by Jim Pulcrano, IMD, & Reference Capital
For decades, venture capital operated on a predictable timeline: invest early, build companies over a ten to twelve-year fund lifecycle, and return capital to limited partners through IPO or acquisition. That model has come under sustained pressure.
In the United States, over 40% of active unicorns raised their first venture round more than a decade ago¹, and cumulative cash flows from US venture funds to limited partners have been negative by an estimated $197 billion since 2022². The secondary market has grown directly in response.
It’s important to point out for context, while secondary transactions create liquidity for sellers they do not always create liquidity for the market. The underlying assets remain private and illiquid; what changes is who bears the exposure. This article offers a practitioner’s assessment of both the utility and the potential risks associated with a rise in secondary transactions.
What Are VC Secondaries?
A venture capital secondary is the sale of an existing stake in a private company or fund to a new investor, rather than a primary financing in which the company issues new shares. No new capital reaches the underlying business. The asset does not become more liquid as a result; it remains private and subject to the same exit timeline as before.
The market operates across three structures, each with a distinct risk profile.
LP-led secondaries occur when a limited partner sells its stake in a VC fund to another investor. The buyer acquires exposure across the fund’s portfolio rather than a single company, and LP portfolio pricing averaged 90% of NAV globally in H1 2025 (though US venture and growth-specific pricing was closer to 78%³). The practical risk for buyers is that the return case rests almost entirely on one or two of the strongest companies in the fund. This follows from the power-law structure of venture returns: in a typical VC fund, the overwhelming majority of value is generated by a small handful of outlier investments, often just one or two, while the rest of the portfolio contributes little or nothing to net returns. When a secondary buyer acquires a fund stake, they are buying into this distribution as it already stands, meaning their outcome depends almost entirely on whether those top performers ultimately exit at valuations that justify the secondary purchase price.
GP-led secondaries, increasingly structured as continuation funds, allow a fund manager to transfer selected assets into a new vehicle. Existing LPs choose to take liquidity or roll their stake forward. US GP-led VC secondaries reached $14.6 billion in 2025⁴. The structural problem is that the GP controls every dimension of the transaction, price, buyer selection, and information flow, while simultaneously crystallizing carried interest and earning new management fees. Access to the most attractive processes is also frequently tied, explicitly or implicitly, to commitments to the GP’s next primary fund.
Direct secondaries involve purchasing a stake in an individual company from a founder, employee, or early investor. US direct secondary volume reached a midpoint estimate of $91.7 billion in 2025, though PitchBook’s own range spans $62.5 billion to $120.9 billion, reflecting both market opacity and the outsized weight of a small number of transactions. Access is typically gained through SPVs (Special Purpose Vehicles), which in 2025 had a median raise of $930,000 from nine LPs over just 16 days⁴. This is the closest position to the asset, but often with the least time and the least transparency to validate what is being purchased.
Why Secondaries Exist
Secondaries have existed as long as VC. The secondary market did not grow because investors discovered a new asset class. It grew because the traditional venture liquidity cycle broke down simultaneously at three levels: the companies that should, or could, have exited did not, and the funds that should have been distributing capital were not. The LP consequences followed directly from these failures. LPs that should have recycled proceeds had nothing to recycle, and their decision to sell was driven by portfolio constraints.
At the company level, the traditional exit pathway narrowed for three compounding reasons.
1. More companies are remaining private longer. The post-2021 rate cycle compressed public market multiples and collapsed near-term IPO activity. US VC-backed listings fell from 198 in 2021 to 42 in 2022, recovering only modestly to 48 in 2025.² But the rate cycle is only part of the story.
2. Public markets have changed structurally in ways that raise the listing threshold independent of rate conditions: passive strategies now dominate equity flows, analyst coverage of new listings has dropped, and meaningful index inclusion requires a scale that almost no VC-backed company reaches at IPO. Exits credible at $100 million in the early 2000s required $1 billion a decade later, and require $10 billion or more today (cf. Reference Capital, The Age of Day-Zero Unicorns). These structural changes are unlikely to compress meaningfully even as the rate cycle normalizes.
3. A third factor is behavioral: companies that raised at 2021 peak multiples face the prospect of a public listing that formally confirms losses their private investors have not yet recognized. The median 2025 unicorn IPO priced at 0.97 times its last private round, with 14 of 17 unicorn listings below their private peak.⁵ For many, delaying is not a liquidity calculation, it is a reputational one.
At the fund level, larger vehicles have extended the duration of the problem. US VC AUM have grown from under $400 billion in 2015 to over $1 trillion in 2025². The exits required to return this capital have grown proportionally, and nearly half of US unicorns completed their first venture round in 2016 or earlier⁴. GPs are holding assets longer not only because they choose to but because the exit markets described above give them little alternative.
At the LP level, the consequences are now acute. Cash flows to US VC limited partners have been negative by $197 billion since 2022, with distribution yield falling to a trough of 7.5% in 2023 against a historical average of 15%². LPs unable to fund new commitments from existing distributions must either reduce venture allocations – only 537 US venture funds closed in 2025, the fewest in a decade – or generate liquidity through secondary sales. The decision to sell is driven by portfolio constraints, not necessarily asset-level conviction.
