Kurz gesagt: As megafunds absorb nearly three-quarters of new venture capital, the single-digit-billion exit becomes an orphaned asset class, and the investors who capture it will be those who build diversification across many vehicles at once.
Five firms took 73 per cent of new venture money, and the middle of the market emptied out
The first quarter (Q1) of 2026 marked a new extreme in venture fundraising concentration. According to the PitchBook and National Venture Capital Association (NVCA) Venture Monitor, 73.1 per cent of US venture capital committed in Q1 went to five firms[1], out of $47.8 billion raised across 172 funds. Experienced managers captured 90.9 per cent of the quarter’s capital, the highest share in the dataset’s history. The same report shows the median fund size falling to $15.3 million from $25 million in 2025, and first-time funds raising $2.1 billion across 29 vehicles, after $7.3 billion across 106 in the whole of the previous year.
Those three numbers describe one movement. Capital is pooling at the very top, while the long tail of managers shrinks in both count and ticket size. In Allegory Capital’s reading, the result is a barbell: a handful of multi-strategy platforms with tens of billions under management at one end, and a crowd of micro-funds at the other. The layer that used to sit between them, the disciplined $300 million to $1 billion early-stage fund, is the one that loses ground.
The last holdouts of small-by-choice venture have scaled up
The clearest signal came this year, from two of the best-known firms that had stayed deliberately small for decades. One announced $900 million in new funds on 8 October 2026[2], against a previous core fund of $275 million. The other raised $2 billion in June across a $750 million early-stage fund and its first-ever growth fund of $1.25 billion[3], ending a two-decade habit of keeping funds near $425 million. A widely read October analysis, The Death of Venture Capital’s Short Kings[4], frames the logic plainly: the rounds that decide a company’s fate now happen later and at larger sizes, so a small fund protects its founders by growing big enough to sit at that table.
The behaviour is rational for each firm. The aggregate effect is the part limited partners (LPs) should study. The same analysis concludes that bigger funds raise the ceiling on outcomes while their number stays fixed; the power law gets taller and keeps the same narrow base. A $10 billion fund needs exits in the tens of billions to return its capital. A company heading for a $2 to $6 billion outcome is a rounding error on that book.
The single-digit-billion exit is losing its natural owners
One founder of those small-by-choice firms still expects healthy returns from winners in the single-digit billions[4]. That is the precise outcome band that the megafund model structurally deprioritises. Fewer investors are now built to want it, which leaves more of it available on better terms to the investors who are.
For a family office, a corporate venture arm or a regional institutional LP, this reframes the question. The opportunity lies in the outcome band that the largest pools of capital have priced out of their mandate. The difficulty lies in owning enough of it, at a scale small vehicles can sustain, while keeping in check the concentration risk that destroys small funds.
Portfolio maths punishes the small vehicle that concentrates
Venture returns follow a power law, and the data on it is consistent across two decades. An outcomes study built on more than 21,000 US venture financings found that about 65 per cent failed to return capital and roughly 4 per cent returned 10x or more[5]. A 2019 analysis of US venture-backed exits by the same research team reported that about half of all capital invested lost money[6], with fewer than 4 per cent of dollars earning 10x or better.
When one deal in twenty-five carries the fund, the number of positions decides the shape of the result more than any other choice a manager makes. To show the effect at the scale a family office or corporate vehicle actually operates, Allegory Capital ran a Monte Carlo simulation of 200,000 portfolios at each size, drawing deals from a distribution calibrated to the 21,000-financing study, which sits close to the later capital-weighted figures.
Forty positions cut the chance of loss from 23 to 3.5 per cent
The model assigns 65 per cent of deals to a loss (50 per cent written off, 15 per cent returning between 0.2x and 1x), 25 per cent to 1x to 5x, 6 per cent to 5x to 10x, and 4 per cent to a heavy tail starting at 10x. Every deal carries the same cheque and the model holds capital at the first cheque. The average deal returns about 2.2x gross, so every portfolio size has the same expected value. What changes is the distribution around it.
| Positions | Median gross | Chance of loss | Owns a 10x | Top 5% |
|---|---|---|---|---|
| 5 | 1.49x | 34.8% | 18.4% | 5.9x |
| 10 | 1.70x | 22.6% | 33.3% | 5.1x |
| 20 | 1.85x | 11.4% | 55.9% | 4.4x |
| 40 | 1.94x | 3.5% | 80.5% | 3.9x |
| 60 | 1.98x | 1.2% | 91.4% | 3.6x |
| 100 | 2.03x | 0.1% | 98.3% | 3.4x |
Source: Allegory Capital Monte Carlo simulation, 200,000 portfolios per size, October 2026. Deal distribution calibrated to published outcome data on 21,000+ US venture financings. Gross multiples, equal cheques, no follow-on, before fees and carry.
