concessions within our Brazilian electricity transmission operation and our Mexican regulated natural gas transmission business, both of which contributed to results in the comparable period.

Moving on to our transport segment. FFO was $311 million, representing a 7% increase over the prior year after normalizing for capital recycling activity. The increase was driven by broad-based strength across our operations with volumes across our rail, port and toll road operations increasing between 3% and 7% year-over-year.

In addition, results benefited from the contribution from our North American railcar leasing platform, which closed on January 1. These contributions were partially offset by the foregone earnings associated with the sale of a 49% interest in our Australian export terminal, the sale of our Australian container terminal business and a partial sale of our U.K. port operation, all of which closed last year.

Our midstream segment generated FFO of $183 million, up 17% compared to the same period last year. The increase reflected strong organic growth across the portfolio, particularly at our Canadian diversified midstream business, which benefited from strong asset utilization as well as elevated commodity pricing. Results also benefited from the contribution of our recently acquired U.S. refined products pipeline system, which more than offset the lost earnings with the sale of our U.S. gas pipeline last year.

Lastly, FFO from our data segment was $154 million, representing an increase of 36% compared to the prior year. The increase was driven by the contribution from our U.S. bulk fiber network acquired last September as well as income generated by our data center developers and the initial contribution from our partnership with Intel to construct semiconductor foundries in Arizona.

Turning to our balance sheet and capital recycling program. Public markets have been increasingly effective exit channel for us. So far in 2026, we have generated meaningful proceeds from public market transactions, reflecting both the quality of the businesses we have built and the depth of investor demand for scaled, high-quality infrastructure platforms.

IPOs and follow-on public market monetizations provide us with an attractive path to crystallize value, broaden the buyer universe and retain flexibility to participate in future upside. They also give us optionality alongside private sale alternatives, supporting value maximization across multiple potential exit paths.

The most recent example was the IPO of our U.S. colocation data center operation on the New York Stock Exchange. Since our initial investment in 2018, we have transformed the business into a scaled platform with large presence across major U.S. markets, serving more than 1,700 customers. A key value driver in this transformation was the acquisition of over 40 sites through a bankruptcy process, which scaled the platform, optimized the portfolio and accelerated its growth.

During our ownership, we have increased EBITDA by more than 4x and expanded capacity from 115 megawatts to approximately 390 megawatts. The IPO represents the next step in our value creation plan. The transaction generated gross proceeds of approximately $1.2 billion at an attractive valuation. Brookfield retains a 64% ownership interest in the business and will continue to participate in future value creation, including the potential to grow the platform to approximately 1 gigawatt of capacity through equipment optimization and under-roof expansion.

In the quarter, we also advanced monetizations across 2 listed businesses in India. At our Indian telecom tower portfolio, we sold a 7% interest through the capital markets. Similarly, at our Indian gas transmission operation, we completed several smaller sell-downs to public market investors following our inaugural sale last year, exiting a further 14% of the business. Combined, these transactions generated nearly $100 million of proceeds net to BIP.

Adding to our asset sale progress, we executed a second transaction under our established framework for monetizing derisked and contracted container portfolios at our global intermodal logistics operations. On July 1, we completed the sale of a majority interest in a portfolio of contracted containers, generating approximately $60 million to BIP.

Finally, at our North American railcar leasing platform, we generated approximately $20 million in proceeds at our share. These proceeds were primarily generated through our structured investment framework, which provides for the transfer of ownership to our partner over time.

Now, together, these transactions further support our ability to self-fund growth while recycling capital at attractive valuations. So far, in 2026, we have generated nearly $1.2 billion of proceeds from our asset sales with several sale processes well underway that give us confidence in achieving our capital recycling objective for this year. That concludes my remarks this morning, and I'll now turn the call over to Sam.

All right. Thank you, David, and good morning, everyone. The first half of the year was active on both sides of our asset rotation strategy. In addition to the asset sales David just discussed, we have secured or deployed over $800 million into new investments. This includes the acquisition of Clarus, New Zealand's leading gas infrastructure utility with closing expected in the coming weeks and an increased equity commitment to the Bloom Energy framework to support an additional CapEx project.

Looking beyond the projects already secured, momentum in AI infrastructure is accelerating with our AI factory strategy gaining traction globally and expanding our pipeline of investment opportunities. In the U.S., Brookfield was selected by the Department of Energy to develop an AI data center campus in Kentucky designed to support over 1.2 gigawatts of compute capacity. We have formed a consortium to advance the project through a bring your own power model.

In South Korea, Brookfield, NAVER and NVIDIA announced plans to develop 200 megawatts of sovereign compute capacity. Under the proposed arrangement, Brookfield would act as the exclusive capital partner to finance the deployment of GPUs at the campus, supporting one of South Korea's largest planned sovereign compute developments.

We also expanded our framework with Bloom fivefold from $5 billion to $25 billion of total CapEx, creating a significant pipeline of future deployment opportunities for behind-the-meter power solutions for leading hyperscale customers. Together, these initiatives demonstrate the breadth of our AI infrastructure opportunity set and our ability to originate large-scale projects on a bilateral basis by combining our digital infrastructure and power expertise with flexible capital at scale to support leading energy and technology partners globally.

As these opportunities progress, we will only commit material capital once appropriate commercial arrangements are secured and our risk-adjusted return objectives are met. With a broader opportunity set in front of us, converting our growing pipeline to capital deployment is a key focus for the balance of the year. We are advancing opportunities across sectors and geographies through traditional M&A and strategic capital partnerships, where leading companies are seeking long-duration capital at scale and an aligned operating partner.

Together, these channels provide multiple avenues to deploy capital into high-quality opportunities at attractive risk-adjusted returns. One of our strategic initiatives for the year is to complete the recently announced corporate simplification to convert BIP and BIPC into a single publicly traded corporation, Brookfield Infrastructure Partners, Inc. We believe the simplified structure will provide improved trading liquidity, increased demand from index funds and ETFs and broader access to investors who prefer a traditional corporate structure, among other benefits.

We expect the simplification to be tax-deferred for Canadian and U.S. investors and completed without any meaningful cost to the business. Special meetings of BIP unitholders and BIPC shareholders will be held on October 14, and we anticipate completing the simplification in the fourth quarter of 2026. Ultimately, we expect this simplification to drive long-term value for all security holders.

In closing, we entered the second half of 2026 from a position of strength. Resilient operating performance, a healthy balance sheet and meaningful proceeds from recent asset sales provide significant flexibility to pursue attractive growth opportunities.

This concludes my remarks, and I'll pass it back over to the operator, Kristal, to open the line for Q&A.

[Operator Instructions] And our first question will come from Cherilyn Radbourne from TD Cowen.

So clearly, the market has become more anxious about the CapEx going into AI and data centers and the timing of the payoff. So I was hoping you could speak to the opportunity set and just how BIP is able to maintain its investment guardrails against that backdrop? And maybe you could touch on whether you're seeing degradation in contract terms more broadly across that space.

Cherilyn, maybe I'll start off and then I can ask Lief Williams, who's with our AI infrastructure group to add further color. Maybe just to begin with, as far as the momentum in the sector and the demand signals that we're seeing, we've definitely seen no reduction in the developments underway or the speed to which our clients are looking to bring forward projects.

So while capital markets have obviously pulled back in the last couple of weeks, customers -- and our customers are the largest hyperscalers in the world are obviously thinking about longer-term trends as opposed to short-term gyrations. I would say the -- your question regarding degradation of contracts, we've always told our investors that we will only deal with the highest quality customers and invest in those projects where we have, as I mentioned earlier, proper risk-adjusted returns.

And the main guardrail, to be honest, is the fact that all these projects require a significant amount of debt capital. And in order to source that debt capital, you need to have highly high-quality counterparties. And if you don't, then you're not going to be able to raise the equity capital, to be honest.

And so while there might be some smaller projects that others might be pursuing where they're taking on lesser quality counterparties. In our case, we're only dealing with the best, and we're not seeing any degradation in terms. Maybe since we have Lief on the line, Lief, do you want to talk about any trends that you're seeing as far as new developments?

Yes. Thanks, Sam, and thanks for the question, Cherilyn. I think from a commercial terms perspective, I think as Sam said, we continue to see strong contracts from our customers. I think in terms of development yields, I would say it's still kind of high single digits, low double digits. I think that you see that move a little bit with interest rates.

And so we are in a slightly higher interest rate environment than maybe in the past. And I think you see that, that ultimately flows into development yields and, as well as, the annual escalator. And so again, whereas historically, that's fluctuated between 2% to 3%. I think right now, you're seeing that really at the higher end of that range. So really more 2.5% to 3%.

And then the last key commercial term I would highlight is on lease term. And so again, typically, the focus for greenfield projects is 15 years plus. And we are starting to see customers who are open to a 20-year initial lease term. And again, from our perspective, that is -- that's a crucial input to developer returns. And so overall, I would just characterize it as a strong market on the private side. We see very good demand, and we think that it's a great opportunity to deploy capital at attractive risk-adjusted returns.

Just to make this call not all about data, I thought I would ask about where else you're seeing opportunities outside of data. One area where we're seeing some sort of news and potential activity is industrial carve-outs with resource companies looking to sort of focus on core operations and carve-out utilities and things of that nature. Are you seeing that in your pipeline as well?

Yes. Maybe just to touch on the first part of your question, and I can come back to maybe carve-outs. But you're asking where we're seeing knock-on effects in other parts of our business. And one area where you might not expect there to be a lot of impact is in our transportation business, where we're seeing a lot of -- what we always refer to as the domino effect of all these developments are requiring products and assets from different parts of the world. And so we're seeing that reflected in trade flows. And so maybe Dave Joynt, who runs our transportation business, we have them on the line here. He can talk a bit about what we're seeing through Triton from a transportation perspective.

Yes. Thanks, Sam, and thanks, Cherilyn, for the question. The -- overall, I think you've seen a very strong quarter for us on transportation. But what might be a little bit hidden by that is that a lot of that strong demand is actually coming from the big build-out of data centers themselves. And so if you look at Chinese exports on a year-to-date basis, it's up nearly 20%. And what is underneath that is machinery and motors and transformers and pumps and valves and tubing that goes into lots of the machinery that goes into the complex itself. And that is flowing through certainly a very strong demand environment for our container leasing business, but also for our ports and our rails on a global basis.

Yes. And maybe just to answer your last question, and maybe we'll keep it short, but we are definitely focused on strategic partnerships and carve-outs in a number of sectors. That's something that worked well on the railcar leasing side that we recently did. And we're seeing a number of industrial companies looking to take advantage of capital available from the infrastructure players like ourselves to source low-cost capital to grow their operations. And so that is a focus. And hopefully, some of the transactions we'll announce in the coming quarters will demonstrate that.

Our next question will come from Devin Dodge from BMO Capital Markets.

It seems like the AI factory strategy is really starting to gain traction here. It's obviously great to see. And it seems like there -- based on your comments, Sam, there's still a lot of irons in the fire. Just wondering if you can frame how large of an opportunity the AI factory strategy could be over time? And maybe just for the projects and frameworks that you've secured to date, just any thoughts on potential equity commitments or deployment timing from a BIP perspective?

Okay. Well, again, I might ask Lief in a second to just comment on some of the initiatives we have globally. Some are still in the early stage, and so we can't get into too much specifics on them. But I'd say we've been busy developing a number of them for the past year. And to the extent that they are sovereign AI factories, those tend to take a bit of time just because of the nature of dealing with governments.

I think the -- on the potential deployment, I think if we look out over a longer-term time frame on a 3- to 5-year time frame, I think we see the potential for BIP to be significant and a major component of our investments. But what I would say is because many of these opportunities are development related, there is a delayed draw component to them. The capital gets deployed over a period of time.

So I think the significant dollars for the AI factories will come in a couple of years as opposed to the next year or 2. So I think I would just caution you from that perspective, even though we're discussing large dollars here, I think they're somewhat back-end loaded to use that terminology. But maybe now just to get into some of the projects we're working on. Lief, do you want to just give a quick update?

Yes, absolutely. Thanks for the question, Devin. So I think as Sam articulated, we see a massive opportunity in the space. We think that in excess of 100 gigawatts of incremental load will be required over the next decade. And I guess when you think about that, really hyperscalers are looking for partners who can engage at scale and can really help move the needle from that perspective.

And so when you think about building out a gigawatt-plus scale campus, that's a huge, huge undertaking. And we announced a project yesterday in West Kentucky that is located on a Department of Energy site that will ultimately serve a data center with in excess of 1.2 gigawatts of IT load. And that type of project will require up to $100 billion in private capital that will support both the data center itself as well as the compute inside and the power generation that will support it. And we think that, that last piece is a really crucial component to building out these AI factories.

Being able to indicate that there will not be an adverse impact to local ratepayers and that these AI factories are bringing their own generation. We think that, that's crucial both for the data center itself from a practical perspective in terms of having the electrons available, but also from a social license perspective and ensuring that there is strong local support and that these sites are bearing the cost of the grid that is ultimately required.

And so from Brookfield's perspective, we've been looking for sites like this around the world, given our global footprint. As mentioned, we announced a large-scale project yesterday in the U.S. But we also have large-scale sites in Canada, in Europe. We recently announced one in South Korea. And so we think that we're very well positioned to be a partner of choice for these large technology companies and sovereign governments around the world.

