Evercore says record H1 — a hundred and twenty-one billion. Biggest first half ever. And I've got exactly one question before anyone pops a cork. Quick catch-up before we dig in. We've been watching record volume, GP-led continuation vehicles, NAV discounts, tender limits, and financing risk. Goldman's Q2 update, via Finvaulta, pegged the venture secondary market at an annualized $112.2 billion — and flagged AI companies holding 45% of total U.S. venture value as of June 30. This is Infrastructure Secondaries Daily. Today, Evercore's record number leads us into a South African market that's building the plane mid-flight. And I've got a cap-rate spread that keeps me honest. Let's start with the headline. This story isn't over: Secondary market volume and NAV pricing. Follow us wherever you're listening, and the next chapter comes to you. Evercore writes:
First-half volume surpassed $120 billion, the strongest first half on record and up nearly 20% year-over-year. That builds on a landmark 2025, when full year volume topped $226 billion, a gain of more than 40% over 2024. The advance was led by GP-led activity at $65 billion, which outpaced a steadily growing LP-led segment at $56 billion.
So, the number's in — Evercore pegs H1 2026 secondary volume at $121 billion, the strongest first half on record, up nearly 20%. That's real. But set it next to dry powder: $194 billion, down only 10% year-to-date. Record deployment, and buyers barely drew down their reserves? Either fundraising is refilling the tank faster than they're spending, or the deals are smaller than the headline wants you to think. Neither one is a clean health story. And here's the split that actually tells you something — GP-leds at $65 billion, up 35 percent. LP-leds, $56 billion, up four. Four. GPs are solving GP problems nine times faster than LPs can rebalance. Single-asset continuation vehicles are the biggest slice of that, and that's exactly where a fairness opinion needs teeth — not a checkbox from an adviser angling for the next mandate. Evercore calls the pricing backdrop 'constructive.' Sure. Firm bids and narrowing spreads look great right up until you ask what date the underlying NAV was struck. Here's my thing on the 35 percent — once the market's this big, missing governance stops being a quirk. $65 billion moving through structures where the LPAC might see the fairness opinion days before the vote? At this scale, you've got a policy failure. Here's Newmark:
Newmark Valuation & Advisory has released the Mid-Year Update for the North American Market Survey, a semiannual, city-by-city survey of capitalization rates, discount rates and investment metrics across major North American commercial real estate asset classes.
Newmark's mid-year survey dropped yesterday — city-by-city cap rates and discount rates across North American real estate. Buried in it: the spread between prime and secondary industrial compressed from 72 basis points to 69. Semiannual, as of mid-year 2026. Small move, but it's a clean, dated one — which puts it ahead of half the pricing claims I read this week. Three basis points sounds like nothing. But if the market's paying almost the same for B-quality industrial as for A, the underwriting's getting lazy — and that complacency doesn't stop at real estate. It's the same argument I keep hearing to defend second-tier infra marks. Right — long-duration cash flows, so we'll wave the mark through. A tightening A-to-B spread tells you the buyer's stopped pricing the difference between good and adequate. And if the liquid corner of the market stops pricing the difference, the illiquid corner won't magically do better. That's what should worry the pension underwriting these things. Here's Dylan Cunard at Business explainer:
South Africa is sitting on a growing backlog of mature private equity assets that managers cannot exit and investors cannot access. The global solution already exists but we are not using it yet. Exits depend on a cooperative stock exchange, willing trade buyers, and investors patient enough to wait out a fund’s full life. In South Africa, none of those conditions are reliably available.
So the pitch is that South Africa should import the continuation-vehicle model — and Dylan Cunard at Werksmans calls the fund-manager-led version 'more impactful.' More impactful for whom? That's the part I'd underline. And look at the backlog they're describing — mature assets that managers can't exit. You're looking at those 2013-to-2018 vintages, exactly where the NAV mark and any real clearing price drift furthest apart. Right, and they want to drop the CV structure into a market that hasn't built the LPAC, the fairness-opinion standard, or a tender clock. You're handing GPs the exit tool before the pensioners have a single guardrail. It's a clean before-and-after picture, though. A market without independent fairness opinions leaves assets stranded and never priced. Put that next to the Evercore review we just covered — GP-leds up 35 percent — and you can see where that road ends. Cash out at 'fair value' — their words. Fair value as of when, and signed off by whom? Get that answer in the first paragraph or this is a wealth transfer with a nicer brochure. Have feedback, story ideas, or corrections? Email us at infrastructure secondaries daily at lantern podcasts dot com. We’d love to hear from you.
You’ll find links to every story in today’s show notes, so take a closer look at any that caught your attention.
That’s Infrastructure Secondaries Daily for today. This is a Lantern Podcast.