Eighty-seven billion in LP secondaries, pricing at eighty-nine percent of NAV — and not one of those figures means anything until I know who counted it and as of when. This is Infrastructure Secondaries Daily. Today: a second named volume source finally lands next to Jefferies, a seven-year-old ILPA playbook nobody's auditing, and a dry-powder number that might explain the whole pricing story. And the question I can't shake — if both reports show records, are they even counting the same deals? Let's start there. So Setter's H1 2025 report, 25th edition, gives me a second named source for the same period as Jefferies. Which is exactly what I needed — because now I can ask whether they're measuring the same market. They're not. There's no central exchange, so Setter's counting brokered flow and Jefferies is tracking adviser deals. Two different networks, same calendar — of course the totals diverge. And nobody reconciles them. We get a headline, “record volume,” and the methodology footnote that would let you trust it lives in two PDFs that never talk to each other. Here's a clean example. Commonfund cites Evercore at two-sixteen in buyer dry powder. Jefferies cited two-eighty-eight yesterday. Those aren't the same number — and they're not even measuring the same thing. Seventy-two billion of disagreement, and both get printed as “capital on the sidelines.” Pick a denominator, people. But sit with the two-sixteen for a second. If that's right and capital's actually scarce — Commonfund literally says growth is constrained by lack of investable and human capital — then the buyer with the best sourcing network sets the price. And the pension fund with a liquidity mandate is still on the wrong end of that. Scarcity doesn't help the forced seller. It hands pricing power to whoever already knows where the deals are. So I don't read eighty-nine percent of NAV as proof the assets got better. With that much buyer capital out there, the compression in infra pricing looks like a supply-of-capital story wearing an asset-quality hat. Right. The marks didn't improve. The auction got crowded. Now here's the story nobody's filing. ILPA's GP-led restructuring guidance is in today's rundown — published in 2019. Fairness opinions, LPAC engagement, disclosure standards. Seven years old. And right next to it, Commonfund calling 2025 a “blossoming new era.” So in a record-volume year, has anyone actually checked whether the 2019 architecture showed up in the documents? That's the audit I want. ILPA put LPAC engagement and disclosure in paragraph one back in 2019. Volume's roughly tripled since. Did those protections scale with the deals, or did they get left behind? And the timeline is pretty clean here — ILPA asked in 2019, the SEC didn't mandate a fairness-opinion floor until 2023. That's four years of the industry running on volume growth before the regulatory floor caught up. Commonfund points to GP-led funds as a primary driver. So the adviser conflict ILPA flagged in 2019 is still growing in real time, deal by deal. And now with a new beneficiary at risk. Commonfund ties this growth partly to forty-Act fund expansion. So retail capital's flowing into GP-led structures — without the LPAC protections ILPA was writing for in the first place. There's the Setter detail I actually want, though — it breaks out the buy side by profile and the assets purchased by type and geography. That texture tells you who deployed, not just how much. Which finally gets at the question I keep asking — who was this liquidity built for? The breakdown's right there in the report instead of buried under one aggregate headline. And against that, Pantheon's April piece reads record 2024 volume as straightforwardly good and projects more. Volume tells you about liquidity appetite. It doesn't tell you the assets are worth eighty-nine cents. Bigger numbers, same blind spot. We've spent a week stacking records — today's the day to ask whether the people reporting them are even measuring the same market. This one's from Setter Capital:
This report summarizes the results of our 35-question survey of the most active global buyers in the secondary market for alternative investments, conducted at the end of June 2025. Volume is defined as total exposure (NAV + unfunded in USD) purchased by the respondents, including only deals where a binding agreement was entered into during H1 2025.
Setter's H1 2025 report — 25th edition — gives me a second named volume source for the same period as Jefferies. And that's exactly what I've wanted all week: a way to test whether these two numbers are even measuring the same thing. They're not, though. Setter surveys principals directly — brokered flow, deal by deal. Jefferies tracks adviser deals. Two different sourcing networks, same calendar, structurally different totals. Right, and nobody reports any of this to a central exchange. So when both point at a record H1, the headline says growth — but there's no way to reconcile the numbers underneath it. Two surveys, no referee. And watch the dry powder figures collide. Commonfund's got Evercore at 216 billion. Jefferies gave us 288. Those aren't measuring the same pool, and they're sitting in the same conversation about why infrastructure pricing compressed. Which is my problem with the compression story. If buyers are drowning in capital, tighter infra discounts can come straight from excess buyer demand, dressed up in an asset-quality jersey. And the Setter breakdown by buyer profile tells you who's sitting on that capital. Deepest sourcing networks set the price. The pension LP with a liquidity mandate is still on the wrong side of the information gap, record volume or not. When a Jefferies or Campbell Lutyens report drops a headline number — say, infrastructure secondaries hit $25 billion last year — how are they actually counting deals that are private by definition? And why might two firms looking at the same market land on different totals? Great question, and this is why you have to read the methodology before the headline. None of these transactions are reported to a central exchange — there's no TRACE for private fund stakes — so every firm is building its volume estimate from a different data collection method. Jefferies, for example, counts deals that flow through its own advisory pipeline, then supplements that with disclosed transactions and counterparty surveys. Its H1 2025 report put total secondary market volume at $103 billion for the first half alone, a 51% jump from H1 2024 — but that figure reflects what Jefferies could observe or verify, not a comprehensive regulatory dataset. The infrastructure-specific slice is even harder to isolate, because many LP-stake transfers involve diversified real-assets funds that hold infrastructure alongside other asset types. So one analyst might classify a deal as infrastructure; another might log it as real assets or energy transition. Allianz Global Investors, citing third-party data, put 2025 infrastructure secondary deal volume at approximately $25 billion, while the broader secondaries market hit a record $240 billion that same year, per Ropes and Gray's Q1 2026 market commentary. Then you get definitional choices: what counts as infrastructure, whether a GP-led continuation vehicle gets booked in the year it closes or the year it prices, whether partial-portfolio sales are counted whole or pro-rated. Stack enough of those choices, and headline numbers can diverge even when everyone's describing the same market. So if the underlying data is self-reported and deal-definition-dependent, should listeners treat these volume numbers as directional signals rather than hard facts? Exactly right — treat them as directional, but not disposable. They're high-quality sentiment data with a known methodology bias. I wouldn't treat them as audited figures. What you want to watch is whether multiple independent surveys are moving in the same direction at the same time, because that convergence matters more than any single headline. The pricing data — discounts or premiums to NAV with a specific reference date — tends to be more actionable, because it's anchored to actual bids on specific assets rather than a firm's best estimate of market size. From Institutional Limited Partners Association:
Such transactions require the Limited Partner’s full attention, but the timing of the process is often difficult to predict and is therefore potentially disruptive for LPs, particularly when multiple deals overlap. The structure of these transactions can also be highly bespoke, making evaluating the impact of an election to buy, sell or hold challenging.