Okay. And then just the follow-up to that is just for your data center businesses, there seems to be growing pushback around the build-out of these facilities. We've definitely seen that more recently in the U.S. Just how do you think this plays out over time? Do you build where there's less resistance? Or are there different approaches being pursued that could have address at least some of the concerns from governments and local communities?

Yes. Thanks, Devin. It's Ben here. And I think as you noted, there definitely is an increased nimbyism or pushback against certain data center developments. And there are also -- I'd just make the observation, there's lots of sort of false narratives and perceptions out there about the industry itself. Our experience is that many local communities do welcome data center investments.

And from a geography perspective, I think as you noted, it's probably -- the nimbyism is probably most prevalent in the U.S. right now. We are noting that it's growing, I would say, in the European market. And we are starting to see some of this type of pushback in smaller markets like Canada as well. So it's definitely a dynamic.

In terms of the false perceptions themselves, they mostly relate to things like water consumption, rising electricity rates and noise, a perception that data centers create a lot of local noise. And so what the industry broadly is focusing on are very fulsome solutions to those types of issues because there are examples where those issues do manifest themselves, although by and large, the industry is good at these things. But the specific things are closed-loop water cooling as an example, where the consumption of water is de minimis, being neutral to actually positive on electricity rates and in supporting local utilities in supporting their local grids and minimizing noise.

So with that, I'd just say our businesses that we're focused on and our operating companies, we're leaders on all of these fronts in the industry, and we are seeing support from many local communities. So in terms of your question of where data centers get developed, I do think that the industry itself, there are lots of false perceptions and the industry is actively working on ensuring that it has solutions to all the concerns. And we're just focusing on the geographies where developments are welcome.

Our next question will come from Maurice Choy from RBC Capital Markets.

Just a quick question on AI for a moment. Notwithstanding all the comments that you just made about nimbyism and an earlier response about timing of payoff for the CapEx in AI. You've obviously been quite successful in signing a number of AI-related deals in Kentucky and South Korea. Maybe we are front-running the Investor Day a little bit, but I'm curious whether you see these AI opportunities is progressing in line with your prior projections? Or are there pockets of the AI infrastructure value chain that you think may be accelerating?

Well, maybe Dave can talk about just from our business plan perspective, how it's playing out. And then -- sorry, and your second question was -- can you repeat that?

Whether or not you felt all these AI opportunities were progressing in line with your prior projections? Or were there pockets of the value chain that you think may be accelerating?

Yes. I'll start. Maurice. Look, I think from a business plan perspective, I'd say it's fairly in line with what we had expected from a deployment perspective. I think maybe the pace of the announcements and the initial frameworks agreed to maybe have accelerated a little faster than we would have thought. But otherwise, from a pure investing perspective, I think we're right on track.

You will recall at Investor Day last year, we said, look, if the AI infrastructure strategy kind of unfolds the way we thought it would, we would be deploying $500 million of equity a year into this strategy. And look, I think with Bloom to date, we've probably done -- the contracts we have in place would probably put us close to $100 million year-to-date. And if we do upsize the framework and participate, that will certainly get us closer to the midpoint or the high end of that range we gave.

And then so as we look ahead to the following years to come, I think as we progress the commercial fronts on the AI factories in the various regions that Lief referred to, I think that will help achieve that $300 million to $500 million of equity invested in AI infrastructure on an annual basis. So I think we're kind of in line with where we thought we'd be.

And maybe just on your second part of your question about are we seeing all the various pockets of the value chain generating opportunities for us. And our AI group commonly refers to $7 trillion of potential opportunities, which obviously is a big number. But you can already see with the magnitude of the projects that we've signed, the market is massive.

We generally talk about 4 areas where we see opportunities. One would be AI factories. Two would be compute. Three would be behind-the-meter power opportunities. And then we kind of have a catch-all area of adjacencies related to AI. And so in terms of the first 3, I think you can see that we are actually advancing the strategy quite well. So in terms of compute, we've established Radiant, which is our in-house Neo cloud, and we've already signed agreements with customers to provide compute and the most recent one would be with NAVER, obviously, which will scale that up dramatically.

Behind-the-meter power opportunities, Bloom is a poster child for that. And I don't think we could have asked for anything better than that relationship. And then in terms of the AI factory, Lief just described all the different ones that we're pursuing. And we expect those to be shovel-ready, hopefully, in the next number of quarters. So in terms of the first 3 components, I think, absolutely, we've demonstrated the opportunities.

And on the adjacencies, that's really in all parts of our business, seeing opportunities come out of AI. And as Dave mentioned, we're seeing trickle-on effects into our transportation business. So I think the opportunity set is absolutely developing as we expected. And I think as you alluded to our upcoming Investor Day presentation, we'll probably touch on this a lot more.

Looking forward to that. If I could finish off with a comment that you've made. I think quoting you in the press release, the public markets have been increasingly effective exit channels to maximize value in your capital recycling program. Can you unpack that a little bit for us, especially relative to the private channels that you've utilized more in the past or perhaps how the private channels that they have changed?

Yes. Maybe I'll answer that question. So in short, nothing's changed on the private channels. So the private channels remain open and all the various -- we referred to our tools in our toolkit to exit remain very relevant, and we're executing them as we speak. What has changed in the last, let's call it, 9 to 12 months is just the fact that the equity capital markets opened up and we're very receptive to new IPOs, which we hadn't really seen for a couple of years. I think prior to doing the Rockpoint at IPO, I think the previous one was probably DBI, which was like 2020 or something like that.

So going back 3 before, I guess maybe, longer, 5 years. And so we're just taking advantage of the market as it exists as, really, just a competing source of capital to the private markets. And in some industries, it's competitive. In other industries, it's less so. And the market -- the window opens and closes. It's probably closed for the next little bit, but I suspect it will reopen just given some of the exciting companies that we know are coming to market in the fall. So we definitely don't think this market has shut for sure. It's going to reopen. And we'll continue to consider it on other opportunities.

[Operator Instructions] And I am showing no further questions from our phone lines. I'd now like to pass the conference back to Sam Pollock for any closing remarks.

All right. Well, thank you, Kristal, and thank you to everyone for joining the call this morning. We hope everyone is enjoying their summer so far for those in the Northern Hemisphere and look forward to hosting all of you for our Investor Day in Toronto on September 29, and we look forward to providing you an update on all our strategic priorities and our growth outlook. In the meantime, thank you again, and I hope you have a great day.

Thank you. This concludes today's conference call. Thank you for your participation. You may now disconnect.

Good day, and welcome to the Brookfield Infrastructure Partners 2026 Results Conference Call and Webcast.

[Operator Instructions]

Please be advised that today's conference is being recorded. I would now like to hand the conference over to your speaker, Mr. David Krant, Chief Financial Officer. Please go ahead.

Thank you, Sherry, and good morning, everyone. Welcome to Brookfield Infrastructure Partners First Quarter 2026 Earnings Conference Call. As introduced, my name is David Krant, and I am the Chief Financial Officer of Brookfield Infrastructure. I'm joined today by our Chief Executive Officer, Sam Pollock; and our Chief Operating Officer, Ben Vaughan. Also joining us today is Dave Joynt, a managing partner on our investments team.

I'll begin the call today with a discussion of our first quarter 2026 financial and operating results, followed by an update on our capital recycling initiatives. I'll then turn the call over to Sam, who will provide an update on our recent strategic initiatives before concluding with an outlook for the business.

At this time, I'd like to remind you that in our remarks today, we may make forward-looking statements. These statements are subject to known and unknown risk factors and future results may differ materially. For further information on known risk factors, I would encourage you to review our latest annual report on Form 20-F, which is available on our website.

So with that, Brookfield Infrastructure had a strong start to the year, delivering record results while continuing to advance a number of strategic initiatives across the business. We generated funds from operations, or FFO, of $709 million or $0.90 per unit in the first quarter. This is a 10% increase compared to the prior year. This performance was driven by strong base business results, highlighted by FFO from our data and Midstream segments increasing 46% and 12%, respectively, compared to the prior year. Results in our Utilities and Transport segments reflected resilient underlying performance with the current period impacted by higher levels of capital recycling activity achieved during 2025.

I'll now go through our results by segment in more detail. Our Utilities segment generated FFO of $201 million, up 5% year-over-year. The increase was primarily driven by inflation indexation and the benefit of over $500 million of capital commissioned into rate base, along with the contribution from our recently acquired South Korean industrial gas business.

Moving on to our Transport segment. FFO was $283 million, slightly below the same period last year. The decrease was primarily attributable to loss contributions from our successful asset sales. As a reminder, this included our Australian export and container terminal operations, the partial sale of a U.K. port operation and the majority interest in a portfolio of fully contracted containers at our global intermodal logistics business.

This was partially offset by the acquisition of our North American railcar leasing platform that closed on the 1st of January. After adjusting for all these factors, FFO was ahead of the prior year, reflecting higher volumes and tariffs generally across our rail and road operations. Our Midstream segment generated FFO of $190 million, up 12% compared to the same period last year. The increase reflects attractive commodity pricing, strong asset utilization and robust customer activity levels across our portfolio. Lastly, FFO from our data segment was $149 million, representing a step change increase of 46% compared to the prior year. The increase was driven by the contribution from our U.S. bulk fiber network, which we acquired in the third quarter of last year as well as organic growth across our data storage businesses, which included the commissioning of over 200 megawatts of operating data centers into earnings over the last year.

In addition to the strong financial and operating results we have delivered, we also made meaningful progress towards our 2026 capital recycling goal with proceeds secured of $1 billion to date. This includes closing the initial tranche of our partnership on a portfolio of stabilized and under construction data centers in North America and the closing of the sale of the largest of 4 concessions within our Brazilian electricity transmission business. We also completed a secondary sale of a 12% interest in our North American gas storage business. And finally, in April, we signed an agreement to sell our bulk liquid storage business, the largest independent storage provider in Scandinavia.

These asset sales improved our strong corporate liquidity position, which was $2.5 billion at the end of the first quarter. Our balance sheet remains well capitalized and our proactive approach to managing debt maturities has allowed us to remain opportunistic in the capital markets. During the quarter, we refinanced approximately $1.5 billion of nonrecourse debt on a net to bid basis, with no incremental borrowing costs for the business.

Before turning the call over to Sam, I would like to briefly note that we have recently begun exploring whether a single combined corporate structure would be the best path forward for the business. The goal is to determine if on a tax-free basis, we can create a single corporate security that would enhance liquidity, increase index inclusion and create value for investors. We are in the early stages of this evaluation, and we'll provide an update when appropriate.

So that concludes my remarks for this morning. I'll now turn the call over to Sam.

Okay. Thank you, David, and good morning, everyone. For my remarks today, I'm going to discuss our strategic initiatives before concluding with an outlook for the year ahead.

We have had an active start to the year with business development activity resulting in new strategic capital partnerships and continued progress under established frameworks. These partnerships are bilaterally sourced with high-quality counterparties and gives us exclusive access to investment opportunities that require long duration capital at scale. Increasingly, these frameworks are becoming a more meaningful avenue for growth, reinforcing our position as a partner of choice and expanding our opportunity set to deploy large-scale capital at attractive risk-adjusted returns. During the quarter, we established a new framework with a leading global investment-grade OEM, and launching an exclusive leasing platform for industrial equipment.

Through this platform, we will provide long-term leasing solutions that are expected to generate predictable cash flows without residual value interest rate or refinancing risk. We will have the sole discretion to enter leases under the framework with BIP's share of the equity investment expected to be upwards of $375 million.

Our $5 billion strategic partnership to install up to 1 gigawatts behind-the-meter power generation advanced further this quarter as well. We secured an additional $430 million CapEx project, bringing the total capital committed under the framework to approximately $1.6 billion. BIP's total equity commitment associated with the framework to date is approximately $60 million. Given the success of the behind-the-meter solution and strong customer demand based on speed to market, we may have the ability to expand the platform in the coming months.

We also remain on track to close Clarus. This is New Zealand's leading gas infrastructure utility in the second quarter. For an equity purchase price of approximately $70 million at our share.

Now moving to our outlook. We are progressing through 2026 from a position of strength, and remain very constructive on the backdrop for infrastructure. While recent geopolitical developments have contributed to greater market volatility, the essential nature of our businesses and the regulated or contractual profile of our cash flows continue to provide resilience and growth.

More broadly, demand for additional power, connectivity and logistics capacity continues to expand. This is being driven by digitalization, accelerating power demand, the rapid build-out of AI infrastructure and the ongoing reconfiguration of global supply chains. These tailwinds are expanding our opportunity set and providing attractive avenues to deploy capital at compelling risk-adjusted returns. Coupled with strong operating performance and a visible pipeline of organic growth projects, these factors position us well to deliver 10% plus per unit FFO growth in 2026.

As David mentioned, our capital recycling program and balance sheet continues to provide the flexibility to fully self-fund the growth ahead. With multiple sale processes underway across our business and continued access to capital markets during windows of opportunity, we are well positioned to fund our investment pipeline while maintaining financial discipline. Taken together, this supports our confidence in the outlook for 2026 and our ability to continue compounding value for our unitholders over the long term.

That concludes our remarks, and I'm going to pass it back to Sherry for -- to open the line for Q&A.