A Market Built on Power Law
The secondary market’s most underappreciated feature is its extreme concentration. Secondary activity does not distribute across the venture ecosystem; it clusters around perceived winners, replicating venture’s power-law return structure rather than diversifying it. In Q4 2025, the top 20 companies on the Hiive platform accounted for 86.4% of global secondary trading value, with the top five alone at 55.6%.⁴ OpenAI’s single October tender offer ($6.6bn) represented 6.2% of full-year US secondary volume. SpaceX was 12.5% of all Augment platform activity in Q4.
A common misconception is that LPs use secondaries to exit underperforming managers. In practice, the opposite is true. The assets generating secondary trading volume are often the strongest in any given portfolio. Weaker managers and struggling portfolios have no secondary market. There is insufficient secondary capital relative to primary capital to absorb the long tail, and buyers compete for access to the same handful of well-known names. As a result, the market clears only for assets that are already perceived as winners, leaving the majority of the venture ecosystem effectively illiquid.
This creates a wave dynamic with a specific forward risk. Today’s secondary market is effectively a market for a small number of elite companies that happen to still be private. When SpaceX, OpenAI, and Anthropic list publicly, as expected in 2026, they will take the majority of secondary volume with them. What remains is the second tier: less proven companies, less competitive buyer interest, and less price discovery. Only 70 new companies saw their first secondary trade in 2025, totaling $492 million⁴. That is not a replacement pipeline; it is a fraction of the activity being lost.
For allocators considering commitments to dedicated secondary funds, this sequencing matters. The track record being presented today was built on pre-IPO exposure to a small number of exceptional names that reached their best spread in summer 2023 due to the combination of events mentioned before. As those names list, that specific source of return disappears. What secondary funds deploy into next is a less proven, more heterogeneous pool of assets, and that shift in underlying quality is not visible in the historical numbers.
The Buyer Risks
The risks in the secondary market are not isolated; they emerge from the interaction of pricing, governance, and information asymmetry.
Because direct access to the most sought-after companies is tightly constrained, limited by transfer restrictions, company-controlled buyer approval processes, and concentrated existing ownership, most secondary buyers gain exposure indirectly, through fund stakes or SPV structures that sit between the buyer and the underlying asset. These structures introduce complexity and opacity that direct ownership does not carry.
In GP-led continuation funds, the structural conflicts are inherent and multiple. The GP sets the transfer price, selects the lead buyer, controls the information provided to existing LPs, and determines which assets are transferred, while simultaneously crystallising carried interest and earning new management fees on the new vehicle. No independent party is required to validate the price or the process. Smaller LPs face a binary roll-or-sell decision under compressed timelines, typically with limited visibility into the buyer syndicate, the transfer valuation methodology, or the revised economics. The SEC’s 2023 rules that would have required independent fairness opinions for GP-led processes were vacated in full by the Fifth Circuit in June 2024; ILPA’s May 2023 guidance is non-binding. There is no mandatory fairness opinion requirement in the United States as of March 2026. Access to attractive GP-led processes is also frequently tied, explicitly or implicitly, to commitments to the GP’s next primary fund, a bundling that raises genuine questions about whether the secondary investment decision is being made on pure investment merit or as part of a relationship maintenance calculation.
In SPV structures, especially those layered across multiple vehicles, buyers may face significant uncertainty about the chain of ownership, the validity of transfer mechanics, applicable information rights, and fee structures embedded at each layer. When the SPV is controlled or endorsed by the issuing company, the company retains effective gatekeeping power over who holds indirect exposure to its equity. As a result, LPs are frequently asked to make decisions on allocation with little information in days or even hours if they want access to these names, often times being charged exorbitant fees. By 2025, these risks had moved from theoretical to documented: Linqto, a retail-facing secondary platform, filed for bankruptcy; a Sestante Capital manager was indicted for investor fraud in pre-IPO transactions; and FINRA’s 2026 Annual Regulatory Oversight Report explicitly flagged misrepresentation and disclosure failures in pre-IPO investments6.
The apparent pricing recovery and discount compression reflect a change in what is being sold, not a broad improvement in asset quality. Companies whose last primary round dates to 2023 were already priced under post-correction assumptions, and they trade at modest secondary discounts of around 19%. Companies carrying 2021 marks still trade at an average 68% discount⁵. These are not two points on the same recovery curve. Presenting these two groups as a single trend obscures the fact that pandemic-era assets remain deeply discounted and that the correction in that segment is far from over. In reality, the secondary market concentrates new price discovery in the subset of assets that are actually transacting, leaving the rest marked on stale assumptions.
To add more context, over a quarter of US unicorns are estimated by PitchBook to have already fallen below the $1 billion mark on a mark-to-market basis, even as they retain the unicorn designation on the strength of their last primary round. While not perfect, given the lack of demand and volume for the bottom 90% of the market, secondaries can still act as a rough guide to the valuation of some assets and even spotlight potential structural issues when there is no volume traded at all.
