Dan Primack: IPO market is down 20% year-over-year, VC median returns trail the S&P 500 for 25 years

Summary

Dan Primack, Axios

VC median returns have trailed the S&P 500, the Nasdaq, and the Russell 3000 for 25 years. Primack describes the industry as running on hope over experience, a characterization he attributes to an LP he spoke with. The top 5% of funds still generate strong returns, but more capital flooding into the asset class means more money sitting at or below that median, which is a problem for the pension funds, endowments, and insurance companies increasingly exposed to it.

401(k) exposure

The Trump administration is moving toward allowing 401(k) plans to invest in private funds, a change the VC and private equity industry has wanted for years. Primack is bluntly skeptical. At the very moment some of the smartest institutional money, Yale and Harvard endowments among them, is doing secondary sales to reduce concentration in private portfolios, retail defined-contribution capital is about to rush in as replacement. Those endowments are not fully exiting, but they are signaling they were overweight.

IPO volume

Despite a brief January uptick, IPO volume is down more than 20% year-over-year in both pricings and filings, from what Primack already describes as a weak 2025. The market is not dead so much as concentrated. Databricks is the name he says people are sleeping on, and he considers an H2 listing almost certain. OpenAI and Anthropic are the headline race, with Primack giving OpenAI the edge partly because CFO Sarah Fryer has done this before, having helped take Square public and run Nextdoor. He thinks the market will be shocked when it sees the actual loss figures in black and white from either company.

The deeper problem is cultural. Companies have been scared to go public for years, and Primack does not think that has changed. There is always a reason not to, and this cycle it is the SaaS selloff, Middle East tensions, and market choppiness. Private markets now solve both traditional reasons to IPO. Balance sheet capital is available privately, often more easily than through a public offering, and secondary tender programs handle employee liquidity. Stripe runs one annually, including one roughly two weeks before this conversation. Primack says if the Collison brothers ever do take Stripe public he will be stunned. They clearly do not want to, and Sequoia building a continuation vehicle to buy out other Stripe investors rather than pushing toward an IPO reinforced that read. When he asked a Sequoia contact how they would ever get liquidity, the answer was essentially faith.

Michael Grimes returning to Morgan Stanley fits the same logic. SpaceX is coming, and the value in a landmark IPO is not the fee on the deal itself but the secondary offerings, debt issuances, and M&A mandates that follow, along with wealth management relationships with newly liquid founders and employees.

Secondary market

The secondary market has grown from a cottage industry a decade ago into a core piece of private market plumbing. Major PE firms have moved in, with Carlyle buying Alpenvest and EQT acquiring Coller Capital. Without this channel, Primack argues the broader VC ecosystem would start to grind because so many LPs have not seen meaningful cash distributions via traditional exits. DPI has been scarce, and secondaries are the pressure valve.

Private credit

The private credit selloff looks like a liquidity story rather than a credit story. There are no significant defaults in portfolios yet. What is happening is redemptions as investors try to get out. Insurance companies are the dominant LP base in private credit, and if they run into trouble the spillover is severe. For now, fund structures can gate redemptions in a way banks cannot, which managers treat as a feature.

VC firm IPOs

Primack names General Catalyst as the most likely VC firm to go public. It has already evolved well beyond venture, owning a hospital, running a wealth management business, and increasingly resembling a diversified asset manager on the Blackstone model. Andreessen Horowitz is possible but he is skeptical, noting that quarterly earnings calls do not fit how Marc Andreessen or Ben Horowitz operate. Sequoia has been discussed for years and shows no appetite.