[Operator Instructions]

And our first question will come from the line of Devin Dodge with BMO Capital Markets.

Wanted to start on the recently launched equipment leasing business. I'm just trying to get a sense if this is part of the strategy for investing inside data centers, what kind of time frame you'd expect to deploy that $1.5 billion of capital and what do you view as the main risks associated with that investment?

Devin, this is Sam. I'll tackle that one. So the opportunity, I think, will likely be broader than just data centers, but initially, a good portion of the investment will be equipment for data centers. The -- we get a lot of comfort over the transaction itself because we're able to provide capital to essentially high-quality counterparties with investment-grade profiles with fully self-amortizing cash flow streams. So it's very attractive from that perspective.

We think we can scale this up as far as timing and deployment of capital. I think our hope is that on a gross basis, we'll deploy $1 billion to $2 billion of equity capital. So BIP share would be 25% of that. And we expect -- it's hard to predict flow, but I think we would hope to do that within a 24-month period.

Okay. I appreciate that. Second question, I was going to ask about that Intel JV. I didn't see any mention of it in your release, but I think Intel disclosures suggested the payments to the JV may have started in Q1. I guess, first, was that the case? And how quickly can return on investment ramp up in the coming quarters as both those fabs come online?

Devin, it's Dave here. I'll take that one. Look, I think in the past, we generally won't provide many updates specifically on the project. I think those generally come, as you said, from the Intel side. I think largely, the project has gone well. It's coming online, in line with our targets in terms of scheduling. They did make their first small wafer payments in the quarter. I would expect the initial -- the final capital contributions to go in over the next 6 months. And as those go in, I would expect the earnings to start to ramp up. And so I would expect to start seeing that come through our transmission and distribution segment of our data business in Q3 and then full run rate will be in 2027.

And that will come from the line of Maurice Choy with RBC Capital Markets.

I wanted to start with this concept of a single combined corporate structure. I have to assume that when the BIPC shares were first created back in 2020, something like this was contemplated. And if so, what were some of the obstacles back then and how those may or may not no longer be as big of an obstacle or even at all this time round.

Maurice, it's David again. But I think -- as we talked about this morning, we are in the early stages of considering this with the Board direction. And so it's probably hard to say what the obstacles are today. That's the word we'd like to complete over the next little while. As you know, the 2 companies for the last 6 years have served us well, but we're always looking at ways to improve our access to capital. And we think following a few things, obviously, the completion of our sister company BBU's process. We can now have some insights into how 1 simplified corporate structure will trade in the market. An early indication to that that's been positive, that this is the right time to reassess. And so I think that's probably all we can say at the moment in light of that, and I'll leave it that.

Fair enough. Maybe if I could just finish off more on your energy portfolio. Obviously, we've seen a series of support of federal and provincial government changes in Alberta and broader Canada. And also, there's been the conflict in the Middle East. Just your thoughts on the outlook for your business in the province notably into pipeline and NorthRiver?

Yes. And it's Ben here. I'll take that question. And look, all those developments are very positive for our Midstream business in Canada. We're seeing really strong demand from all of our clients for more access to our facilities and our pipelines. We completed about $400 million worth of growth projects in the past several months that are now starting to ramp up in terms of the revenue profile and delivering results.

And probably most importantly, we have a really tangible, meaningful pipeline of pretty bite size, relatively straightforward to execute and very low build multiple and highly accretive growth projects right in front of us. So I would expect the backlog -- our pipeline to grow. The backlog in Midstream is really attractive right now, and we expect to bring a number of those projects to FID in the coming quarters.

And maybe just to put in perspective, the magnitude of the projects that we're looking at the opco level would be roughly in the $8 billion range from a pipeline perspective. So it's a fairly meaningful size number of projects we have to look at.

One moment for our next question. And that will come from the line of Robert Catellier with CIBC Capital Markets.

I'd like to follow up on the potential corporate conversion that the Board is exploring, understanding it's quite early days. I wondered if you had any time lines for us in terms of what's reasonable to expect in terms of when a decision might be made?

Rob, as I said, unfortunately, there's probably not a ton we can share on the time line as of yet. As I said, we're just kicking off the process now.

Okay. No, that's reasonable. I just wanted to just dig into the Csquare IPO, which I understand they filed a confidential registration statement. I'm just curious as to how you chose the IPO route versus private sale and maybe you're dual tracking it, but maybe you could comment on that and how much you're expecting to sell by way of IPO.

Robert, I'll tackle that one. Look, I think in discussions with our advisers, the capital markets for IPOs are quite open at the moment. And obviously, there's a lot of anticipation for the upcoming SpaceX IPO. The one thing that public investors are looking for businesses that generate high cash flow, have still strong growth prospects and have great tailwinds related to the AI sector. And our business Csquare basically ticks all those boxes. We think it has the potential to be one of the leading IPOs of the year. And we're really excited about bringing that forward. And so just stay tuned.

One moment for our next question, and that will come from the line of Cherilyn Radbourne with TD Cowen.

I wanted to ask a couple of questions on the data segment. And I appreciate the data centers and Intel are the major growth drivers at the moment. Just curious how you think about the balance of your data portfolio in towers, fiber and so forth? And what value that provides in terms of diversity but also what you see in terms of inorganic opportunities there?

Cherilyn, maybe I'll start. But then I think you've given us a good segue to maybe talk about what's going on in the AI infrastructure sector, in particular, and we have Lief Williams here. who I think can expand on some of the things going on. But maybe just talking about some of our other businesses, we're seeing continued strong growth across the sector. One of the situations. And our colleague, Scott Peak mentioned it a number of months ago at our Investor Day. There's this domino effect. And the huge growth in data centers and AI is having impacts across basic utilities, power, Midstream and our other data businesses, including fiber -- our towers.

And the types of things that we're seeing is all our customers are looking to expand density across their networks. And so on the tower side, we have a number of build-to-suit opportunities that we continue to execute in all our businesses across Europe and Asia.

On our fiber businesses, there's still a huge amount of the U.S. in particular, but other parts of the world that have not been fiberized that are still operating on copper. And so that remains a lot of white space for us to continue to build out those networks and allow people to run all these new devices and programs more effectively.

So we see this as a continued 5- to 10-year build-out. And so all our businesses are well positioned. But maybe turning to some of the more, I'd call it even more exciting stuff going on in the AI infrastructure space. Peak, maybe just give us a little update on that.

Yes. Thanks, Sam. And good morning, Cherilyn. So the large users of AI factories and data sectors are highly, highly active in the market. There's effectively no data center inventory remaining for 2026. And even 2027 is quite scarce. What's interesting as well is that the demand profile has broadened from just data center capacity into also looking for compute. So leasing GPU as a service as well as behind-the-meter power opportunities. And so we see a large opportunity for groups like Brookfield, who have tremendous access to capital and an asset base to participate across all of those different asset classes.

Great. And then maybe just a quick follow-up on the data center side. I imagine that site selection and acquisition is particularly competitive and secretive. And so I'm not going to ask you to reveal anything proprietary. But to what extent is having a sister real estate business help in that regard?

I would say it certainly helps. And certainly, the scale of Brookfield is helpful in that regard. Just to give a sense, I would say there is a kind of dual track search for powered land. There's front of the meter options and then there's behind-the-meter options. Front of the meter options are challenging in the sense that the number of load applications going to utilities. It just kind of massively overwhelms the grid. And so utilities are now increasingly requesting large levels of credit or financial deposits, which is a huge disincentive to many of the parties out there looking for those front of the meter power solutions.

Behind the Meter also has its challenges in terms of delivering baseload power. At speed, the low emissions and highly modular. And so that's a place where, again, we think our Bloom partnership will be tremendously effective. I would say more broadly, we're starting to see some pushback in some locations in terms of the scale of these AI factories and then the risk of it pushing up rates for local ratepayers. And so again, I think Brookfield has been doing large-scale projects across a number of asset classes for many years. And so we think that we're very well positioned to help identify credible powered land sites. And help bring them to fruition.

One moment for our next question. That will come from the line of Robert Hope with Scotiabank.

Hoping to dive a little bit deeper on the AI factory and digital hub strategy. How are we progressing on those discussions with counterparties. And should we be in a position in 2026 to see some notable or sizable project announcements?

Robert, it's Peak again. I'll take this one as well. Yes, look, I think you do see, as I mentioned, tremendous demand from the large technology companies. The demand profile has broadened a little bit as well. There used to be a handful of large hyperscalers. We now have a wave of foundation model companies and inference operators who are also looking to secure the capacity.

So I would say there's a tremendous amount of noise in the market in terms of number of sites available and when does the power ramp. But what we're increasingly seeing is these users are looking for credible partners who have placed long-lead equipment orders and you have true dates for when the power is available. And we think that, that will benefit groups but Brookfield who are institutional and who have tangible sites that have a real ramp.

And so I would say the demand profile is very strong. I think you will see and we have seen strong leasing activity on the Brookfield portfolio over the last couple of quarters. And I think you'll continue to see strong demand through '26 and certainly through '27.

All right. Appreciate that. And then maybe moving over to kind of to the 10% organic growth rate or to the 10%-plus FFO growth rate for 2026. Can you comment how you're tracking on that? The organic growth at 8% seems strong, and the commodity price environment seems to be helping. Though it does appear that asset sales are coming a little quicker than M&A activity on the other side. So can you maybe talk about what the headwinds and tailwinds that you're seeing there are.

Yes, Rob, it's Dave here. And look, I think you did a great job summarizing my answer probably, but I think the -- I'll start by saying the first quarter was an excellent start. We delivered on our 10% target. And so with that behind us, I think we feel good with how the year is progressing. It's always hard to predict the timing of new investments and asset sales. But to your point, we have had some good initial success on the capital recycling. Front, I think from an all-in cost of capital is very attractive. The yield we'll see on that $1 billion, somewhere in the mid-single digits probably. And so I think from an accretion perspective, I don't expect that to be a meaningful drag on the business as we look ahead.

So all in all, I think we started the year off well. I think we feel confident with our 10% for the year still, and we'll continue to provide an update as we progress through the year.

[Operator Instructions]

our next question will come from the line of Frederic Bastien with Raymond James.

Good morning. You've historically leaned into periods of uncertainty in this location to pursue large acquisitions. How are you thinking about your ability to deploy capital into a sizable transaction this year?

Fred, I'll talk -- take that one. Yes, I think you're right. We've been historically successful in I think, taking advantage of dislocations. And when others have paused, we've seized the moment. At the moment, I'd say the market remains relatively calm and constructive, given all the volatility. There's still a fair amount of buyers out there. So I wouldn't describe this as an opportunistic market environment.

Nonetheless, I do think we're always on the lookout for large value opportunities. We -- what we're seeing is, today, the opportunity set around our AI infrastructure strategy is extremely strong. And so I'm very optimistic about us being able to do some exciting transactions there. We're also seeing an uptick in activity across Europe. And I think there's some larger value opportunities there that I think we can take advantage of.

And then we keep on monitoring the capital markets. Often some of our best acquisitions are when the public market will pull back, and we can take advantage of a take private opportunity. So those are the things that we're up to I can't give you like a time line on when we'll do our next large deal, but we're always out there. We have tremendous partnerships so that we can execute on those things and optimistic there will be some exciting deals in front of us.

That's helpful. I just wanted to tack on a Midstream related question as well, thinking switching gears on monetization. Are you -- how are you thinking about NorthRiver and a potential monetization there?

Sorry, that was about monetizing NorthRiver.

Yes.

Well, look, I could turn it over to Ben in a second if he wants to add anything. But what I would just say is the business has had tremendous commercial success in the past couple of years. We've extended our contract term, I think, to close to 12 years now. And we think it's really well positioned. Our debate internally is whether or not we continue to build out the business and take advantage of some of the additional growth that's there or whether we bring it to market and sees what is probably a pretty constructive environment for Midstream businesses.

So we're weighing those considerations. It's performing really well, and we just don't have an answer for it at this point in time.

I'm showing no further questions in the queue at this time. I would now like to turn the call over to Mr. Sam Pollock for any closing remarks.

Great. Thank you, everyone, and thank you, Sherry, for hosting this call. We appreciate you joining us today to hear about our results. And we look forward to the warmer weather that's in front of us and the hockey playoff season. I hope all you Habs fans are cheering for my team. And we look forward to providing an update in the quarter ahead.

Thank you. This concludes today's program. Thank you all for participating. You may now disconnect.

Good day, and thank you for standing by. Welcome to the Brookfield Infrastructure Partners Fourth Quarter 2025 Results Conference Call and Webcast. [Operator Instructions] Please be advised that today's conference is being recorded.

I'd now like to hand the conference over to your speaker today, David Krant, Chief Financial Officer. Please go ahead.

Thank you, Liz, and good morning, everyone. Welcome to Brookfield Infrastructure Partners' Fourth Quarter 2025 Earnings Conference Call. As introduced, my name is David Krant, and I'm the Chief Financial Officer of Brookfield Infrastructure. I'm joined today by our Chief Executive Officer, Sam Pollock; and our Chief Operating Officer, Ben Vaughan. Also with us today is Dave Joynt, a Managing Partner; and Udhay Mathialagan, Head of our Global Data Center businesses.