The median vehicle with ten positions returns 1.70x gross and loses money in 22.6 per cent of runs. At forty positions the median rises to 1.94x and the probability of loss falls to 3.5 per cent, because the vehicle now owns at least one 10x outcome four times in five. Beyond sixty positions the gains flatten: the median moves by only five hundredths between 60 and 100 companies, which places the practical range at 40 to 60.
Diversification buys safety and caps the jackpot
The top 5 per cent of ten-company vehicles return 5.1x or more; at forty companies that ceiling compresses to 3.9x. A concentrated portfolio is a lottery ticket with a one-in-five chance of losing the stake. A diversified portfolio is an asset allocation decision with a narrow, reliable band of outcomes around 2x gross.
For most family offices and corporates, the second profile is the one an investment committee can defend for a decade. Arithmetic is the obstacle: a 5 million Swiss franc (CHF) vehicle that wants forty positions writes cheques of CHF 125,000 before fees and reserves, too small to secure information rights, board observation or pro-rata rights (the right to invest in later rounds) in a competitive round. A diversified programme needs a structure built for many small positions, and a single small fund struggles to provide one.
Regulated industries are where the orphaned exit lives, and vehicle design decides who captures it
Allegory Capital’s thesis is that the outcome band megafunds now deprioritise is disproportionately populated by companies in regulated sectors: fintech, insurtech, health and medtech, energy and climate infrastructure. These businesses grow more slowly in their first years, because licences, certifications and procurement cycles take time. They also build moats that compound once in place, and the natural buyer is often an incumbent that needs the licence and the customer base, which in Allegory Capital’s reading places many of them in the low single-digit-billion exit band.
Compliance compresses the power law in the investor’s favour
In a regulated market, the licence, the audit trail and the integration with incumbent systems are themselves assets. Allegory Capital has argued that procurement fit is an investable asset in regulated Europe[7], and that regulated-industry startups capture an outsized share of European Series B capital[8] for exactly that reason. Barriers to entry reduce the number of credible competitors per category, which narrows the left tail of outcomes for the survivors. The trade is a lower probability of a 100x and a higher probability of a 5x to 20x, which in Allegory Capital’s reading is the band where sector knowledge gives a specialist investor its edge over a generalist.
This is also where the barbell creates the widest pricing gap. A megafund screens a regulated-industry Series A against a target exit it needs in the tens of billions, and frequently passes. A specialist vehicle screens the same company against a $2 to $4 billion strategic sale and sees a clean 10x. The same company therefore carries two prices, set by the size of the fund looking at it.
Small investors reach forty positions by stacking vehicles
The cheque-size problem has a structural answer. An LP can hold a set of single-purpose vehicles, each concentrated on one deal or one tight theme, and reach diversification at the level of the whole programme. Special purpose vehicles (SPVs) make each position legally ring-fenced, let the investor choose which deals to join, and let a family office or corporate run the programme under its own brand while a specialist operator handles deal flow, diligence, compliance and capital calls.
Allegory Capital’s own SPV platform is built on that model. According to its 2025 SPV prospectus[9], an investor can run a venture programme under its own brand and Swiss company from CHF 5 million of assets, using Swiss-based SPVs and a network of vehicles in the United States, the United Kingdom and Israel to accept cross-border capital, while investment decisions stay with the investor’s own governance.
The stack costs more to set up and buys control in return
A stack of SPVs carries a different cost profile from a blind-pool fund, and an investment committee should price it explicitly. The same prospectus[9] sets a one-off set-up fee of CHF 150,000, an annual fee of 2 per cent on assets and a horizon of three to nine years. On a CHF 5 million programme, the set-up fee alone absorbs 3 per cent of committed capital before the first cheque, so the structure earns its keep only when the investor commits enough capital and enough positions to amortise it. A family office writing four or five deals a year reaches forty positions in eight to ten years, which stretches past the nine-year ceiling, so a programme aiming for forty positions needs a faster pace of six to eight deals a year or a horizon agreed up front.
