Venture Capital Funds Explained: 2 and 20, the Power Law, LP Access, and What the 2026 Numbers Show

In the first half of 2026, US venture investors put a record $412.7 billion into startups, but only 405 venture funds closed, fewer than half the 907 that closed in all of 2025. Money is going to a shrinking number of companies from a shrinking number of funds. For anyone on either side of a venture deal, the useful question is how a fund's own arithmetic shapes its behavior.

This guide looks at venture from the fund's side: what limited partners (LPs) pay, how the return math forces funds to chase outliers, how an individual can gain access, what the latest PitchBook-NVCA data shows, and which term sheet points matter to founders. The individual investor's route, writing angel checks, is covered separately in our angel investing guide. Nothing here is investment advice, and venture capital can lose all invested money.

How a venture fund is built

A venture fund is a limited partnership. LPs (pensions, endowments, family offices, wealthy individuals) commit capital, and the general partner (GP) calls it in stages. The main terms:

• Management fee. Kruze Consulting's explainer describes a 2% annual fee typically applied over the first five years, the investment period. Smaller and larger funds vary, and many funds step the fee down after the investment period. Our illustration charges 2% for all 10 years, which slightly overstates fees for those funds.
• Carried interest. The GP keeps about 20% of profits. Some funds pay carry only after LPs get all their contributed capital back, which is how we model it below. Others pay it deal by deal, which favors the GP, so read the partnership agreement.
• Fund life. Ten years is standard, often with extensions. Companies are bought in the first three to five years, and the rest of the time is spent supporting them and waiting for exits.
• Reserves. Part of the fund is kept back for follow-on investments in the companies that are doing well. A fund that reserves 40% of its investable capital writes an initial check and then decides who deserves more.

Fees are charged on commitments, not on invested capital. At 2% for 10 years, a fund spends 20% of what LPs commit on fees. That leaves 80 cents of each dollar to invest, and the portfolio has to grow that 80 to clear the 100 the LPs put in, before any profit.

The power law, with numbers

Venture returns are dominated by a few investments. The early angel research covered in our angel guide shows the pattern at the deal level. At the fund level, the Kauffman Foundation looked at its own LP portfolio in a 2012 report and found that only 20 of 100 venture funds beat a public-market equivalent by more than 3 percentage points a year, while 62 of 100 failed to exceed public-market returns after fees and carry. That is one foundation's portfolio and it ends before the 2020s, so treat it as evidence about the shape of the outcome distribution and not a forecast.

Illustration: a $100 million fund with 30 companies

Assumptions: $100 million of commitments, $20 million of fees over 10 years, 30 initial checks of $1.5 million ($45 million), and $35 million of reserves put into the 8 best companies. Carry is 20% of profit after LPs recover all $100 million, with no preferred return. All figures are in millions of dollars and come from a script we ran and checked.

The 29 other companies return $114.75 million on $74.1 million. To give LPs a 3.0x net return, the fund must distribute $350 million: LPs receive 0.8 x $350 million + $20 million = $300 million after $50 million of carry. So the outlier must return $235.25 million on $5.875 million invested, which is 40 times its cost and 67% of all the fund's proceeds.

What exit value does that take? With an 8% stake at exit, the company needs an exit value of about $2.9 billion. At 5% it needs $4.7 billion, and at 3% it needs $7.8 billion. Ownership at exit is lower than at entry because later rounds dilute the fund.

The net result depends heavily on that one company:

With no outlier at all, the fund still returns 1.12x net, which is a positive but weak result for a decade of illiquidity. Two things follow for anyone on the other side of the table. First, a fund cannot make its returns from the company that "does fine." A $100 million fund wants a company that can return the entire fund on a single position: at 8% ownership, that is a $1.25 billion exit. A $1 billion fund needs a $12.5 billion exit for the same test. Second, that is why funds pass on good businesses whose plausible exit is small, and why founders should ask what exit a firm needs from them.

The J-curve

Cash flows are negative first. Using the same fund and an assumed schedule of calls and distributions that adds up to the $350 million gross:

The LP is $78 million out of pocket at the low point in year 5 and does not break even until year 8, and the net IRR is 22.6% only because the fund hits the 40x outlier. The timing of the payouts is our assumption. For the metrics involved (TVPI, DPI, IRR, and public-market comparisons such as KS-PME), see our private equity due diligence guide.

The 2026 market: the PitchBook-NVCA Venture Monitor

The Q2 2026 PitchBook-NVCA Venture Monitor, with data as of June 30, 2026, shows a market that has never been larger and never been narrower.

Deal value. US venture deal value reached $412.7 billion in the first half. Rounds of $100 million or more made up 87.5% of it, and AI companies took $355.9 billion, or 86%. Rounds under $100 million drew $51.4 billion, which was 12.5% of the total, compared with 43.8% in 2024 and 33.1% in 2025. Anthropic's $65 billion round was the largest, though the report notes $15 billion of it had been committed earlier.

Fundraising. Funds raised $72.4 billion across 405 funds in the first half, just under the $74.9 billion raised across 907 funds in all of 2025. Funds of $1 billion or more raised $49.5 billion across 16 vehicles. Andreessen Horowitz, Thrive Capital, and Founders Fund together took 48.1% of the total. Experienced managers raised 89% of capital ($64.5 billion against $7.9 billion for emerging managers), and first-time funds raised $3.4 billion across 53 vehicles, against $11.4 billion for the whole of 2025. The median time to close a fund fell to 6.4 months from 15.1 months in 2025, which the report reads as a sign that only well-placed funds are closing.