I'll begin the call today by highlighting our results for 2025, followed by a recap of our record year of capital recycling. I'll then hand the call over to Udhay who will elaborate on our approach to AI infrastructure investing and how we have been able to turn sector tailwinds into durable value for unit holders.

Finally, Sam will provide an update on our recent investments before concluding with an outlook of the business.

At this time, I would like to remind you that in our remarks today, we may make forward-looking statements. These statements are subject to known and unknown risks, and future results may differ materially. For further information on known risk factors, I would encourage you to review our latest annual report on Form 20-F which is available on our website.

2025 was another strong year for Brookfield Infrastructure. Our key accomplishments include exceeding our capital recycling target of $3 billion. investing approximately $2.2 billion of equity into growth initiatives and completing approximately $16 billion of financings to further derisk our operating company balance sheets.

From a results perspective, we generated FFO comes from operations of $2.6 billion during 2025. Normalized for the impact of asset sales and foreign exchange, FFO increased 10% compared to 2024 in line with our target and reflective of our operational performance and the strength of our business. This result includes record FFO during the fourth quarter of $0.87 per unit.

Given this performance, a conservative ratio for the year of 66% and a strong outlook for 2026, I'm pleased to report that the Board of Directors have approved a quarterly distribution increase of 6% to $1.82 per unit, on an annualized basis. This marks the 17th consecutive year of distribution increases of at least 5%.

I'll now go through [indiscernible] the base business continued to perform well during the year. driven by inflation indexation across the portfolio and the contribution of roughly $500 million of capital commissioned into rate base over the last 12 months.

Moving on to our Transport segment. FFO totaled $1.1 billion, in line with the prior year after normalizing from $1.8 billion of capital recycling initiatives. The loss of earnings from these sales was partially offset by higher revenues across our transportation networks, particularly in our rail and toll road segments where volumes and rates grew on average by 2% and 3%, respectively.

Our Midstream segment generated FFO of $668 million for the year, representing a 7% year-over-year increase. This growth reflects higher volumes and activity levels across our midstream assets particularly at our Canadian natural gas gathering and processing operations and our recently acquired U.S. refined products pipeline system.

Lastly, FFO from our data segment was $502 million a step change increase over 50% compared to the prior year period. The increase is attributable to several new investments completed over the last 12 months, the most recent being our U.S. bulk fiber network, which is now fully contributing to earnings. -- the fourth quarter. In addition, we achieved strong organic growth across our data storage business, which included the commissioning of 220 megawatts of capacity at our hyperscale data center megawatts of new billings at our U.S. retail colocation data center operations and income generated by our global data center developers. Our global data center platform now has development potential approximately 3.6 gigawatts, including contracted capacity of over 2.3 gigawatts today.

Before turning it over to Udhay, I would like to briefly touch on our record liquidity which totaled $6 billion at the end of 2025 and included just under $3 billion at the corporate level. Contributing to this strong position was a record $3.1 billion in asset sale proceeds raised in 2025. We believe that the elevated pace of capital recycling will continue into the year ahead.

We already have 2 transactions secured that crystallize attractive returns. The first which is we agreed to sell the largest of 4 concessions within our Brazilian electricity transmission operations. We expect proceeds of approximately $150 million net to BIP generating an attractive IRR of 45% and over 8x multiple capital. closing for the transaction is expected at the end of the first quarter in 2026.

Secondly, we formed a capital partnership for a portfolio of stabilized and under construction data centers in North America. Proceeds from the sale are expected to be used to support the buildout of our powered land bank within the business.

That concludes my remarks for this morning. I'll now pass the call over to Udhay.

Thank you, David, and good morning, everyone. AI is justifying dominating headlines with many bold predictions ranging from data centers and space to breakthrough in quantum computing that could one day redefine how the world operates. At the same time, many are questioning the merits of the magnitude and velocity of capital flowing into AI and where the demand will materialize at a level that justifies this spending. The sheer scale of investment underway to build the physical backbone that makes AI possible is staggering.

In 2025 alone, corporates invested approximately $500 billion into AI-related infrastructure, with capital investment over the next 2 years expected to rise further. Much of this build-out is fundamental to the development of AI, enabling power intensive workloads to run reliably, securely and at scale in well-connected locations. The reality is driving a sustained wave of investment into the backbone infrastructure that enables AI, including data centeric capacity, good resiliency, power generation and transmission. The sector remains exposed to overbuilding, technological change and disruption.

With capital moving quickly, not all participants will be rewarded and there will be mistakes made. Our approach is designed to protect against such exuberance. Brookfield Infrastructure is applying a prudent risk-focused approach to participating in the build-out of AI infrastructure, maintaining strict guardrails to safeguard our capital.

First, our development projects are underpinned by long-term contracts with favorable terms. We do not build speculatively and earn an attractive return within the initial contract period, mitigating technology risk. Second -- the second guardrail is that we selectively focus on the strongest investment-grade counterparties was some of the largest, well-capitalized and most profitable technology companies in the world.

Third, we concentrate on top-tier workload agnostic locations for our data centers. that can support the full spectrum of demand, reducing the risk of the single theme exposure and increases the durability of demand through cycles. The fourth is our disciplined strategy, we are deliberate in how much land and powered shelves we control and develop.

We have created a self-funding model that provides funding for future development and locks in attractive developed economics. -- as well as reduces the size of our platform while maintaining the benefits of scale. And fifth, we've matched the capital structure to the tenor of the contracted cash flows with a focus on preserving flexibility and ensuring that we can finance growth responsibly.

To illustrate the benefits of our approach during 2025, we experienced exceptional demand at our data center platforms, securing record growth, commercialization, capital recycling and capital markets activities. For example, at our U.S. core location data center business, we experienced 11 consecutive quarters of record bookings, and it's now fully utilized across several markets.

During the quarter, we signed several large contracts at a data center in Illinois, achieving 100% occupancy and adding approximately $45 million of annual EBITDA on a run rate basis commencing later this year.

Without investing any further equity, we acquired and added a 40 site data center portfolio at January 2024 to our existing business and subsequently increased EBITDA from a combined base of approximately $200 million to approximately $500 million on a contracted basis. The exciting part that the growth journey is expected to continue, led by high returning under-roof densification and in-footprint expansion capacity, which total over 600 megawatts of identified growth potential.

Across our global data center platform, we achieved a significant lease-up of our land bank during the fourth quarter, which is expected to be commissioned over the next 3 years. We executed agreements for approximately 800 megawatts of capacity predominantly in North America. The vast majority of these leases are with investment-grade customers and underpinned by long-term contracts.

Since acquiring our North American and European platforms, our adherence to the guardrails outlined above has allowed us to maintain a consistent greenfield data center yield on cost. In 2025, we partnered on almost 850 megawatts of stabilized and operating sites in North America and Europe, crystallizing developer premiums and demonstrating strong demand.

Taken together, we hope these examples highlight both the strength of demand we are seeing and importance of disciplined execution converting demand into durable returns.

As AI workloads scale, the value well-located powered infrastructure intensifies. In this environment, scale, reliability and access to capital are differentiating factors to counterparties and we believe our global operating capabilities and long-standing relationships benefit us.

Our risk-focused approach and strict adherence to guardrails will enable us to continue investing in the core infrastructure needed to deliver AI at scale while protecting our downside.

That concludes my remarks for this morning, and I will now pass the call over to Sam.

All right. Thank you, David, That was great, and good morning, everyone. For my remarks today, I'm going to discuss some of our strategic initiatives and then conclude with an outlook for the year ahead. In 2025, transaction activity accelerated and as a result, we deployed approximately $1.5 billion into new investments. We expect this momentum to carry into 2026 based on our robust pipeline of new investment opportunities that continues to be diversified across sectors and geographies.

During the quarter, we completed the inaugural project under the framework agreement with Bloom Energy, installing 55 megawatts of behind-the-meter power for a data center site in the United States. We have since secured additional projects under the framework for several hyperscaler customers, bringing the total to approximately 230 megawatts of power generation. These additional projects have contract terms of at least 15 years in length. BIP's total equity investment associated with these projects to date is expected to be approximately $50 million and fully deployed by mid-2027.

Also during the quarter, we closed the acquisition of a South Korean industrial gas business which is the leading supplier of industrial gases to investment-grade semiconductor manufacturers in the country. The total equity purchase price is $125 million for our share. And on January 1, we closed the acquisition of a leading railcar leasing platform in partnership with a best-in-class railcar lessor. The business is highly cash generative, providing stable cash flows that are supported by a diversified and large investment-grade customer base. BIP's total equity consideration is approximately $300 million.

Now turning to our growth outlook. We see a highly constructive backdrop for infrastructure in 2026. The asset class has a long history of delivering resilience growing cash flow through a variety of market environments and is squarely positioned at the center of 3 powerful structural themes, which we've talked about quite a bit in the past, digitalization, decarbonization and deglobalization.

Together, these forces are driving an infrastructure investment super cycle that is broadening in both scope and scale. We have entered 2026 from a position of considerable strength as well. Our base business is delivering resilient growing cash flows, and we have clear visibility into a multiyear runway of organic growth and capital deployment.

In addition, the rapid build-out of AI-related infrastructure is materially expanding our opportunity set across data centers, power and network connectivity. As a scaled global owner and operator of critical infrastructure, we are well placed to deploy capital into these teams at attractive risk-adjusted -- these factors, combined with a stable interest rate and foreign exchange backdrop, position us well to return to our 10% or higher per unit growth target in 2026 and beyond.

So that concludes my remarks. I'll now pass it back over to Liz to open up the line for Q&A.

[Operator Instructions] Our first question comes from the line of Maurice Choy with RBC Capital Markets.

I'll just ask one question, but I'll admit it is a multipart question on data centers and data infrastructure. Udhay in your prepared remarks, you highlighted how your contract approach aims to mitigate technology risk. Can you elaborate a little bit more on that? And also what risk do you think is underappreciated by the market.

And my quick follow-up is going to be on returns. Obviously, I would expect the returns are superior to the 12% to 15% target range. So maybe you could help us understand a little better how much more better, even if it's just a range, driving some factors for us to consider and quantify these premium returns?

Hi, Maurice, it's Sam here. Maybe I'll start off with the returns. And then I'll have Udhay talk about the contract items and the risk that you also asked. So on the return front, I'll keep it high level and simple. But in essence, we develop new data centers at a yield the cost anywhere on average between 9% and 10%. And and we monetize them at cap rates actually 5.5% and 6% on average. And so that gives us a rough development profit 10 basis points.

And with leverage in the development, 70% range, that pencils into equity returns if we do everything right. into high teens or 20s. And I think that is a profitable industry. So that's the rough pull, and I think we'll leave it on that from returns perspective?

And then maybe I'll throw it over to Udhay to answer your first 2 questions.

Sure. Thanks, Sam. Look, I think taking a step back, the basis of pretty much all our data center businesses is around providing the core infrastructure and staying out of the real -- the technology that our tenants, our customers use. And so my earlier remarks around being managing the technology risk is really around the way the environment within the data centers are being designed for longer-term use and for changes that are happening at the compute infrastructure level. That predominantly translates into how power and cooling works in the data centers.

So by making sure we've got very long-term contracts, so let's say, 15-year contracts, which are very specific in terms of what we deliver, we're staying completely out of any technology change that could take place in that 15-year period at a customer's sort of end. And in this -- in case it necessitates any change in the underlying infrastructure, then those as specific changes that are not to our cost at that point in time. So that was the underlying sort of comment around how we're managing our -- the committed cash flows, I guess, over that period of time. in terms of technology risk.

Our next question comes from the line of Devin Dodge with BMO Capital Markets.

All right. Maybe to the extent that you're able, can you provide some additional color for the transaction where KKR acquired a stake and portfolio of data centers from Compass. And just trying to get a sense for how many assets are included, the timing? It sounds like it might be phased into that partnership and maybe the net proceeds to BIP. .

Dan, it's Sam here. I can't really speak to the details of it because we don't get into those level of granularity on specific transactions that are private -- what I can tell you is that -- and we mentioned this earlier in the call, we've -- we effectively entered into JV arrangements with a number of institutional investors, which I would include KKR in that group across not just North America, but Europe as well. totaling about 850 megawatts.

And effectively, the way that the intense work is these are, for the most part, passive vehicles in the sense that we retain operational control of the assets and retain a significant ownership stake to have alignment with our partners. And so we've done this, as I said, in markets, and it's kind of part of our playbook to recycle capital from developments to crystallize some profits to reinvest back in the business. So we can fund future growth.

Okay. Okay. Second question for Brookfield's is a $10 billion AI infrastructure fund. I believe BIP is 1 of the pools of capital that could be used to meet Brookfield's commitment. I was just wondering if you could provide a framework or thoughts on what types of investments made by the fund may be suitable or not suitable for BIP? .

Devin, so that's correct. So BIP is one of the entities that will fund opportunities that come from that strategy. And I think the way to think about it is transactions that have the profile that we have in our flagship funds. So returns that are let's say, 12% and higher in sectors that are suited for BIP things that are outside of renewable energy and investments that probably don't have a development profile that's too, too long.

If the development cycle is excessively long, then that may not make it appropriate for a bit. But otherwise, I think keeping in mind for pool construction objectives for BIP. If it's in the data center sector, if it's gas-related if it's utility related, those are all sectors and if the returns fit, then we would invest through BIP for those type of transactions.