Conclusion
The venture secondary market provides real but narrow utility. It allows LPs to manage liquidity constraints, gives GPs a mechanism to generate DPI in the absence of exits, offers employees partial liquidity in companies that have remained private far longer than anticipated, and can shorten the J-curve and overall holding period for secondary buyers entering mid-fund-lifecycle. GP-led continuation vehicles, in particular, have delivered competitive outcomes in specific contexts, but those results are highly sensitive to manager quality, asset selection, and period, and should not be generalized across the market.
What the data shows more clearly is that the secondary market is not a broad liquidity solution. It is a selective market that clears only for a small set of perceived winners. Activity is concentrated in a handful of companies, and outside that group, assets are not simply discounted; they are often untradeable. The issue is not pricing inefficiency, it is the absence of liquidity altogether.
This distinction matters. A discount implies future upside; illiquidity implies the absence of an exit path. In a power-law system, buying outside the narrow set of assets that dominate secondary demand is not a contrarian strategy – it is a different risk category entirely.
The structural risks follow from this dynamic. Governance in GP-led processes remains uneven, information asymmetry is persistent, and access to the most attractive opportunities is frequently conditional on broader GP relationships. In more intermediated structures, particularly layered SPVs, complexity and time constraints further reduce the ability of buyers to validate what they are acquiring. These are not isolated frictions; they are embedded features of how the market operates.
As a result, sophisticated capital does not allocate to secondaries because the market as a whole is attractive. It allocates because, in specific cases, secondaries provide selective access to a narrow set of elite, oversubscribed private companies that are otherwise inaccessible through primary markets. That access is scarce, relationship-driven, and not scalable across the full opportunity set.
The key question for any buyer is therefore not whether secondaries are growing, but whether the specific asset they are acquiring sits within the narrow segment of the market that will ultimately exit, and on what timeline. Historical performance, largely driven by pre-IPO exposure to a small number of exceptional companies, may not be indicative of what capital deployed today will experience.
The secondary market does not resolve the underlying tensions in venture capital. It is a consequence of them. It concentrates liquidity around a small number of assets, leaves the majority of the market untouched, and, where misapplied, transfers risk from informed sellers to less-informed buyers. In that sense, it does not smooth the venture cycle; it makes its structural asymmetries more visible.
This prompts a set of important considerations regarding the optimal level of secondary capital in venture capital, and whether it can mature into an asset class capable of delivering consistent performance comparable to private equity.
Prof. Pulcrano directs IMD’s Venture Capital Asset Management program, has cofounded 5 startups and is currently cofounder/COO of LeaderInsight.ai
Reference Capital is a specialized Venture Capital investment firm based in Geneva, Switzerland.
References
1Unicorn age (US): PitchBook Q4 2025 Venture Monitor: “roughly 40% of private unicorns are more than 10 years old.” PitchBook 2025 Annual US VC Secondary Market Watch: 48.5% of US unicorns had their first VC round in 2016 or earlier. As of December 31, 2025.
2Cash flow, distribution yield, IPO counts, fundraising, AUM (US VC): PitchBook Q4 2025 Venture Monitor / NVCA, February 2026. Negative cash flows: $196.9bn since 2022. Distribution yield: 7.5% trough (2023), 11.2% (2025), ~15% historical average. IPO counts: 198 (2021), 42 (2022), 48 (2025). Funds closed 2025: 537. US VC AUM: over $1tn.
3LP portfolio pricing (global, all strategies): Jefferies Global Secondary Market Review, July 2025 (H1 2025 data). LP-led: 90% of NAV (all strategies); US venture/growth-specific: 78% of NAV. Geography: global, all private market strategies.
4US VC secondary market sizing, concentration, and SPV data: PitchBook 2025 Annual US VC Secondary Market Watch, February 20, 2026. Direct secondaries midpoint: $91.7bn (range $62.5bn–$120.9bn). GP-led: $14.6bn. Concentration: Hiive50 Index (global), Augment (global). Dry powder: $11.8bn as of June 2025. SPV data: Sydecar platform (median fundraising: 16 days, $930K raised, 9 LPs). US VC. As of 12/31/2025.
5Valuation discounts by cohort, IPO step-down data (US): PitchBook 2025 Annual US VC Valuations and Returns Report, February 10, 2026. 2021 cohort: −68.2%; 2022: −52.1%; 2023: −19.4%. Median IPO step-up 2025: 0.97x. 14 of 17 unicorn IPOs below private peak.
6Regulatory and enforcement: FINRA 2026 Annual Regulatory Oversight Report, December 9, 2025. Linqto bankruptcy: 2025. Sestante Capital indictment: 2025. SEC fairness opinion rules vacated: Fifth Circuit, National Assoc. of Private Fund Managers v. SEC, No. 23-60471, June 5, 2024. ILPA Continuation Funds guidance, May 2023 (non-binding).
7GP-led continuation fund performance: Ares Wealth Management Solutions, “Assessing the Quality of GP-Led Secondary Transactions,” August 1, 2025, citing HEC Paris research.