Fees and reserves turn the gross figures in Exhibit 1 into a lower net number, and an investment committee should see both. Applying the prospectus terms to a CHF 5 million programme over a nine-year horizon, and charging the 2 per cent annual fee on commitments for simplicity, the set-up fee and nine years of fees absorb CHF 1.05 million, leaving CHF 3.95 million, or 79 per cent of commitments, at work. The 40-position median of 1.94x gross then returns about 1.53x on committed capital, against about 1.34x for the 10-position median, before any performance fee. On the same net basis, the chance of returning less than committed capital is about 12 per cent at forty positions, against about 33 per cent at ten. Follow-on reserves move the result again: holding back part of each cheque to defend ownership in later rounds concentrates capital in the winners the investor can already see, which lifts the upper tail and leaves fewer first cheques for the same budget. The gap between ten and forty positions holds on a net basis, and the gross-to-net bridge belongs in the programme policy from the first deal.
What the investor buys with that cost is control over the three variables a blind pool decides on its behalf: which deals enter the portfolio, how large each cheque is, and when to stop. That control carries a discipline of its own. The programme earns its diversification only when the investor fixes a target number of positions in advance and keeps writing cheques through a weak vintage, because the Exhibit 1 curve rewards breadth and punishes the investor who stops at twelve deals after two early losses. A policy written before the first deal, with a target count, a cheque size and a sector mandate, makes the forty-position target binding on the people who manage the programme. It also gives the investor a clear reporting line to its own board: positions written against target, capital deployed against plan, and the share of the programme in regulated sectors.
The orphaned exit rewards the investors who plan for it now
The concentration of venture capital will likely continue while megafunds keep raising faster than smaller managers. That creates a durable gap in the single-digit-billion outcome band, and the gap is widest in regulated industries, where sector knowledge and patience price risk better than scale.
Each type of investor has a different first move. A family office should set a target number of positions before choosing a vehicle, and treat forty as the floor for a programme expected to return capital with high confidence. A corporate venture arm should treat its own industry knowledge as an underwriting edge in exactly the companies megafunds pass on, and use deal-by-deal structures to keep that edge with a lean in-house team. An emerging manager should pick portfolio size from the outcome it promises its LPs: concentration for the shot at a top-decile fund, breadth for the near-certainty of returning capital. A founder in a regulated sector should ask every lead investor what exit returns its fund, because a $3 billion strategic sale is a triumph for one cap table and a footnote for another.
This article has been reviewed by Haider Alleg, General Partner at Allegorisches Kapital.
Referenzen
- PitchBook and NVCA. Q1 2026 PitchBook-NVCA Venture Monitor, April 2026. https://nvca.org/wp-content/uploads/2026/04/Q1-2026-PitchBook-NVCA-Venture-Monitor.pdf
- USV. The Next Chapter of USV, 8 October 2026. https://www.usv.com/writing/the-next-chapter-of-usv
- TechCrunch. Benchmark raises its first-ever growth fund as part of $2B capital raise, 3 June 2026. https://techcrunch.com/2026/06/03/benchmark-raises-its-first-ever-growth-fund-as-part-of-2b-capital-raise/
- Kyle Harrison. The Death of Venture Capital’s Short Kings, Investing 101, October 2026. https://investing101.substack.com/p/the-death-of-venture-capitals-short
- Seth Levine. Venture Outcomes Are Even More Skewed Than You Think, August 2014. https://sethlevine.com/archives/2014/08/venture-outcomes-are-even-more-skewed-than-you-think.html
- Correlation Ventures. Venture Capital: No, We’re Not Normal, Medium, September 2019. https://medium.com/correlation-ventures/venture-capital-no-were-not-normal-32a26edea7c7
- Allegory Capital. Users Buy the Product. Organisations Buy the Company., October 2026. https://allegory.capital/procurement-fit-vs-product-market-fit/
- Allegory Capital. Why Regulated-Industry Startups Are Capturing Disproportionate Series B Capital in Europe, July 2026. https://allegory.capital/european-regulated-startups-series-b-thesis/
- Allegory Capital. SPV Prospectus, 2025 (first-party document, available on request from Allegory Capital).
This article is for information only and does not constitute an offer, solicitation or investment advice. Venture investments carry a high degree of risk, including the loss of the entire amount invested.