Exits. Exit value was $2.19 trillion in the first half, of which $1.83 trillion came in the second quarter, driven by SpaceX's IPO. That IPO raised $75 billion, and its acquisition of xAI in the first quarter was valued at $250 billion. The report says that without SpaceX, the quarter's exit value would sit at a level consistent with recent constrained years, and that the distributions "do not signal a reopened liquidity market." Unicorns reached a record 945 with aggregate value of $5.3 trillion, which widens the gap between paper value and cash returned to investors.

What it means for LPs and founders:

• For LPs, the stated distribution drought is the reason new commitments concentrate in the largest managers. The report expects cash to reach LPs only after lockups expire.
• For founders outside AI, the money is thinner. Corporate VCs took part in just 21.1% of deals in the first half, the lowest share in a decade.
• For a new fund manager, raising a first fund is hard now, though the report says many emerging managers are spinouts of established firms.

Getting access as an LP

Securities law shapes who can join. Most venture funds are unregistered and rely on one of two exemptions in section 3(c) of the Investment Company Act: no more than 100 beneficial owners (250 for a qualifying venture capital fund, a small fund capped by statute at $10 million of capital and commitments, indexed for inflation), or ownership only by qualified purchasers (a category of investors with large investment portfolios). The offering also normally requires investors to be accredited, and the SEC's accredited investor definition includes net worth over $1 million excluding a primary residence, income over $200,000 ($300,000 with a spouse) for two years, or certain licenses such as Series 7, 65, and 82.

Routes for individuals:

1. A small fund directly. Minimums are set by each partnership agreement, and access usually depends on relationships. We did not find a reliable industry-wide minimum, so ask.
2. A feeder fund or special-purpose vehicle, which pools several investors into one commitment. This lowers the minimum and adds a second layer of fees; the pre-IPO guide shows what that does to returns.
3. An evergreen or interval fund. ARK Venture Fund's SEC filings state a $500 minimum initial investment and a policy of quarterly repurchase offers for at least 5% and up to 25% of shares at NAV, subject to proration when requests exceed the offer. The fund invests partly through private vehicles run by unaffiliated managers, so their fees stack on top of the fund's own; read the prospectus expense table. Our private equity for individuals guide covers how evergreen structures behave under stress.

Compare any route against a public-market alternative. In Kauffman's sample most funds did not beat public markets after fees, and a feeder or fund of funds adds a further layer of cost.

What founders should know about term sheets

A term sheet is mostly a list of how proceeds and control are split. The main points:

Liquidation preference. Preferred investors get paid before common shareholders in a sale. Cooley's Q2 2026 venture financing report found 95.8% of reported deals had a 1x preference and 96.4% had non-participating preferred stock, so founders should treat those as the market and question anything richer. Illustration: an investor puts $10 million into a company for 25% of its equity, with a 1x non-participating preference. At an $8 million sale the investor takes all $8 million. At $30 million, the preference ($10 million) beats converting to common (25% x $30 million = $7.5 million), so the others share $20 million. The investor converts once the sale price passes $40 million, and at $60 million the others receive $45 million. With a participating preference, the same investor at $30 million would take $10 million plus 25% of the remaining $20 million, or $15 million. This is also why a 409A valuation of common stock is lower than the investor price; see our valuation guide.

Pro rata rights. The right to buy a share of future rounds to keep an ownership percentage. It matters most to the fund because it decides how much of its reserves it can put into winners. Founders should ask whether the firm will use it and whether the right applies to every investor or only those above a threshold, since many small pro rata rights make future rounds harder to allocate.

Down-round and structure terms. Cooley's Q2 2026 data: 83.6% of deals were up rounds, 4.3% flat, and 12.1% down (10.9% in Q1); pay-to-play provisions appeared in 8.4% of deals, redemption rights in 5.4%, and accruing dividends in 3%. Anti-dilution protection decides how much a down round costs founders.

Fund-level questions to ask the investor:

1. How big is the fund and what exit does your company have to reach for it to matter? (See the return-the-fund test above.)
2. How far along is the fund, and how much is reserved for follow-ons? A fund in year eight has different incentives from one in year two.
3. Will you lead or follow in the next round, and will the firm use its pro rata?

Convertible notes and SAFEs come before a priced round, and the convertible debt guide covers their caps and stacking. Between equity rounds, venture debt can stretch runway but adds a lender with a claim ahead of everyone. Later-stage rounds often come from growth equity investors, and venture is one of the options within alternative investments.

Before you commit as an LP or sign as a founder

1. Read the fee and carry terms, and check whether carry is calculated on the whole fund or deal by deal.
2. Ask a fund manager what exit is needed from a typical portfolio company, and how the fund's reserves are set.
3. Ask for realized distributions (DPI), not only paper value; the Monitor shows the gap between the two is wide right now.
4. Keep venture to money you can leave untouched for a decade or more.
5. As a founder, have a lawyer experienced in venture financings read the term sheet before you sign.

This guide is for informational purposes only and is not investment, tax, or legal advice. Venture capital is illiquid and high risk. Consult qualified professionals before investing or signing financing documents.