Our next question comes from Cherilyn Radbourne with TD Cowen.

Thanks very much, and good morning. On the data center side, I did want to ask if you could talk about how you think about sovereigns versus hyperscalers of counterparties. And how you think the mix of your basket of counterparties could end up between those 2 groups?

Hi, Cherilyn, that's great to have you on the call. So maybe I'll touch on this and Udhay, can add anything else you'd like to. I think we like both of the counterparties because it gives diversity. One of the things that serves us well across all our business is diversity of counterparties. And obviously, the hyperscalers will amazing credits, are few in number and have similar exposures to AI and other data-related cash flows and sovereign nation diversifies from those risks.

It also -- the other reason we've been focused on some of these sovereign AI factories is because we think it gives us a differentiated strategy than many others who are just focused on building the large mega sites for the hyperscalers. Here, we can on a more bespoke basis to assist sovereign nations to build ecosystems in their countries. The challenge with it is that governments tend to move a bit slower with than corporates.

And so the time to market can sometimes be a bit longer. But as far as what the mix will be, that's a little bit too hard for me to predict at this stage. I mean we'd love to have a broad base of both hyperscalers and sovereign credit. But it's a little premature for me to speculate on that.

That's helpful color. And then more of a straight-up question for David. Can you give us a sense of what we can expect from inflation indexation across your various geographies in 2026?

Look, I think as we look forward, the 2 biggest drivers of growth from an organic perspective in 2016 will be the inflation indexation you highlighted as well as a significant commissioning of CapEx out of our backlog. As you've seen, it's a record level now. On the inflation front, I'd say in OECD markets, we're probably averaging between 2% and 3% on our escalators.

And then on the emerging markets, it ranges depending on which metric you're looking at. But I'd say between India and Brazil as the 2 biggest emerging market exposures we have inflation pass-through in -- it's probably also in the 2% to 4% depending on the metric. So I think it's more manageable, still above probably 25, 50 basis points above our historical averages that you would have observed, but certainly not as elevated that you saw in 2022.

Our next question comes from Robert Hope with Scotiabank.

Can we dive a little bit deeper into the data operations capital backlog? It looks like it's up just over $1 billion versus Q3 with about $900 million of that driven by the hyperscale backlog -- so can you maybe dive a little bit deeper into what is driving the significant increase in Q4 as well as kind of what is the outlook and how large can this get?

I can start and Udhay or Sam can jump in. Look, I think, Rob, you certainly pointed out. I think across the data segment, the data center platform had the most growth. We've also onboarded the bulk fiber backlog and order book in hot wire that we acquired in the fall. So those are another key driver in the increase in the last half of the year.

But on data center itself, I think it's a little chunky in terms of when we sign new on. And as we highlighted this quarter, we had significant momentum on the leasing activity. So there was about 100 megawatts signed globally. When we sign those contracts, that's effectively when we'll put in the backlog associated with those. And so up until then, as you heard through our call, there's very little investment until a contract is signed. And then at that point, we then effectively consider the project FID and adds into our backlog. And so as you heard, it's probably a mixture of North American, European as well as a few in Asia Pacific that drove those -- those signings drove the addition to backlog.

And then maybe sticking with data centers. So $3.9 billion of the backlog relates to Intel. Can you update us in terms of timing, how you're thinking about cash flow and returns there and any potential follow-on investments?

Ben, do you want to give an update on the in-service date for the Intel facility?

Yes, sure. So for the Intel facility, our JV has 2 fabs and 1 of which has its in-service date. So it's now producing wafers, which is great. And the second fab is making good progress towards completion. So the actual underlying operations our JV and the construction activity is progressing very well for Intel.

Next question comes from Robert Catellier with CIBC Capital Markets.

You seem to have a pretty high level of conviction in the capital recycling, having just come off a record year and you're also continuing to make new investments. But I'm curious about the rate of commissioning capital from the backlog in 2016, given this is an important part of the capital allocation process -- so wondering if you could maybe quantify and characterize what you see coming in the next couple of years there relative to $1.5 billion commissioned in 2025.

Rob, it's David here. I can give you some color on the shape of that commissioning. I think I'd split our backlog into 2 various components. As you heard from the previous question, there's an Intel component which is about $3.9 billion in the number. We'd expect that to commission from our earnings profile in the back half of 2026.

The other -- the balance of our backlog would be diversified across our utilities transport midstream and data businesses. And typically, as you've heard, it's a 3-year outlook. So those projects tend to be smaller, lower, shorter development cycles and build cycles. So I'd expect roughly 1/3 of that backlog, which should be close to $1.5 billion to come in online 2026 as well throughout the year. It is -- I wouldn't say there's a chunky element to it. It's pretty -- it will be pretty smooth across our utilities and our data centers driving the bulk of that.

Right. So exing out Intel, which is obviously a unique investment. You're really looking at about $1.5 billion-ish a year then. Is that...

What goes to our backlog, excluding Intel is $5.3 billion. And so assuming average of 3 years, you're looking close to $1.75 billion probably and the $1.5 billion to $2 billion.

Okay. Excellent. And then my other question was just what are your views on the Canada, Alberta MOU as it relates to energy development. It looks like there's a momentum building towards a bolder energy strategy here. So I'm curious how it impacts how you manage your midstream investments in Canada. So do you hang on for more growth -- or does this derisk the asset to a point where you might consider more asset sales?

Robert. Look, look, I think it's -- it's too early to say whether or not the MOU is going to have any material impact on the growth trajectory of our businesses. Irregardless of that, though, we've already have plenty of growth in the -- there's been significant producer expansions underway, which has led to some additional tie-ins particularly in our IPL facility IPO network. And we've recently undertaken a number of growth initiatives North River.

In terms of our plans to monetize the businesses, I think we have business plans in place for each of the businesses that we're looking to continue to develop. And I guess the only thing that would accelerate monetizations would be market conditions to the extent that they to us bringing some or part of these businesses to the market. We might look at that. So I appreciate some of that it's very loose.

But I think the takeaway is that the businesses today operate in a very strong environment. And with the added push by the federal government with the provincial government to encourage further growth in the sector. We think that's only helpful to our businesses and makes them more attractive to potential buyers on Yes, I totally agree with you. I think it's too early, and I too would want to say a couple of more cards slipped on how they have a MOU plays out and if they achieve the milestones as intended. So thanks for your answers.

Our next question comes from the line of Frederic Bastien with Raymond James.

During your Investor Day, you noted that Brookfield Ad form partnerships to build 7 AI factories totaling 6 gigawatt of compute capacity. Can you provide an update on how that's going and whether you get more developments to announce soon?

Fred, maybe I'll tackle that and Udhay might add some further comments, but I think the answer is relatively short. We continue to progress all those very initiatives. And today, we have discussions underway with probably 5, at least in Europe. -- some in North America as well as some of the Middle East and one actually in Oceana.

So basically, across the globe, as I mentioned I think it was Cherilyn, the -- it -- these discussions do take time, and we're probably a little disappointed that they haven't gone a little faster given the importance that each of the put towards these initiatives.

Nonetheless, I think we're hopeful that during the year, we'll have 1 or 2 of these progressed. And I think the only thing that I would caution you is that they tend to be smaller than some of the mega sites that you see announced with the hyperscalers. So most of these are anywhere between as small as 50 megawatts up to as much as maybe 250 in phases. Nonetheless, those are still represent meaningful dollars, and we're pretty excited to see it through.

And I guess your relationship with Bloom Energy is still fairly young. You've committed to delivering just under 300 megawatts of power generation. I think your original agreement was to -- was for up to 1 gigawatt of behind-the-meter power generation is. Are you comfortable that you will see through this agreement all the way to that 1 gigawatt?

Obviously, we'd be speculating on the future, so it's hard to predict. But at the moment, with the level of demand that they're seeing and the amount of developments underway, I feel pretty optimistic that we'll get to that and maybe even above that level. There's no doubt there's for Bloom at the moment.

So I mean your relationship obviously is strong and growing, obviously. .

Yes. Yes, it is.

That concludes today's question-and-answer session. I'd like to turn the call back to Sam Pollock for closing remarks.

All right. Thank you, Liz, and thank you, everyone, for joining the call this morning. With you've all had a good start to the year, and we look forward to providing our first quarter results at the end of April. Thank you, and take care.

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

Good day, and thank you for standing by. Welcome to the Brookfield Infrastructure Partners L.P. Q3 2025 Results Conference Call and Webcast. [Operator Instructions] Please be advised that today's conference is being recorded. [Operator Instructions]

I would now like to hand the conference over to your speaker today, David Krant, Chief Financial Officer.

Thank you, Josh, and good morning, everyone. Welcome to Brookfield Infrastructure Partners' Third Quarter 2025 Earnings Conference Call. As introduced, my name is David Krant, and I'm the Chief Financial Officer of Brookfield Infrastructure. I'm joined today by our Chief Executive Officer, Sam Pollock, as well as Ben Vaughn and Dave Joynt, who will be available for the question-and-answer portion of the call. .

I'll begin today with a discussion of our third quarter 2025 financial and operating results, followed by a discussion of our financing activity and strong balance sheet position. I'll then hand the call over to Sam, who will provide an update on our strategic initiatives and conclude with outlook for the business.

At this time, I would like to remind you that in our remarks today, we may make forward-looking statements. These statements are subject to known and unknown risks, and future results may differ materially. For further information on known risk factors, I would encourage you to review our latest annual report on Form 20-F, which is available on our website.

Brookfield Infrastructure had another solid quarter, achieving strong financial results and executing on our strategic initiatives. Beginning with our financial and operating results, we generated third quarter funds from operations, or FFO, of $654 million or $0.83 per unit. This was 9% higher compared to the previous year, driven primarily by strong organic growth, highlighting the financial strength and stability of our base business. These results were delivered despite FFO contributions for following a year of record asset sales and only a partial contribution from the new investments we've made.

Turning to our results by segment. Our Utilities generated FFO of $190 million, slightly ahead of the prior year. Results benefited from inflation indexation in addition to contributions from over $450 million of capital added to the rate base. The strong underlying performance was partially offset by higher borrowing costs and the sale of our Mexican regulated natural gas transmission business in the first quarter of this year.

Moving to our Transport segment. FFO was $286 million for the quarter. Headline results are lower than last year due to the sale of our interest in our Australian export terminal, DBI, and the sell-down of stabilized containers within our global intermodal logistics business. After adjusting for these capital recycling initiatives, our results were slightly ahead of the prior year.

The solid underlying performance reflected strong volumes across our networks and rate increases on our rail networks and toll roads.

Our Midstream segment generated FFO of $156 million, representing a 6% increase over the same period last year. We experienced strong customer activity levels and asset utilization across our portfolio, particularly at our Canadian diversified midstream operation. Notably, we completed the acquisition of Colonial Enterprises this quarter. The partial earnings contributions were offset by the lost income associated with the sale of our U.S. gas pipeline in the second quarter of this year.

Lastly, FFO from our Data segment was $138 million, representing a step change increase of over 60% compared to the prior year. The increase is driven by a full quarter contribution from the tuck-in acquisition of a tower portfolio in India completed last year as well as strong organic growth across our data storage businesses. This growth included income earned by our developers, the commissioning of 80 megawatts of capacity at our hyperscale data centers and 45 megawatts of new billings initiated at our U.S. retail colocation data center operation.

Before turning the call over to Sam, I'd like to provide an update on recent financing activity. Debt capital markets remained favorable during the quarter, with significant new issuance activity and further tightening in credit spreads. During the period, we completed financings to enhance our liquidity, support growth initiatives and refinance near-term maturities. This included a $700 million corporate issuance of medium-term notes in September. The issuance had a weighted average interest rate of approximately 4% and was priced at the tightest credit spread in our history.

As a result of our proactive approach to refinancing, less than 1% of our nonrecourse debt is maturing over the next 12 months. We maintain a well-laddered maturity profile with a weighted average maturity of approximately 7 years. Our balance sheet remains well capitalized with liquidity at the end of the third quarter totaling $5.5 billion, which includes $2.5 billion at the corporate level and over $1.4 billion in cash across our operating businesses. This strong liquidity position positions us with the confidence to pursue a variety of growth opportunities as they arise.

That concludes my remarks for this morning. I'll now turn the call over to Sam.

Great. Thank you, David, and good morning, everyone. For my remarks today, I'm going to provide an update on our transaction activity, and then, I'll conclude with an outlook for our business.

Now, starting with investments. We've already met our deployment objective for the year, securing 6 new investments totaling over $1.5 billion. This quarter, we secured 3 new investments across diverse regions and sectors, whereby BIP will deploy approximately $225 million in total.

The first investment is a $1.3 billion enterprise value New Zealand natural gas infrastructure operations. The business primarily operates a leading gas transmission, distribution and storage business that is comprised of regulated and long-term contracted revenues with inflation taxation. This value-based acquisition is highly cash generative, resulting in a short payback period of approximately 7 years. We expect the transaction to close in the second quarter of next year, subject to customary regulatory approvals.

The second acquisition is a $1 billion enterprise value company that is a South Korean industrial gas business that supplies industrial gases to industry-leading and investment-grade semiconductor manufacturers. The majority of the business is underpinned by 20-year minimum take or pay off take agreements with significant cost pass-throughs. This transaction is expected to close later this quarter.

