Somebody's bidding fifteen to thirty percent under NAV today — and for once, they actually told us how they got there. If you're just joining Infrastructure Secondaries Daily, here's where we are: the pricing arc's been running hot — 2025 volume framed around 226 to 240 billion, with GP-led continuation vehicles at roughly half. Buyout stakes have been printing around 87 to 94 percent of NAV; venture, closer to 78. So the marks, and the cost of liquidity, are still sitting at the center of every new print. And today: a tender offer that names its inputs, a six-way bid fight over tank terminals, and a bank quietly building a financing product on top of everybody's NAV. Let's start with Cox. Wall Street Journal Markets writes:
Data-center builders and operators across the U.S. are working with bankers to sell majority equity stakes worth tens of billions of dollars in their companies this summer, according to people familiar with the efforts.
Netrality, DataBank, Edged, EdgeCore — bankers are shopping majority stakes worth tens of billions, from Phoenix to Atlanta. That's the sell side lining up in the open. And here's what I can't get past: these developers are choosing to sell majority equity right into maybe the hottest bid they'll ever see. When the person who built the thing wants out at the top, I want to know what they see that the buyer doesn't. The thing I'd press on is the pricing basis. Owning the physical building behind AI is the pitch — fine — but a majority-stake sale has a reference date and a marked NAV somewhere in the deck, and none of that is on the wire. Six named operators, one banker-driven summer, and the number we get is 'tens of billions.' The demand is obvious. The as-of valuation on any one of these — that's the part I still can't reconcile. Jensen Huang's out there estimating the cost of a gigawatt of compute — and the shortages are real: electricians, turbines, memory. So replacement cost is climbing, which is exactly the story a seller wants in the market while they hand you the majority. Yahoo Finance writes:
The initial offers represent aggregate consideration of approximately $30.5 million, at prices representing discounts of 15% to 30% to each fund's reported Class I net asset value ("NAV") as of May 31, 2026.
Finally — a pricing sentence with actual inputs in it. John Cox at Cox Capital actually gives you the inputs: portfolio quality, liquidity characteristics, and observable traded-market discounts. For once, there's a methodology behind the number. And here's the number that methodology produces: Cox is bidding 15 to 30 percent below the May NAV. So, 'we don't have a negative view of these funds' — but also, thirty points off. But watch what he's leaning on, Daniel. 'Comparable traded BDCs trade at meaningful discounts.' Sure, there's an observable market for BDCs. For an infrastructure fund stake, there isn't a traded comparable — so if you borrow that logic, the discount gets pretty hand-wavy fast. Right, and look at why these holders are cornered. Q2 repurchase requests ran 13, 16, 14 percent — against a 5 percent program cap. They got prorated. So Cox shows up offering to clear the rest, and the rest is made up of people who already couldn't get out. Here's Noh Ja-woon, Kim Jong-yong at CHOSUNBIZ:
According to the investment banking (IB) industry on the 14th, KKR and sale lead Nomura Securities recently shortlisted five to six qualified bidders for Central Terminal Korea (CTK) and Japan's Central Tank Terminal (CTT). The bidders are currently conducting due diligence, and the main bid is scheduled for mid next month.
Finally, a tape I can actually read. KKR's selling the Korea and Japan tank terminals — CTK and CTT — and Nomura's shortlisted five or six bidders, doing real due diligence, with the main bid due in the middle of next month. An actual auction: arm's-length buyers on physical assets. Six-way competition for operating energy infrastructure tells you where buyer appetite actually sits — less on the AI-power headline than on boring liquid-cargo storage that can print fees for decades. And here, we have a named seller and a bid count. That's more than most of this week's wire gave us. But watch the Actis detail. Highest preliminary bid, per CHOSUNBIZ — and no confirmed record of ever running a liquid cargo terminal. They've done data centers and commercial real estate in Korea, not operating infrastructure. So the top bid may be the least equipped to close. Highest number, thinnest operating track record — that's the gap between a preliminary bid and a cleared price, and it's why I'd wait for the main bid next month. Right. A preliminary bid is a flirtation. Actually running tanks full of oil and chemicals is a different question from topping the price sheet in round one. Here's Private Equity Wire:
According to people familiar with the discussions, the bank’s fund finance team is proposing a structure under which limited partners (LPs) acquire subscription line facilities after they have been drawn by a fund. The arrangement would enable investors to earn interest on committed capital that would otherwise remain uninvested, while allowing Goldman to reduce the amount of capital tied up on its balance sheet.
Goldman's fund finance team is marketing a structure where the LP buys the capital-call loan drawn against its own fund. Moody's puts the fund finance market at $1.3 trillion now — so this is a product built to squeeze balance-sheet efficiency out of that pile. Follow the collateral, though. A capital call line is secured against LP commitments. Those commitments get drawn against assets the GP marked — the same GP-set NAVs we've been talking about all week. Right, and that's the escalation. When those marks were just feeding a quarterly report, a missing reference date was a disclosure problem. Now a bank is structuring financing on top of them — so the mark becomes what the whole facility rests on. And Goldman collects a spread in the middle. The LP earns interest on capital that would've sat idle, sure — but now they're financing a draw against a valuation nobody's independently checked. The opacity is still there; it just has a lender attached to it. Have feedback on today's briefing, a story idea, or a correction we should know about? Send us a note at infrastructuresecondariesdaily at lantern podcasts dot com. We read every message.
What we're watching next: main bids for Central Terminal Korea and Japan's Central Tank Terminal are scheduled for mid-August. We'll keep an eye on both as that window gets closer.
As always, we've put links to every story from today's briefing in the show notes, so you can dig into the pieces that matter most to you. That's Infrastructure Secondaries Daily for today. This is a Lantern Podcast.