15
Private Assets and Something About Ostriches
Any investment decision, and any decision made within an enterprise, must account for risk. Discounted cash flow valuations are completely invalid, have no significance whatsoever, and are useless for decision-making if they don’t properly account for risk. Market values derive from an estimation of risk.
—JPP
NO MATTER WHICH OPTIMIZATION METHODOLOGY WE use—multiperiod, full scale, mean variance, tail-risk-aware versions of mean variance, risk parity, and/or judgment and experience—a perennial challenge in portfolio construction has been to determine the optimal allocation to alternative asset classes. Institutional investors such as public pensions and endowments have grappled with this question for years. Recently, alternative asset classes have grown in popularity with individual investors as well.
The comparison between public and private equities is particularly important given the size and growing popularity of private equity as an asset class. Also, this comparison can be generalized to different alternative asset classes, because it relates to the broader question of how to allocate between liquid and illiquid markets.
What’s the role of private equity in investors’ portfolios? What’s the optimal mix between private and public equities? As I mentioned in Chapter 9, a naïve answer—often reinforced by consultants armed with mean-variance optimization results—is that private equity is awesome. It’s essentially a free lunch. If the investor’s time horizon is long enough, perhaps the entire portfolio should be invested in private markets because they seem to deliver high returns, low volatility, and low correlation to the business cycle. Based on my extensive review of the literature, and my own research, I would argue that private equity is not a free lunch. It has a role to play in some investors’ portfolios, but it’s a smaller role than suggested by a naïve interpretation of the data.
A study by Hamilton Lane, an asset management company in private markets, claims that as of September 30, 2018, the 20-year return-to-risk (Sharpe) ratio of private equity was 270% higher than that for public markets (as proxied by the MSCI World Index). Moreover, from 2000 to 2018, the same study claims that the lowest five-year annualized return for developed markets buyouts (a large sub–asset class within private equity focused on leveraged buyouts) was +2.4%, compared with –5.7% for the MSCI World Index. Essentially, from this five-year lens, cash has more downside than buyouts, as global cash rates have dipped to zero and below.1
Similarly, in an article by Austin, Thurston, and Prout (2019) that encourages wealthy individual investors to invest in private equity, the consulting firm Cambridge Associates finds that the more an institution invests in private equity, the higher the returns. The article implies that a 40% allocation to private equity might be optimal:
[Our analysis] highlights the meaningful returns that institutions with higher allocations to private investments have achieved over the last 20 years. The median annualized return for a greater than 15% average allocation was 8.1%, 160 basis points higher than the group with a less than 5% allocation. [It] also shows that the higher the allocation the better and . . . top decile performers have steadily increased their allocations over the years to a mean of 40%.
The message seems to be that private investors should emulate sophisticated endowments. The authors add that “illiquidity does not translate into greater risk, unless liquidity needs have been miscalculated. In fact . . . institutions with higher allocations to private investments fared better in most major market downturns.”
Public pensions apparently also benefit from private equity’s advantage over public markets. The American Investment Council, in its 2018 Public Pension Study, claims that across 163 US public pension funds, on a 10-year annualized basis, private equity returns were 8.6%, compared with 6.1% for public equities.
Despite higher valuations and the usual disclaimer that past returns aren’t indicative of future returns, private equity investors remain optimistic. In its 2018 Global Private Markets Review, McKinsey reports that 90% of limited partners (the asset owners that provide capital to private equity managers) expect private equity to continue to outperform public markets. “And so, capital keeps flowing in,” the McKinsey folks observe. The Hamilton Lane study shows that more than $6 trillion has flowed into private market investments over the last 10 years.2 This flood of money should only intensify over the next decade. Hamilton Lane projects that between $11 trillion and $15 trillion will be raised over the next 10 years.
Footnotes and Fine-Print Disclaimers Matter
From the perspective of finance theory, and common sense, it’s hard to justify why private companies should consistently outperform public companies. Imagine two companies that provide the same products and services. Suppose they have the same expected cash flows, have no debt, and are managed by twin sisters with the same business acumen and experience. The only difference is that one is a private company, while the other is publicly traded.
To value companies, investors discount their expected cash flows. These two hypothetical companies have the same expected cash flows. The discount rate for both should vary as a function of the risk-free rate and a premium for risk. Over time, why would a stake in the private company appreciate much faster than a stake in its public twin if they are both valued based on discounted cash flows? And why would the private company’s business risk be lower than the public company’s?