And then lastly, we've secured our first AI-related project under a newly established $5 billion framework agreement with Bloom Energy to install up to 1 gigawatt of behind-the-meter power solutions for data centers and AI factories. This project provides a hyperscale customer with 55 megawatts of behind-the-meter power for an AI data center in the United States.

Now, with respect to capital recycling, the momentum in our asset sales program has continued. During the quarter, we progressed a number of initiatives, and we've now generated over $3 billion in proceeds for the year and are on track to achieve a further $3 billion over the next 12 to 18 months. One of the most significant asset sales completed in mid-October was the parcel sale of our North American gas storage platform and what is the largest IPO in the TSX since May of 2022. In total, we raised CAD 810 million, and BIP's share of the net proceeds from the offering was approximately USD 230 million. Since the formation of Rockpoint, which is our natural gas business, which was done through a series of acquisitions, EBITDA has grown by more than 4x, driven by operational improvements and favorable market fundamentals. As a result of various strategic initiatives, which enhance the stability and quality of earnings, along with the sale of 2 noncore assets in 2023, we have now realized a 3.2x multiple on our invested capital while continuing to own a significant interest in the business.

Now, the outlook for Brookfield Infrastructure for the balance of the year and looking into next year remains favorable. As mentioned at our recent Investor Day in September, we believe BIP is at an inflection point in its growth profile. Each of our new investments this year is expected to deliver returns above our 12% to 15% target range, and it's backed by credible business plans that support potential upside returns over 20%. We also have a robust pipeline of new investment opportunities across each of our existing segments driven by the long-term megatrends that we've talked about many times in the past of digitalization, deglobalization and decarbonization.

We're also seeing a significant new growth vertical emerging from the rapid build-out of AI infrastructure, a $7 trillion opportunity set that remains in its early stages and continues to expand. We expect to deploy up to $500 million annually into AI-related infrastructure in the coming years with AI factories and behind-the-meter power solutions, representing a natural compelling extension of our investment activities. These growth factors paired with a macroeconomic backdrop that is trending very favorably, set the stage for BIP's FFO per unit growth to inflect higher.

So that concludes my remarks, and I'll pass it back to Josh to open the line for Q&A.

[Operator Instructions] Our first question comes from Maurice Choy with RBC Capital Markets.

If I could just start off with the capital deployment opportunities, it is clear that data infrastructure and energy had been quite thematic of late, and that's consistent with your Brookfield day messaging. Just your thoughts on the rising competition for these types of assets. And if you could break it down, whether that be types of subsectors within a new vertical or even geographically? What does that mean to your ability to source these opportunities?

Maurice, this is Sam. I'll tackle that one. I think what we've -- what we flagged at Investor Day, and I guess, I think we've talked about for a number of years now, is that we have seen a significant increase in the number of opportunities to deploy capital, particularly in the data sector and in sectors impacted by digitalization.

And you're right that there's always -- there are new players who are competing for those opportunities with us. And frankly, over the last 15, 20 years, we've seen new entrants into the Infrastructure sector. We remain confident that because we have a global franchise and access to the most significant amounts of capital of any player that we remain with a distinct advantage in sourcing the best opportunities to deploy capital. And in sectors and regions where capital is plentiful, we'll avoid entering into those cost of capital shootouts and look for areas where capital is more scarce or where people are looking for partners that they can rely on and have trust in to be long-term counterparties. And those are the things that I think, historically have made us successful.

And I think today, many of the large tech companies who are making significant investments in the many billions of dollars, they're looking for players like us that they can have confidence in to deliver the projects and be there for them through thick and thin. So I know that was probably a bit of a general answer to your question, but I think, hopefully, you get the tone that, yes, there are new players; however, the opportunity set is still very large and our specific expertise and skill sets position us to get the best returns and not be competed down to the lowest common denominator, which is, I think, what you're worried about. And I don't think that's an issue.

That's great color. If I could just finish off with the question about the LP unit repurchases and the establishment of the ATM program for the BIPC shares. I wanted to get your view about the timing of any action, and also, what does success for both these 2 actions and programs look like in your view? How do you measure that? Over what time period would you measure that? And conversely, what would be an unsuccessful outcome?

Maurice, it's David. I can start. Look, I think at this point, as you would have seen in our release, it's something we're contemplating at this time. We still have, obviously, filings to do to be in a position to execute the program. So I'll just caveat with that first. I think the key point, and I don't think we'll get into measuring success or failure of the program. I think what we're looking to do and one of our objectives has been to increase the liquidity of BIPC. And I think that is, in essence, what the announcement today will look to do if we go ahead with it. And I think we -- given the success we've had on the capital recycling front, we don't need the capital. So issuing BIPC under an ATM on its own isn't helpful to the business, so we felt pairing it potentially with an NCIB would be a way to avoid any dilution to our existing shareholders and to our business. And so I think that was the ultimate intent of the program if we do decide to go ahead with it.

And maybe just a quick follow-up. And maybe this is also somewhat related to the float of the BIPC shares is the secondary target also to perhaps tighten up the spread between the 2 securities?

Yes, Maurice. Look, we obviously don't know what will happen. So that would be purely speculative I think at this stage. We're just focused on the objective that Dave mentioned, and we'll see what happens.

Our next question comes from Devin Dodge with BMO Capital Markets.

I wanted to start with Rockpoint. With the IPO now obviously completed, should we expect that to pursue more exits via the public markets going forward? Are there for the midstream assets or across the broader portfolio? Or was Rockpoint more of a one-off?

Devin, it's Sam here. Look, we've always had the public market as a potential exit strategy. We did do it with DBI a number of years ago. I think we just considered as 1 of the tools in our toolkit to monetize assets. And for a period of time, there wasn't the market conditions that allow for good execution of IPOs. I think that has changed. And to the extent that the market remains open, then we would very much look at it as a potential option. It doesn't change, though our views on favoring 1 or the other. It's really about execution, maximizing value and basically position the companies themselves for future growth and success. So we weigh all those factors, and we were thrilled with the outcome that we had, and we think Rockwood is going to be an amazing company in the Canadian markets for a long period of time. .

Okay. Second question, I wanted to ask about CenterSquare. Look, lots of progress after combining a Volcan Xtera, and there were some more sites added more recently. Just wondering what's the investment basis from here? Is this a platform that you're going to continue to add scale? Is there more proven opportunities? Or has a lot of the heavy lifting already been done?

Again, I'll start, and I don't know, Ben might jump in as well. Look, I think we're just at the beginning. I think this is an unbelievable business, and the opportunities haven't been more favorable since we've owned it, which is -- it sounds crazy given how well it's done. I think we've increased EBITDA there 4x over the last number of years through acquisitions and leasing of space. But what's really exciting is the fact that many of these sites, which are legacy telco sites, are significantly overpowered, and we have a tremendous number of under-roof expansion opportunities that have built multiples in the 3 or 4 times, which is just unheard of in this sector. So we're going to take advantage of that.

I think the capital deployment opportunity in the business, Dave, correct me if I am wrong, is in order of magnitude of $300 million to $400 million over the next couple of years of CapEx. And so you can kind of do the math on how much EBIT that can drive. In fact, the opportunity may even be bigger than that, I'm probably understating it. But suffice it to say, and I know I'm running on here, the company is incredibly well positioned. It's unique in the market in the sense that there's not really many others who are serving that smaller 1 to 5-megawatt demand from customers. And I think this -- the explosion in AI inference and agent models is going to provide us tons of customers, along with the corporate customers. So it's a great story, and I think there's lots more to come.

[Operator Instructions] Our next question comes from Patrick Sullivan with TD Cowen.

Can you just talk about the level of market interest you saw in the stabilized data center portfolio that you monetized during the quarter? I guess, how are you making decisions on the size of the portfolios you're bringing to market right now?

Yes, I can start, and it's David here. And just to make -- I'll repeat the question, I think it's around the sizing of our stabilized data center program and how we determine what to bring to market. Look, I think this was just the first of hopefully several programs that we undertake over the next few years. Data for -- in Europe had probably the largest operating portfolio of data centers when we acquired it. So it was a logical candidate to be the first for our capital recycling initiatives within it. .

As you would have seen, it was over 200 megawatts of operating sites. They weren't all commissioned, so we still have some build-out to complete. So think of it as a bit more of a program where over time, we'll continue to execute on the sites as we commission them. But essentially, we were able to identify the program in terms of the perimeter of assets that we felt would be out of revenue generating today or generated in the next 6 to 12 months. And we felt that for -- it was about $1.4 billion equity check. We thought that was the right size in the market in Europe to target financial investors to come in and partners to join us to own these types of assets. And we're seeing significant demand for this return profile in the market. So as I said, Europe will be the first of our programs, and we'll look to replicate that in other markets as well.

Yes. The only thing I'd add is we've returned capital 2 ways. One is through issuance of ABS securities, which we do very programmatically almost monthly, it feels like, but let's just say, quarterly as the facilities get established. And then similarly, we'll -- our goal would be to -- as facilities get completed to programmatically sell down pieces of the equity quarterly or semi-annually, as they get built because that's that will be what the profile will look like over the next number of years.

Okay. Great. And then, I guess more on data center stuff. So like in a recent Brookfield podcast on the Data4 deal, you discussed that Europe has 1 of the largest infrastructure gaps relative to the other regions and essentially no sovereign compute. And you've made some announcements related to these sovereign compute opportunities. But can you just talk about some of the differences between that sovereign compute opportunity set versus the more hyperscale AI lab driven opportunity set that we see in the United States. Is there anything there to contrast and compare between the two?

Yes. Look, I think the -- I think they're both very exciting. For us, we've had a focus on the sovereign compute primarily because, as I kind of alluded to in some earlier answers, it's a way to use our particular skill set to create bilateral transactions in the sector, where we can come up with more complex solutions for governments to bring all the various players together to solve what for them is a sovereign issue in a sense that they want to retain data as well as AI capabilities in their home markets. and not be dependent on just all the U.S. hyperscalers. And so it really is a distinct market. Obviously, we still focus on servicing our hyperscale customers, but we thought we could create kind of a niche market for ourselves working with -- today, it's probably 6 or 7 sovereign nations on building smaller, but more dedicated facilities for their home needs.

And we're excited about the progress we've made. The only drawback to it is often dealing with sovereigns tends to be a bit slower. And the investment cycle has probably taken longer than we might have thought initially. But we're excited about the projects we're working on, and we hope to have announcements in the coming quarters.

I would now like to turn the call back over to Sam Pollock for any closing remarks.

All right. I guess that -- it looks like there's 1 more question we can take.

One moment for questions. Our next question comes from Frederic Bastien with Raymond James.

Appreciate it. Guys, are you able to quantify the organic growth rates your various data businesses are enjoying. And how would these be tracking versus your underwriting assumptions?

We're happy to start, it's David here. I think -- look, I think the overarching team across the data businesses and maybe we'll stick to towers and transmission and then data centers are 2 separate categories. I'd say on the transmission and tower side, I'd say are -- I would say that's a much more predictable, stable execution of our backlog there. We are building out towers for our customers in France, in Germany, some in India as well in terms of rooftops and antenna. So I'd say that the case there is much more predictable. And I'd say it's slightly ahead of underwriting, but generally in line. So that's going well. .

When we bought these hyperscale platforms, we underwrote a land bank that they had in place, and we've executed and continue to actually that on schedule and on pace. And I think we're excited about what the next, call it -- what we call the shadow backlog looks like in these businesses, where we could see up to a gigawatt across the globe of new projects coming in the coming years. And that is something we never underwrote. So the pace of growth and organic growth in those businesses will be dramatic. Like the percentage, I don't think is that meaningful because it's coming from such a small base in these platforms, but that's what you're starting to see come through the numbers. I think in the last year alone, Fred, we've commissioned 175 megawatts across the globe, which is pretty impressive.

That's great color. Just building on that, you flagged some promising AI factory opportunities shaping up for a bit during your Investor Day. And I was wondering if these are along the lines of the partnership you signed with Bloom Energy to install BTM solutions? Or do they vary depending on the partnership you're pursuing?

Brad, look, as you can imagine, they vary. I think there are some that are very similar to the Bloom Energy arrangement, where we're facilitating capital needs for a number of the data center and hyperscaler companies to finance all the capital that they need to put into these data centers. But then, obviously, building out AI factories is going to be a different type of arrangements, and we expect that probably to be the majority of what we do in our AI activities.

Okay. So I think that's probably it. I'm glad we got Fred in there. So yes, thank you, operator, Josh, for today, and thank you to everyone on the call for joining us. I know this is our last call before the end of the year, so on behalf of everyone here at Brookfield Infrastructure, we'd like to wish you a healthy and happy upcoming holiday season. And we look forward to providing you more updates on the fourth quarter and year-end results early in the new year. So take care. Thanks. .

This concludes the conference. Thank you for your participation. You may now disconnect.

Well, good afternoon. I guess the only bad news is unfortunately there's not going to be a break that I think we had planned at this moment, but it's been a pretty exciting day so far. We had Bruce Flatt, we had fire drill, and then BBU just wowed at us just now. So, we're going to try and keep the momentum going and welcome you to Brookfield Infrastructure's portion of today's event.

The theme for our presentation is growth, and more specifically our view on our current growth trajectory. So, I'm going to begin by providing context on our historical growth and where we think we're headed. Essentially, what we're going to describe is our inflection point. My colleagues will then come up and provide more details on our playbook and how we're going to achieve that growth.

