Back to The Twenty Minute VC (20VC)

20VC: Inside Carnegie Mellon's $4BN Endowment | Why 90% of LPs...

The Twenty Minute VC (20VC)

Full Title

20VC: Inside Carnegie Mellon's $4BN Endowment | Why 90% of LPs Shouldn't Invest in VC | The $140BN Problem with Multi-Stage Funds | The Hidden Math Behind DPI, TVPI, and Illiquidity with Miles Dieffenbach

Summary

This episode delves into how Carnegie Mellon's $4 billion endowment approaches venture capital, offering an LP's candid perspective on market inefficiencies, risk-adjusted returns, and the challenges posed by the growth of multi-stage funds. Miles Dieffenbach critiques the current venture landscape, emphasizing the need for greater liquidity and alignment between GPs and LPs to ensure sustainable returns.

Key Points

  • Carnegie Mellon's $4 billion endowment allocates 85% to equity and 15% to fixed income, with half of the equity in private investments, making venture capital a significant portion at nearly 25% of the total endowment, which is overweight compared to peers.
  • LPs are generally not adequately compensated for the high risk associated with venture capital, as historical median IRRs (around 8% net) and DPI (1.8x top quartile for 15-year funds) often underperform public market equivalents like the NASDAQ 100.
  • Only LPs with a proven ability to access and select top-decile managers should consider investing in venture, as returns significantly lag public benchmarks even for top-quartile performers.
  • Small seed funds (e.g., $50-$100 million) face structural challenges because average seed round sizes ($45 million) necessitate either subscale ownership or insufficient portfolio diversification.
  • The "selling" pillar of venture success, referring to exits and liquidity, has been a major weakness for many managers, particularly evidenced by the lack of distributions from the 2021-2022 vintages despite high valuations.
  • Large multi-stage funds, some raising $7 billion, face significant hurdles in delivering desired net returns (e.g., 4x net) because achieving such multiples would require unrealistically massive market cap exits, far exceeding historical IPO and M&A values.
  • Current venture fee structures (e.g., 2% management fee and 20% carried interest) are deemed inappropriate for large multi-stage funds that invest in more mature companies, suggesting a need for lower fees closer to those of long-only public equity investors.
  • Carnegie Mellon conducts rigorous due diligence on GPs, including extensive off-sheet reference calls, to assess interpersonal dynamics and partnership risks, as these factors are crucial for long-term LP-GP relationships.
  • The venture capital industry is experiencing a severe liquidity problem, with fundraising at its lowest levels in years and significantly fewer IPOs and M&A exits compared to prior decades, indicating a broken mechanism for returning capital to LPs.
  • The speaker anticipates an eventual "bubble pop" in AI investments, similar to past technological transitions, due to unsustainable capital expenditure and a high dependence on continuous cash injections, making many high-valuation AI companies inherently risky.

Conclusion

Venture capitalists are urged to accelerate taking their portfolio companies public, as favorable public market conditions currently offer viable exit opportunities and are essential for alleviating the industry's liquidity crunch.

LPs should exercise extreme caution and selectivity when investing in venture, focusing exclusively on managers with demonstrable top-decile access and performance, and critically evaluate the fundamental economics and alignment of multi-stage funds.

It is suggested that large multi-stage funds revise their fee structures to better reflect the more passive nature of their investments in mature companies, shifting towards models akin to public equity fees (e.g., 1% management, 10% carry) for improved LP alignment.

Discussion Topics

  • Given the challenges Miles Dieffenbach highlighted for LPs in venture capital, what strategies could new or smaller endowments/family offices employ to achieve top-decile returns?
  • Miles Dieffenbach argues against current fee structures for large multi-stage funds. If you were a GP of such a fund, how would you justify existing fees or what alternative models would you propose to maintain LP alignment?
  • Considering the current "liquidity drought" and the potential for an AI investment bubble, what actions should venture capitalists prioritize to ensure long-term sustainability and healthy exits for their portfolio companies?

