Key Takeaways
- • Family offices now direct 34% of their VC commitments into special purpose vehicles (SPVs), up from 9% in 2019, according to LP Breezer Clarkson as cited on This Week in Startups.
- • Emerging-manager fund closings fell from 927 in 2022 to 448 in 2025 and just 211 through the third quarter of 2026, leaving sub-$500 million funds starved of LP capital.
- • Ben Lerer of Lerer Hippeau predicts sub-$500 million funds will raise less money in 2026 than in any year in the last decade, while capital concentrates in megafunds.
- • Panelists warn SPV fee structures, often stacking to 15–20% over a fund's life, will draw litigation within one to two years, with Jason Calacanis predicting it and Jenny Fielding expecting transparency problems at OpenAI and Anthropic IPOs.
- • Lerer says AI produces excellent analysis but useless recommendations, and his firm's scoring system rarely yields a clear yes or no.
Summary
Venture capital is splitting into two extremes, and the middle of the market is under pressure, according to a This Week in Startups panel featuring Ben Lerer of Lerer Hippeau, Jenny Fielding of Everywhere Ventures, D.A. Wallach of Time BioVentures, and host Jason Calacanis. The panel cited data showing family offices routing 34% of their VC commitments into SPVs, up from 9% in 2019, and emerging-manager fund closings dropping from 927 in 2022 to 448 in 2025 and 211 through the first three quarters of 2026. Megafunds absorb nearly all LP capital, while sub-$500 million funds struggle to raise.
Lerer's structural view is a barbell. He said sub-$500 million funds will raise less money this year than in any year in the last decade, while capital flows into mega-platforms with strong recent paper performance. He also said LPs are cutting platform commitments because multiple funds now hold the same companies, and described the middle of the market as "cooked." Fielding predicted a "decimation" of many emerging managers, while Calacanis welcomed the reduction in less committed capital.
The panel was critical of SPVs as a fee business detached from track record. Calacanis said a family office told him a competitor charges 3% annually, which over a five-to-ten-year hold approaches 15–20% in total fees. Wallach described some emerging managers who could not raise new funds as engaging in "freelance investment banking," brokering late-stage allocations for fees. Calacanis predicted litigation within a year or two over off-book arrangements, nested SPVs, and fee opacity, and Fielding expects the issue to surface at IPO, when many buyers may not know whether they hold direct shares or interests in another vehicle.
On AI, Lerer said the analysis from his firm's models is excellent but the recommendations are not useful, since outputs cluster around middle scores on a 1-to-9 scale with no 5s. Calacanis expects AI to outperform most human analysts on analysis but said the human role remains building trust and staying available to founders. Pivots were described as normal, with Calacanis saying roughly half his companies pivot significantly and Fielding estimating at least a third of Everywhere's portfolio does.
Lerer's structural view is a barbell. He said sub-$500 million funds will raise less money this year than in any year in the last decade, while capital flows into mega-platforms with strong recent paper performance. He also said LPs are cutting platform commitments because multiple funds now hold the same companies, and described the middle of the market as "cooked." Fielding predicted a "decimation" of many emerging managers, while Calacanis welcomed the reduction in less committed capital.
The panel was critical of SPVs as a fee business detached from track record. Calacanis said a family office told him a competitor charges 3% annually, which over a five-to-ten-year hold approaches 15–20% in total fees. Wallach described some emerging managers who could not raise new funds as engaging in "freelance investment banking," brokering late-stage allocations for fees. Calacanis predicted litigation within a year or two over off-book arrangements, nested SPVs, and fee opacity, and Fielding expects the issue to surface at IPO, when many buyers may not know whether they hold direct shares or interests in another vehicle.
On AI, Lerer said the analysis from his firm's models is excellent but the recommendations are not useful, since outputs cluster around middle scores on a 1-to-9 scale with no 5s. Calacanis expects AI to outperform most human analysts on analysis but said the human role remains building trust and staying available to founders. Pivots were described as normal, with Calacanis saying roughly half his companies pivot significantly and Fielding estimating at least a third of Everywhere's portfolio does.
Go deeper than the headline
Research every firm in this story, track the ones that matter and get the daily briefing — free for 7 days.