To set the stage not dissimilar to what [ Anuj ] did, I want to start with what our mission is and a lot of you would've heard this in prior years, but essentially it's to own highly contracted and regulated businesses that generate long-term consistent growth with minimal variability. And if we do that well, we should be able to generate FFO growth of about 10% or more annually, and this will lead to growing distributions within our 5% to 9% target range.

Now, to meet our goals, we're going to execute a full cycle business strategy that has three primary pillars. The first is we deploy capital at or above a 12% to 15% hurdle rate. Second, we'll crystallize value through our capital recycling plan. And third, we'll maintain a strong financial position to ensure that we always have ample liquidity to take advantage of opportunities that may arise. This approach has delivered a long-term track record of stable and growing cash flow with FFO per unit increasing at a compound annual rate of 14% since our inception. This in turn has allowed us to increase distributions for 16 straight years at a compound rate of 9%.

Now, while we're proud of our long-term record, we recognize that our share price will reflect either current and expected growth rates. So, it's a fair question to ask, how have we done lately? So we thought we'd look at a more recent time period, if we just look at the past five years, we've still delivered strong financial performance across all our metrics. Absolute FFO has grown about 13% per annum, cash flow per unit has increased 10% annually, and we've reduced our payout ratio by 11% from 78% to 67%. These are good results when you compare them to our Canadian midstream and utility peers that have generally grown their cash flow per unit over the same period by about 5%, but it's still lower than our long-term average of 14%.

And there's probably two macro reasons that largely explain the difference. The first headwind our business faced was as a result of being an international company was the tremendous strength of the U.S. dollar. From 2020 to the beginning of this year, we saw the U.S. dollar strengthen approximately 15% versus global currencies. In terms of our results, if you adjusted for the impact of FX, we would've delivered a 12% FFO per unit growth instead of the 10% annually.

The second headwind was higher borrowing costs from the rising global interest rates. As everyone knows, the world transitioned from a decade of low interest rates into a higher rate environment at a very rapid pace. The federal funds rate increased by 500 basis points from the trough to the peak, and similarly, the U.S. 10-year treasury rose by a magnitude of about 400 basis points. If you adjusted our results for the impact of higher interest rates, our FFO per unit would've increased by 2% to 3% with some offsetting benefit, obviously, from higher inflation in our revenues.

Now, longer term, we are better off having had the benefit of higher inflation, as that will continue to compound in our results over the long run, but in the short run, we were negatively impacted.

Okay, so you're asking why am I telling you all this? Well first, the macro environment has changed and our business is better than ever. Today, we are at an inflection point and we think we'll see substantially higher growth rates in our business than we've achieved in the last five years.

So, let's start with how the business has evolved. In the last five years, the embedded organic growth from our business, which is reflected in our capital backlog, has grown 4x. This is important because our organic growth is the highest returning investments that we can make in the business. In a similar vein, the number of high growth platform businesses that we own has more than doubled. And lastly, our asset rotation program is also 5x bigger, allowing us to self-fund our growth, which is our lowest cost source of capital to fund new investments.

The strength of the business and the strategy execution has also been showcased during 2025. We've made significant investments in growth already exceeding our annual goal, deploying $2.1 billion, $700 million of that has been deployed into organic growth projects, and $1.4 billion has been deployed into four new investments at returns we believe will exceed our target returns of 12% to 15%. We have also been successful this year in recycling capital to self-fund our growth. We have already secured $2.8 billion of sell proceeds, and that's an annual record for BIP. This includes eight transactions that involve either full or partial exits, and all together these asset sales have results where we've achieved a 20% IRR and a 4x multiple capital. These are terrific results and a testament to our ability to crystallize value.

Now, in terms of the macro environment, several factors are trending in a positive direction, from our perspective. First, in most parts of the world, interest rates are stable or decreasing, a positive for our portfolio. One recent example of the benefit of lower rates is a five-year BIP bond that we did just this week where we issued at a coupon of 3.7%. Just two years ago, in 2023, we issued a similar bond that was over 200 basis points higher than that. These are rates that we saw back five to seven years ago when rates were much lower.

Another favorable trend has been the softening of the U.S. dollar. For the first time in many years, FX rates may no longer be a headwind for us, and in fact, if this trend continues, it may become a tailwind.

And then lastly, as we've been saying for a few years, we are in a massive investment cycle across all our segments. We've been calling this an infrastructure super cycle. Now you've probably heard lots of big numbers, but it's estimated that the world needs a $100 trillion dollars in investment going up to 2040 in order to meet the new and updated infrastructure requirements across the world, and that probably is underestimated, given that AI estimates are only growing from year to year to year.

So, let's just bring this all together. BIP delivered strong FFO growth of 10% annually over the past five years in spite of interest rate and FX headwinds. We believe that with FX and interest rates normalizing, along with the strategic enhancements that we've made to the business, this will result in BIP returning to annual growth rates for the next five years approaching where we have those long-term rates of above 14%. Higher growth rates ultimately will enable us to increase dividends at the higher end of our 5% to 9% range, and we'll do that without increasing our payout ratio.

The content for the remainder of our presentation today will be on our execution playbook, and this is how we're going to deliver higher FFO growth. Now, first up is Lief Williams, a Managing Director in our AI infrastructure business, and he'll come up and preview AI infrastructure and explain why we believe this new investment category will be a massive opportunity for BIP. After Lief, Scott Peak, President of our infrastructure business, will come up and walk you through our deployment this year and how we expect to outperform our return targets. And then lastly, David Krant, our CFO, will discuss the evolution of our capital recycling program and how it creates value and growth for the company.

And with that, I'll ask Lief to come up.

Thank you, Sam. We thought we would start by recapping a little bit about what we said last year at this event, which was really two main themes. Firstly, the digitalization tailwinds continue to accelerate, and secondly, BIP is well positioned to participate both through our existing asset base and through new opportunities. With the passing of a year and the benefit of hindsight, we will be the first to admit that we may have underestimated the sheer scale of the opportunity set. The growth is exponential and is much, much larger than we imagined a year ago.

To take a couple of data points, Hyperscaler CapEx in 2024 was already at an all-time high at $270 billion. A year later, it's increased by 50% year over year up to $400 billion. We're now at the point where this is impacting U.S. economic statistics with this making up about 1% to 2% of U.S. GDP. That's already higher, for the sake of context, than the fiber build-out of the early 2000s and it's starting to approach the build-out of the railroads during the late 1800s. Now, what's driving Hyperscale CapEx is the AI race led by the largest technology companies in the world and this is supported by advancements in semiconductor chips.

So out of interest, to support AI workloads, current chips are using 10x more power density than for non-AI workloads. And so we're up to about 120 kilowatts per-rack relative to non-AI workloads, which are more like 10 kilowatts per-rack. This is already 2x higher than we were forecasting only a year ago. What's even more impressive is, if you look out five or 10 years, the leading chip designers are actually forecasting that we will increase another five to 10-fold from where we are today, and so this will be half a megawatt or 1 megawatt per-rack.

What's really important to emphasize about this is the fact that having more powerful compute infrastructure will enable more powerful agentic AI. For example, even the most powerful compute cluster in the world today is limited by memory bandwidth and by context length. And what we mean by that is that large language models will forget instructions that were previously provided or work that was previously done. And so having more powerful compute will enable us to push out the technology frontier and do things, rather than just a single topic query, it will be able to be a persistent partner that works for hours, days, or even longer periods of time.

And so with that backdrop, it will be no surprise that our data center platform has experienced a step-change in growth. We've sold effectively all of the inventory that we had available and so, in the last few quarters, we've secured additional development capacity. We will be able to deliver this product to our end customers for years to come. This land bank will provide long-term optionality. And what's really exciting about investing in the data center sector is that it hits all of the infrastructure attributes that we target. High barriers to entry, long-term contracted cash flows, investment grade counter parties, inflation protections, and ultimately attractive risk-adjusted returns. But it hasn't just impacted our existing asset base. We see a whole host of new opportunities to invest capital to support the unprecedented scale required to build out AI.

Over the next 10 years we see a $7 trillion investment opportunity in the physical assets supporting AI. This will start with AI factories, which is a next generation of digital hubs powered by specialized networking and liquid cooling to host hundreds of thousands or millions of chips. In addition, behind-the-meter power generation will be required, given shortfalls on the grid. We will also need to invest in compute infrastructure to support the AI factories and additional portions of the AI value chain.

So to dive into AI factories for a moment, what do we mean by this? Well, firstly, this is a $2 trillion opportunity set in and of itself. There are really two raw inputs that are required to feed AI: power and compute. These are also the two constraints that are slowing the deployment of AI around the world and delaying the technology frontier. AI factories are really an extension of our existing investment perimeter. We currently invest about $10 million per megawatt into building out cloud data centers. This would include the purchase of land, connecting it to the grid, investing in the building shell, and then investing in the mechanical and electrical systems that support the IT load inside the data center.

AI factories are all the same things, but even more. We would also need to invest an incremental $30 million per megawatt of IT load, so up to a total of 4x what we invest into a cloud data center. The largest part of this incremental investment is into the chips outright. Now, traditionally, chips have been funded by the data center customers, and so it's been on corporate balance sheets. But even the largest and best-funded corporate balance sheets are unable to address the physical and immense capital needs going into this space.

So again, to provide context, ChatGPT-4 was trained using 25,000 GPUs over a period of 100 days. A modern AI factory will host a million GPUs, and these are more powerful GPUs than the prior vintage of chips. And so the scale here is enormous and technology companies are looking for partners like Brookfield, who have experience owning and operating these assets and have access to capital to deploy into the space.

So, we wanted to spend a few minutes talking about some of the AI factories that we are developing. There are seven AI factories that we are pursuing in five countries. And one thing that's very exciting about the current AI market is that the universe of customers has expanded. Traditionally there were maybe a handful of hyperscale customers who could be the tenants for a data center. Today we see native AI companies, a wave of enterprise companies, and perhaps most interestingly, sovereign governments themselves are looking to be off-takers for this infrastructure.

Governments around the world are looking at AI, realizing that, firstly, for data sovereignty reasons as well as data security reasons and then, perhaps most interestingly, the transformative economic potential that AI offers, AI should be at the top of their agenda. And as a result, they would like to support the buildup of the infrastructure in their region.

The seven AI factories will -- in total comprise 6 gigawatts of compute, 3 million GPUs, and $200 billion of total capital deployment. Now, we realize that these are staggering numbers and so, to contextualize, $200 billion would occur over time. We have selected these sites to be future proof, where they can be expanded to meet our customer needs over decades to come. In addition, it includes both debt and equity. And in each of these regions we are partnering with technology companies, and so ultimately the equity would be shared between ourselves and our partners.

The other thing to note is, in terms of the GPU deployment and the leases that we have with our customers, this will follow exactly the same infrastructure model that we have with the broader data center. And so, these are take-or-pay contracts with investment-grade counterparties, and no technology risk.

To sum it all up, the AI infrastructure super cycle is here. This is a multi-generational investment opportunity. Just as the electric grid build-out took decades but was required to build out a modern world, AI infrastructure will also take many, many years, and this will be required to build out an intelligent world.

Brookfield is a partner of choice for large technology companies and for governments around the world. We are excited to be investing in this space at returns that exceed our targets and expect BIP to deploy $500 million annually to make this future a reality.

And so with that, we'll go next to Scott Peak, the President of our Infrastructure business.

Okay. Thanks, Lief. Well, hopefully you're concluding that it has been a very productive 2025 for Brookfield Infrastructure across all of our business. This year we made excellent investments. We've advanced asset management priorities and team activity levels remain very high. Therefore, we thought it's timely to share an update with you on our recent deployment activity and our deployment outlook for the business.

Let's set the stage. Three key messages. The first is that the infrastructure super cycle continues to serve as a deployment tailwind for our business, enabling us to be very selective. The second is we have a proven track record of executing marquee transactions opportunistically across cycles. And third, our active asset management approach generates a return premium incremental to our initial underwriting. And together, we continue to deliver investment returns above our stated targets.

Our results validate that we're focused on the right themes in the right regions at the right time in the market. Digitalization, deglobalization, decarbonization, our longstanding three Ds now, remain the most topical for our business. Each of these Ds continues to require an enormous amount of capital, creating a deployment tailwind, expanding our opportunity set, and enabling us to patiently select only the best investments in front of us and, together, translating to return on performance. We follow the large-scale capital needs along the economic backbone of the regions where we invest. Ten years ago, our global investment activity was focused on ports, toll roads, gas and electricity transmission, and towers.

Then the opportunity set evolved, primarily due to three reasons. The first is technological evolution led to new businesses which did not previously exist. For example, fiber. The second is certain businesses improved their contracting models to attract large-scale capital, thus becoming more suitable investments for us. Think data centers. And third, with time, frankly, we became more comfortable with certain businesses that we did not previously prioritize, such as residential infrastructure.

And the evolution continues today. This year we've added AI factories, bulk fiber networks, refined products pipelines, rail car leasing, and industrial gas platforms, among others. We have a differentiated ability to apply our learnings across industry sectors, often from and to businesses that otherwise appear unrelated. And by doing this, we unlock opportunities that others, frankly, just don't see. Let's touch on a few examples.

