How venture capital works as a business

#Same fee language, a completely different portfolio shape

Venture-capital funds use the same limited-partnership structure and the same "2 and 20" fee shorthand introduced in Module 01 — a management fee around 2% of committed capital, carried interest around 20% of profits above a hurdle. [Directional] But what a VC fund actually buys is structurally opposite to what a PE fund buys: small minority equity stakes (commonly single-digit to low-double-digit percentages) in a large number of very early-stage, usually pre-profit or pre-revenue companies, using no debt at all — a startup has no stable cash flow to service debt against, so venture investing is pure equity risk. [Established] Where a PE deal's downside is cushioned somewhat by the acquired company's existing cash flow and the fund's ability to influence operations directly, a venture investment's downside on any single company is capped at exactly the amount invested (you can lose 1x your money, never more) while its upside is, in principle, unbounded. [Established] — this asymmetry is the single mechanical fact that explains everything else in this module.

#The power-law distribution, with real numbers behind it

That capped-downside, uncapped-upside asymmetry produces a power-law return distribution rather than the roughly bell-curve distribution most other asset classes exhibit: most individual investments in a venture portfolio return less than the capital put into them, a small minority return a modest multiple, and a very small number of outlier "home run" investments generate the overwhelming majority of the fund's entire return. [Established] — this shape is documented across every independent dataset examined for this course, not asserted from theory alone:

• Correlation Ventures, a data-driven VC firm, published its own analysis of over 21,000 US venture financings from 2004–2013: 65% of investments lost money outright, only 2.5% returned 10x–20x, about 1% returned more than 20x, and roughly 0.5% returned 50x or more. [Directional] — the firm's own published data and methodology (via its "VC by the Numbers" analysis), which this course treats as a disclosed-numbers operator source rather than an independently audited academic dataset; it is nonetheless the most granular outcome-distribution data found in this research.
• Horsley Bridge, a fund-of-funds that has invested in venture managers since 1983, shared aggregated data with Andreessen Horowitz (published on the firm's own blog by general partner Chris Dixon in June 2015) covering roughly 7,000 individual investments made between 1985 and 2014: about 6% of investments, representing 4.5% of dollars invested, generated approximately 60% of the total returns across the entire dataset. [Directional] — Horsley Bridge's own aggregated data as published through a16z; the sample runs only through 2014, so it reflects several decades of venture history rather than the current market specifically.
• Ilya Strebulaev, a Stanford Graduate School of Business professor who runs Stanford's Venture Capital Initiative, has published research (with co-authors including the University of Florida's Blake Jackson) analyzing more than 100,000 venture investors, finding that fewer than 40% of active VC investors are ever credited with even a single successful investment, and that roughly 90% of aggregate VC investment profits are generated by approximately 5% of venture capitalists. [Directional] — Strebulaev is a genuine academic researcher publishing in this specific field (tier 2 on this course's own source-tiering standard), though this specific figure comes from his broader body of published research and popular-press coverage of it rather than one single peer-reviewed paper this course independently verified line by line.

All three datasets — built independently, at different times, from different underlying samples — converge on the same qualitative shape even though their precise percentages differ. That convergence, not any single number, is the actually load-bearing finding: this is not one contested statistic, it's the consistent structure of the entire asset class. [Established]

#What the power law means for how a VC fund actually behaves

A PE fund, per Module 07, is structured to want a moderate, reliable positive outcome across most of its portfolio. A VC fund's math runs the opposite way: because the fund's entire return depends on catching one or two enormous winners, a VC investor is mechanically incentivized to make outlier-seeking bets rather than downside-protecting ones — a venture fund that plays it safe across its whole portfolio, avoiding any investment with a real chance of total loss, will systematically also avoid the rare outlier bets its return model actually depends on. [Directional] This is the reasoning behind the common venture framing that a bad venture investment (a total loss) is cheap and expected, where the truly costly mistake is not investing in the company that would have been the fund-defining winner. [Directional] — a widely stated practitioner framing consistent with the power-law data above, not itself an independently measured statistic.

#Fund life, and why the timeline runs even longer than PE's

A venture fund's stated life is typically around ten years, similar to PE, but the realistic timeline to a fund's defining outcome often runs longer in practice: an early-stage startup investment frequently takes seven to ten-plus years to reach an exit (acquisition or IPO) at all, meaning many venture funds extend beyond their initial stated term to allow their best-performing positions time to mature rather than being forced to exit early. [Directional] This is a further structural reason VC compensation, covered in the next module, skews even more heavily toward a long, illiquid horizon than PE compensation already does.

The next module turns from the fund's economics to what that economics implies about who actually gets hired into VC, and why the recruiting funnel looks nothing like IB's or PE's structured pipelines.