This ILPA document is dated April 2019. Seven years ago they wrote down what a GP-led process should look like — LPAC engagement, disclosure standards, advisor conflicts — and then volume tripled. And read the framing: GP-leds used to be the zombie-fund parking lot, end-of-life or key-person situations. ILPA's own line is that the stigma diminished and they became, quote, solutions. Solutions for whom is the part the documentation has to prove. Here's what nobody's filing, Daniel — we just heard Setter and Jefferies pointing at record H1 2025 volume, and nobody is auditing whether the governance ILPA prescribed in 2019 actually showed up as that volume scaled. Their table of contents puts Advisors to the Transaction on page eight. The issue is the originating adviser blessing its own deal — ILPA flagged that seven years ago, and I'd love to see compliance data. There isn't any. What gets me is the timeline. ILPA asks for fairness-process discipline in 2019. The SEC doesn't mandate a fairness opinion floor until 2023. That's a four-year gap where the market ran on volume and self-policing. And the LPAC engagement piece — page five — only means anything if members get the fairness opinion with real lead time, not a weekend before the deadline. ILPA said that in 2019. The record-volume year is the stress test on whether anyone listened. Lawrence Shoykhet, writing in Commonfund:
Twenty-seven LP transactions greater than $1 billion closed in 2024, a meaningful increase on the 19 such transactions that closed in 2023.2 Moreover, aver-age pricing has ticked up by 400bps from 85 percent of Net Asset Value (“NAV”) to 89 percent.
Commonfund's headline is “$216 billion of dry powder among buyers,” cited to Evercore, and the expectation is all of it gets deployed in 2025. That's a specific, sourced number — and it sits right on top of the methodology question we just chewed through with Setter. Here's what bugs me: average LP pricing ticked up 400 basis points to 89 percent of NAV. If buyers are sitting on $216 billion they have to spend, that pricing move can be explained by too much capital chasing the same paper, not by suddenly better assets. And watch which numbers don't reconcile. Commonfund cites Evercore at $216 billion. Jefferies put dedicated dry powder at $288 billion. Those two figures are not measuring the same thing, and nobody's reconciling them. The line I can't walk past — they say growth is “constrained by a lack of investable and human capital.” So the bottleneck is what buyers can actually deploy into, and who can source it. The buyers with the deepest sourcing networks set the price. And the pension fund with a liquidity mandate is still on the wrong end of it. Then there's the framing word — “blossoming new era.” Put that next to the ILPA GP-led guidance from 2019 we just covered. Seven years old. In a record-volume year, is anybody auditing whether the fairness-opinion and LPAC standards from that document actually show up in the documentation? Right, and Commonfund names the volume drivers — '40 Act funds and the rise of GP-led funds. The retail channel is new here. So the beneficiary at risk isn't just the pension LP anymore. It's the retail investor in a '40 Act wrapper who never got the LPAC protections ILPA was writing for in 2019. From Amyn Hassanally at Pantheon:
Even given the prospect of a recovery in deal and exit markets that many had expected this year, sponsors and LPs are increasingly seeking alternative methods to manage their portfolios, with secondaries sales being one of them. Indeed, 2024 marked a record-breaking year for the private equity secondaries market. Volumes exceeded $160bn, passing the previous record of $134bn set in 2021.2
Pantheon's headline — “from strength to strength,” record year, more growth coming. And the engine they name for it is high rates choking off distributions, so LPs sell on the secondary just to get cash out. So the “strength” is people who couldn't get liquidity any other way. That's a supply story driven by stress, dressed up as a vote of confidence. And notice the dates, Daniel. This is an April 2025 piece, citing Lazard January 2025, Evercore January 2025, Preqin as of September 30. Every one of those numbers has a stamp on it — which is exactly why it lands so clean. But clean is the problem. After the Setter report we just hit, you've got Pantheon projecting straight-line growth off 2024 like the only direction is up. Record volume tells me capital wants in. It tells me nothing about what those assets are actually worth. Right, and “supportive conditions” for the buyer means there's a forced seller on the other side of the table. Pantheon's framing never names who's selling under a liquidity mandate — that's the half of the trade that doesn't make the deck. If Infrastructure Secondaries Daily helps you stay on top of the market, take a moment to subscribe or leave a review wherever you’re listening. It helps other people find the show, too.
We’ve put links to all the stories we covered today in the show notes, so if something caught your ear, you can head there to read more. That’s Infrastructure Secondaries Daily for today. This is a Lantern Podcast.