The debate over the inclusion of private investments in 401(k) plans is a hot topic in the investment community. With more than $8 trillion in assets and a growing asset base the US defined contribution (DC) market is a significant, largely untapped market for privates.
The research paper “Why Defined Contribution Plans Need Private Investments[i]” — a 2019 collaboration between the Defined Contribution Alternatives Association (DCALTA) and the Institute for Private Capital (IPC) — provides an analysis of the potential benefits of including private equity and venture capital in DC plans, with the clear conclusion reflected in the paper’s title.
A balanced view should consider the objectives of the study’s sponsors. Specifically: DCALTA’s mission statement calls for “advocacy on the benefits of including hedge funds, private equity, and other alternative investments within a defined contribution framework.”
Consistent with the organization’s mission, the 2019 study’s bold conclusions include:
Investing in private funds “always increases average portfolio returns” when publicly traded stocks are replaced with private equity (referred to as “buyout” in the study) and venture capital investments.
The study states that “…despite the wide dispersion of returns in private funds, the ability to diversify by investing in multiple funds is sufficient to have nearly guaranteed superior returns historically.”
The message: If you play the game right, private investments always win.
A careful reading of the research should ring alarm bells for the prudent investor or fiduciary:
1. It implies that any outperformance of private investments vs. public markets justifies investment.
2. The study uses mean returns, which improves scenario results, when median results are more appropriate.
3. It assumes that the tiny VC market in the 1990s could have accommodated impossibly large investments in the simulation’s early years.
4. Assumes that the overall size of the venture capital market was equal to the buyout market, when in fact it is much smaller.
5. The cost assumptions for indexing traditional stocks and bonds are relatively high. There are lower-cost options available in the market.
6. The paper’s findings are based on hypothetical returns, while a recent real-world study indicated that the median fund of funds’ return has trailed the S&P 500.
The Devil’s in the Details
The paper compares the historical returns (from 1987 to 2017) of a traditional 60/40 stock/bond portfolio to simulated portfolios in which a chunk of the publicly traded stock allocation is replaced with randomly selected venture capital and/or buyout funds.
To compare results with public markets, the paper uses public market equivalents (PME) — a methodology for assessing the performance of private equity relative to a public equity benchmark — as a key measure. For example, the median PME of 1.06 for private equity means the typical buyout fund return was 6% better (over its entire life, not annualized) than returns from a similar investment pattern in the S&P 500.
Is that good? I think the late David Swensen, esteemed head of the Yale endowment, would have said no. He wrote: “The high leverage inherent in buyout transaction and the corporate immaturity intrinsic to venture investments cause investors to experience greater fundamental risk and expect materially higher investment returns.”[ii]
The authors’ conclusions seem to suggest that even a 1.01x PME is worth the trouble. The prudent investor would disagree.
Source: “Why Defined Contribution Plans Need Private Investments.”
In Fact, You Aren’t Invited to the Party
Despite median VC performance that trailed public markets[iii], mean returns were juiced by a small number of killer VC funds that Acme 401(k) Plan can’t (and couldn’t) access. For simulation purposes, everyone was invited. In practice, there was a velvet rope — even for large, institutional investors. This is no secret. The research acknowledges it:
“Top VC funds are also difficult for most investors to access because of excess demand for these funds and the tendency for VC general partners to limit the size of their funds.”
Temporal Anomalies and Retroactive Re-Weightings
In 1987, the DC market in the US was worth $525 billion.[iv] A 10% target allocation in venture capital, which the simulation assumes, would therefore require a $52.5 billion investment. Unhelpfully, total venture capital raised for the five years from 1987 to 1991 was $31 billion.[v] Marty McFly’s 401(k) plan could have reaped the spoils of the halcyon years. Not all of us have a time-traveling DeLorean.
The simulation also relies on equal allocationsbeing made to both VC and buyout funds, despite the capitalization of the (higher returning) VC funds being much smaller than the buyout market. The simulation massively over-weights the smaller, better performing (based on the mean result) VC funds. Is this what they mean when they say VC investment leads to great innovation?
Finally, the 60/40 Vanguard index funds used for most of the period of the paper, (VTSMX and VBMFX) have annual expense ratios of 14 and 15 basis points, respectively, when much lower-cost options have been available from Vanguard and others for years.
It’s Cheap if You Ignore the Costs
The study’s key scenario calls for plans to invest in 10 funds per year. Most institutional investors in private markets invest in less than three per year. To get to the desired 10+ funds, the plans would likely need to invest in funds of funds. In the unsimulated world, that costs more money. The paper’s assumed added costs of up to 0.5% per annum for privates compares with real world fund of funds costs of ~2%.[vi] In addition, the paper’s claim that returns were virtually guaranteed to perform better than a 60/40 portfolio appears to not reflect any additional costs associated with private investments
A more constructive approach would be to analyze the actual performance of funds-of-funds. Helpfully, academics already have. One study[vii] shows that more than half of the funds of funds underperformed the S&P based on PME. The paper’s authors note: “Our results also have policy implications regarding whether and how 401(k) plans should invest in PE funds.”
Investors and fiduciaries embarking on an alternative/private markets journey take note: Your alternative journey will be in real life, not simulated. Always consider the real-world evidence and consider the motivations of those that are selling to you.
[i] “Why Defined Contribution Plans Need Private Investments,” DCALTA/IPC Research Paper
[ii] Pioneering Portfolio Management, an Unconventional Approach to Institutional Investment. 2009. Swensen, David. page 221
[iii] 25% percentile results: Buyout: 0.87x Venture Capital 0.62x. A lot of funds have underperformed public markets
[iv] US DOL Website page 13
[v] https://www.nytimes.com/1989/10/08/business/venture-capital-loses-its-vigor.html
[vi] “Diversifying Private Equity” by Gredil, Liu, and Sensoy
[vii] “Diversifying Private Equity” by Gredil, Liu, and Sensoy Page 32