You see here, and you'll remember our partnership with Intel. The first of its kind semiconductor partnership that we announced a few years ago. The market viewed it to be quite innovative, but the component parts of the transaction were simply adapted from our prior investments. We took the contracting framework from our hyperscale data centers. We took the project finance framework for investments like our LNG tolling facilities, and we took the governance and risk allocation from various large scale corporate partnerships that we've done.

A similar cross-sector approach was applied for our Hotwire acquisition where we weaved together our experience from fiber and residential infrastructure and stabilized asset carve-outs, which you'll have seen us execute on over the past year in which our European data center platform, as well as our container leasing platform. And cross-sector learnings were also applied this year to our rail car leasing partnership and our investment in Colonial.

Uncertainty often leads to market volatility, creating a buyer's market for those investors who resist the tendency to simply sit on the sidelines. These are the times when we at Brookfield, find that we are the most active and we move decisively with a focus on the highest quality businesses. The large marquee businesses we acquire are rarely available, particularly on a value basis. They perform essential services typically on a regulated or contracted basis without material cross border trade activities. The investments we make in uncertain market periods typically outperform.

During the pandemic, we acquired several excellent businesses at deep value. During the macro rate shock environment, while most paused for consensus on the trajectory for rates, we quickly acquired great businesses with limited to no competition. And earlier this year during the trade disputes, we acquired marquee businesses, which are already exceeding our expectations. And while our teams are busy during the uncertain times, there's no vacations at Brookfield, they're also busy in the interim periods characterized by elevated market sentiment when the sidelines clear. These are the periods where we opportunistically recycle capital from our mature businesses.

Our transactions this year covered our sectors and our typical deal structures. In July, we closed the 100% acquisition of Colonial, a value-based acquisition of a highly cash generative and strategic pipeline system, spanning 5,500 miles from Houston to New York. Earlier this month, we closed 100% acquisition of Hotwire, the leading platform of bulk fiber to the home infrastructure. And we signed a rail car leasing partnership with GATX, a leading corporate strategic and a corporate carve out of an attractive Korean industrial gas platform.

I'll now spotlight our investment in Colonial. Colonial is the largest refined product system in the U.S., supplying almost half of the demand of the U.S. East Coast. With a multi-decade history, as the lowest option for customers with utilization over 90%. Colonial is a large, critical, and marquee midstream business that we acquired at 9x EBITDA with a high cash yield. So how are we able to do this? One, the transaction timeline overlapped with the trade war noise, limiting credible competition, particularly at this scale.

Second is our pipeline and FERC experience enabled us to quickly complete due diligence and move quickly as a single buyer on an all cash offer during an uncertain market period, enabling us to successfully sign the transaction.

Four years ago, at this Investor Day, I shared a case study on our take private of Inter Pipeline, and I outlined our self-imposed midstream rules for our investments that we do in the space. I suspect some of you may have been in the audience then, and like Inter Pipeline, Colonial is a critical midstream business which checks all of our boxes. It's highly cash generative with a quick payback. It's FERC regulated with inflation linked tariffs, an investment grade capital structure, and we underwrote it conservatively with long-term declining utilization in a finite life. And importantly, Colonial came with significant untapped opportunity to apply our operationally intensive asset management approach and materially outperform.

Our business plan for Colonial is focused on bridging from our downside protected target return to our outperformance return by focusing on three main priorities. The first is to align incentives and accountabilities where we eliminate bureaucracy and inefficient work streams. The second is to enhance margins where we look to find win-win arrangements with stakeholders. We look to drive operational excellence while we ensure that safety and reliability remain paramount. And third, to prepare the business for a future opportunistic exit with optionality for diverse products.

Now onto Hotwire. Hotwire is the leading U.S. provider of bulk fiber broadband infrastructure to home and condo owner associations, long-term take or pay contracts, a self-funding model, and an efficient in-place capital structure. We are the natural owners of Hotwire because we bring our experience in residential real estate, our homebuilder relationships, and our complementary residential infrastructure platforms. We have deep fiber and securitized asset financing expertise and an ability to incorporate the learnings from seemingly dissimilar businesses. Hotwire had been in operation for 25 years as a single business.

Given our prior stabilized asset experience, we quickly recognized Hotwire would be optimized as two businesses. Business number one, a development growth platform, including a proven development engine, 2,000 employees, and 200 homebuilder relationships. Business number two, a stabilized portfolio of contracted fiber assets, long-term contracts, high cash yield and high renewal visibility. We split Hotwire concurrently with signing the transaction and exited the majority of the stabilized business. We partnered with lower cost of capital, consistent with the low-risk nature of the stabilized business, and together this unlocked value not historically recognized and not shared with the sellers.

So, with that structural change behind us, what's next for Hotwire? Three-step outperformance plan. One, continue to sell down stabilized assets and recycle capital. Two, expand partnerships with large homebuilders and developers. And three, leverage the Brookfield ecosystem so that we can enhance Hotwire's customer service offering. And if we do this, Hotwire is well-placed to serve as another innovative and outperforming platform for us.

So, in conclusion, our business is ideally positioned. We have a robust and attractive opportunity set in front of us, and we have a differentiated ability to both originate and add value. And taken together, we expect to continue to achieve returns above our stated targets. Thank you.

We remain excited for what's next for Brookfield Infrastructure, and I'll now hand the stage over to our CFO, David Krant.

Thank you, Scott, and good afternoon, everyone. Now, last year we identified a $5 to $6 billion target for our capital recycling program. Just nine months in, we have delivered $3 billion of capital recycling proceeds, and we've done that while crystallizing excellent returns. So today I'm going to focus on how we measure success of that program and how this year's sales alone could unlock up to $6 billion of value for our shareholders in the years to come. Much of it behind the scenes and outside of our key performance metrics. Our capital recycling program has scaled with the growth of our business. It is now a key differentiator that will fuel the next phase of our growth that you've heard today.

Let's take a look at what we've achieved so far this year. We've sold $2.8 billion worth of businesses. We've done this across eight exits, they spanned six different countries, one from at least each of our segments, and truly demonstrate the broad-based global demand for high quality infrastructure businesses like ours.

Now, in addition, we've done it in a variety of ways. Looking at our toolkit, which won't surprise anyone, we have leveraged a variety of techniques to maximize value for our shareholders. As you heard, we've used stabilized asset pools for the first time this year, identifying it in our European data centers and within Triton. That's a great way of pulling out a lower risk, lower growth stream of cash flows from a broader platform to find the right buyer. It generally results in a broader investor universe and a better outcome for us as sellers.

Now, I'm going to focus the next little while on the three key ways in which recycling capital in our business creates value for us. First, it reduces the reliance on public markets to fund our future growth. Second, it creates a significant amount of value that we build and compound during our investment period. And finally, it is an accretive source of perpetual capital that we can use to redeploy and grow faster.

We'll go through each of these in succession, starting with our funding sources. We've broken up the last 15 years into three distinct five-year intervals. As you've seen our investment pace ramp up, so has our capital recycling. In fact, in the last five years, we have funded over 85% of our new investments with asset recycling. In fact, if I shorten that period to just the last three to four years, 100% of all of our new investments were funded with capital recycling proceeds. That's right, every single dollar we deployed into new investments was funded internally.

Now, the second pillar of the value creation we have here is that it crystallizes the value that we have compounded over our investment period. Now investing at 12 to 15% targets like we do is good, but what's better is if we're able to compound those returns over longer periods of time, or we're able to unlock more value and earn a higher return. That's when it becomes exceptional. I think that's what I'd characterize this year as. Across the $2.8 billion of proceeds, you can see our average IRR was 20%.

Now considering our hold period for this portfolio averaged about eight to nine years, that translated into a multiple on our capital of 4x. These are great returns by any standard, but what's more impressive to me is the fact that these were earned on either regulated utilities or highly contracted businesses. How do these returns get captured in our reported results? Well, the truth is they don't, and that's what I'm going to elaborate on in a little bit now. In terms of how this profit we earn on these investments gets captured in net income or in FFO.

Starting with net income, as we've said in years past, this measure is not great for how we track our business on an annual performance because most of the FFO or cash flow we generate and distribute is offset by amortization or depreciation. What that means is that over time, our book value or carrying value of our investments tends to deplete or reduce over time, resulting in a large gain on exit. As you can see here across these four assets, when we dispose of them, we'll recognize a total gain of about a billion dollars net to BIP.

Now, you might be wondering, "I thought you revalued your assets under IFRS?" Well, it's true. We do follow that standard. However, there are limitations to what IFRS allows you to revalue. To put it into non-accounting terms for most of those people in the room, our standards allow us to value the assets in place, the physical pipes, the processing units, the data centers, etc., but it doesn't allow you to value the business. As many know, there's a lot more that goes into running a business than just physical assets. In fact, it doesn't allow us to revalue the growth ahead of it, the intangible customer relationships we've built over time or the platform values we've built during our 10 years of ownership in many cases.

As a result, when you go to exit these businesses, you tend to sell them at premiums. This year, that's no different. We recognized a 2x uplift over our book value on exit. What's even more interesting is our key performance metric, funds from operations or FFO, also doesn't capture this benefit either.

To highlight this, what we've done is we've taken our realized multiple of capital, which is simply just the cash on cash return we earned over the investment period, divided by our day one investment, and we use that multiple to imply the realized profit that we earned on that investment. It's simple. For each of them we've highlighted that in the middle there. We then compared that to the cumulative FFO that we recognized on those same investments over our investment period.

What it tells us is that no matter which investment we look at, there were hundreds of millions of dollars of profit that we earned as the investor that never went recorded through our FFO. Meaning had we been able to capture the benefit we created over time, our reported results would've been much higher and grown much quicker than they had. Now, recognizing these limitations is the very reason we focus on IRRs and multiples of capital when we're explaining how we did as investors.

Now, the third and final value creator, and probably the most important here today, that I want to highlight is how we can create significant value when we take the proceeds from our asset sales and redeploy them into higher returning investments. This boils down to sources of capital. In terms of cost of capital, we've outlined where we think our target cost of capital is for asset sales. Obviously, it's going to be a little more expensive than issuing corporate debt or preferred shares, but we believe it's highly accretive when compared to the cost of issuing new equity to fund our growth. This year's program certainly outlines that merit.

In terms of some of the key headline metrics we delivered, our sales this year triangulated to an FFO yield of 7%. Candidly, that's a helpful number in the short term, but tells us nothing of importance over the longer term. Instead, looking at our implied EV to EBITDA multiple on these sales and the buyer's cost of capital can be much more valuable. The 10% to 11% buyer's cost of capital we achieved is an estimate. The buyers of our businesses are going to make assumptions on their own of how they expect these businesses to perform in the years ahead. However, we can imply them by looking at our business plans for these businesses and using the price they paid. This year, we averaged about 10% to 11% for that cost of equity.

I'm going to touch on what that means in a second, but first we thought we would just highlight the disconnect that we're seeing again between the public and private valuations of our businesses. Interestingly, when you compare that 15x EV to EBITDA multiple that we were able to achieve on our program, which again, not to repeat myself, was across four segments and six countries, so a pretty broad group of our assets, that 15x EV to EBITDA multiple is three to four turns higher than where BIP overall trades at, meaning there's quite a premium in the private markets for the types of assets that we own.

Pivoting back to that 10% to 11% cost of capital, what does that tell us? Well, it's a really important metric in how we decide or determine how much value we think we can create in the years ahead. To determine that value, there are three key inputs that you need. The first is that cost of capital, so for illustrative purposes, we're using 11% this year. Second is the size of our recycling program. As you've heard, we've delivered nearly $3 billion this year. The third is our expectation of what we will earn in the years ahead when we redeploy it. As you've heard, we've had some of the best investments we've made in the last 24 months that we believe will exceed our target returns. For illustration purposes, we've used a range of 15% to 17%.

When you look at the value you create by reinvesting that $3 billion at 17% relative to the 11%, it costs us on our balance sheet, we think we can create up to $6 billion of incremental value by doing this program. Importantly, that's just this year's sales alone. That is the benefit of this asset recycling programs that we're constantly selling de-risked, mature businesses and reinvesting them into higher growth, higher returning opportunities.

Now, to sum up our capital recycling initiatives this year, there are three things I want to leave you with. The first is that we will meet our $3 billion target for this year. In fact, we're already there. Second, we believe we'll sell a further $3 billion of assets in the next 12 to 18 months and then going forward a $2 to $3 billion annual run rate for our program. Finally, we expect to create significant value when we redeploy those proceeds into higher returning opportunities.

So with that, I'll now stand back up for closing remarks and questions and answers. Thank you.

Well, thank you for listening to our presentation, and I'd like to leave you with a few thoughts before we start Q&A. We coined a phrase that BIP was a "grow-tility" a couple of years ago, and we believe that's not only the case, but that we're an even better one today. As we think about the next five years, what we've talked about is the fact that we're in a moment in time when we have all the macroeconomic trends going in our favor, including the huge investment opportunity related to AI infrastructure.

In addition to that, we've had the benefit of all the enhancements that we are putting into the business that we've been able to execute, which is investing at higher returns and also recycling capital at a higher rate, which we think will compound value and increase growth over those next five years. If we do that, that's going to lead to higher FFO growth, which ultimately leads to higher dividend increases.