Key Terms

LP
Limited Partner; an investor in a fund (e.g., venture capital, private equity) who provides capital but has limited liability and no management responsibilities.
GP
General Partner; the managing entity of a fund responsible for investment decisions and operations, receiving management fees and a share of profits (carry).
Endowment
A financial fund established by a non-profit institution (like a university) that provides a consistent income stream for its operations through investment returns.
PME (Public Market Equivalent)
A method used to compare private market investment returns (like venture capital) to public market benchmarks, typically an index like the NASDAQ 100.
IRR (Internal Rate of Return)
A metric used to estimate the profitability of potential investments, representing the discount rate at which the net present value (NPV) of all cash flows from a project equals zero.
MOIC (Multiple on Invested Capital)
A metric that measures the total value of an investment (including distributed and unrealized value) relative to the capital initially invested, often expressed as a multiple (e.g., 2.5x).
DPI (Distributed to Paid-In Capital)
A liquidity metric for private equity or venture capital funds, representing the total cash distributions received by LPs divided by the total capital they have invested.
TVPI (Total Value to Paid-In Capital)
A performance metric for private equity or venture capital funds, representing the sum of distributed capital and the remaining net asset value (NAV) divided by the total capital paid in by LPs.
CapEx (Capital Expenditure)
Funds used by a company to acquire, upgrade, and maintain physical assets such as property, industrial buildings, or equipment.
Strip Sale
The sale of a portion of an LP's interest across multiple funds or across all assets within a specific fund, often done to generate liquidity or rebalance a portfolio.
Multistage Funds
Venture capital firms that invest across multiple stages of a company's lifecycle, from early seed rounds to late-stage growth equity.
Carry (Carried Interest)
A share of the profits of an investment fund paid to the general partner (GP) that exceeds the amount that would be allocated to them based on their capital contributions.

Timeline

00:04:13

Carnegie Mellon's $4 billion endowment allocates 85% to equity and 15% to fixed income, with half of the equity in private investments, making venture capital a significant portion at nearly 25% of the total endowment, which is overweight compared to peers.

00:05:52

LPs are generally not adequately compensated for the high risk associated with venture capital, as historical median IRRs (around 8% net) and DPI (1.8x top quartile for 15-year funds) often underperform public market equivalents like the NASDAQ 100.

00:06:40

Only LPs with a proven ability to access and select top-decile managers should consider investing in venture, as returns significantly lag public benchmarks even for top-quartile performers.

00:07:25

Small seed funds (e.g., $50-$100 million) face structural challenges because average seed round sizes ($45 million) necessitate either subscale ownership or insufficient portfolio diversification.

00:09:08

The "selling" pillar of venture success, referring to exits and liquidity, has been a major weakness for many managers, particularly evidenced by the lack of distributions from the 2021-2022 vintages despite high valuations.

00:18:18

Large multi-stage funds, some raising $7 billion, face significant hurdles in delivering desired net returns (e.g., 4x net) because achieving such multiples would require unrealistically massive market cap exits, far exceeding historical IPO and M&A values.

00:23:19

Current venture fee structures (e.g., 2% management fee and 20% carried interest) are deemed inappropriate for large multi-stage funds that invest in more mature companies, suggesting a need for lower fees closer to those of long-only public equity investors.

00:13:24

Carnegie Mellon conducts rigorous due diligence on GPs, including extensive off-sheet reference calls, to assess interpersonal dynamics and partnership risks, as these factors are crucial for long-term LP-GP relationships.

00:27:15

The venture capital industry is experiencing a severe liquidity problem, with fundraising at its lowest levels in years and significantly fewer IPOs and M&A exits compared to prior decades, indicating a broken mechanism for returning capital to LPs.

00:34:53

The speaker anticipates an eventual "bubble pop" in AI investments, similar to past technological transitions, due to unsustainable capital expenditure and a high dependence on continuous cash injections, making many high-valuation AI companies inherently risky.

Episode Details

Podcast
The Twenty Minute VC (20VC)
Episode
20VC: Inside Carnegie Mellon's $4BN Endowment | Why 90% of LPs Shouldn't Invest in VC | The $140BN Problem with Multi-Stage Funds | The Hidden Math Behind DPI, TVPI, and Illiquidity with Miles Dieffenbach
Published
August 4, 2025