Perhaps there’s a premium for illiquidity. Shareholders in the private company may require a higher rate of return in exchange for the inability to buy or sell shares easily on the open market. The puzzle remains, however, whether this premium justifies a 270% increase in the Sharpe ratio, as reported in the Hamilton Lane study.
Answers may lie in the footnotes and fine print of the studies reported above, which contain statements such as “Performance results do not reflect the deduction of any applicable advisory or management fees” and “based on internal rates of returns.” These studies and projections are from firms that have skin in the private equity game: Hamilton Lane is a private equity manager; Cambridge Associates and McKinsey count many private equity investors as clients; and the American Investment Council is an industry advocacy group that represents private equity firms.
Academics, on the other hand, are incentivized to let their data speak and to provide as much transparency as possible with regard to their assumptions. Perhaps they provide more of a truth-seeking assessment of private equity’s performance compared with that of public markets.
Remarkably, the academic consensus appears to be that private equity does not outperform public markets over time. For most of my career, I’ve tried to provide a bridge between academic research and practice. This conclusion may represent one of the biggest chasms between academic research and conventional wisdom that we have ever faced as an industry.
A wide range of biases in marketing presentations may explain the chasm. The three key issues are that private equity returns are levered, locked up, and self-reported. These biases interreact and compound each other. First, leverage amplifies returns, and when those returns are smoothed, risk- adjusted returns are automatically overstated. Second, private equity managers ask investors to precommit capital that they can deploy and return as they see fit, typically over long lockup periods. Portfolio managers in public markets, in contrast, have no control over capital inflows and outflows. Third, private equity firms are granted substantial flexibility in how they report the value of their investments, as these investments don’t trade frequently, if at all.
In “Private Equity Performance: Returns, Persistence, and Capital Flows,” Steven N. Kaplan and Antoinette Schoar (2005) conclude that “average [private equity] fund returns approximately equal the S&P 500.” Their sample covers a 21-year period, from 1980 to 2001. They do not control for leverage. Hence, they assume that private equity investments have a market beta (sensitivity to the S&P 500) of 1.0. They do not control for the possibility that private equity funds may be more exposed to systematic risks than public equities.
The only adjustment they make is to account for the “locked-up” component, i.e., the timing of cash flows. Internal rates of returns (IRRs) for private equity funds are value-weighted, while public market benchmarks are time-weighted (all years get the same weight in the calculation). Because private equity managers benefit from significant flexibility on the timing of cash flows, this comparison may favor private markets.
To adjust for this bias, Kaplan and Schoar calculate a public market equivalent (PME) benchmark. PMEs provide a value-weighted version of the S&P 500’s returns that assumes the same cash inflows and outflows as for the private equity fund. Hence, they neutralize the timing advantage for private equity.3 PMEs are typically reported as a ratio between public and private market returns. A ratio above 1.0 means that private equity outperforms the public market, and vice versa for a ratio below 1.0.
To illustrate, suppose the S&P 500 returns +5% one year and +20% the following year, for a compound (time-weighted) annual rate of return of +12.2%. Let us assume that over the same period, a private equity fund’s investments delivered the exact same returns as the S&P 500. However, the private equity manager deployed $10 million in capital at the beginning of the first year and added $100 million before the second year. In this case, the private equity manager had more dollars exposed to the very good year ($110 million invested at +20%) than the “just OK” year ($10 million invested at +5%)—good timing.
The private equity manager would have been able to return $132.6 million to the investor at the end of two years, for a value-weighted IRR of +18.6%. If we compare the IRR of +18.6% with the S&P’s return of +12.2%, should we conclude that private equity outperformed public markets by more than +6%? Is this outperformance true “alpha”? In this case, PME is obviously 1.0, as the underlying returns were the same each year. All we did was adjust for the timing of cash flows, and alpha went from +6% to zero.
Kaplan and Schoar found that the average PME for 746 private equity funds was 0.96 if calculated on an equally weighted basis and 1.05 if calculated on a size-weighted basis (large funds tend to perform better). These results were not statistically different from 1.0. Recent studies have confirmed these PME-based results for a wide range of datasets and time periods.4
These studies mainly adjusted for the timing of cash flows. Others have highlighted additional biases in published private equity returns. In the first few sentences of their 2009 article “The Performance of Private Equity Funds,” Ludovic Phalippou and Oliver Gottschlag get straight to the point:
The performance of private equity funds as reported by industry associations and previous research is overstated. A large part of performance is driven by inflated accounting valuation of ongoing investments and we find a bias toward better performing funds in the data. We find an average net-of-fees performance of 3% per year below the S&P 500. Adjusting for risk brings the underperformance to 6% per year.
These numbers are remarkably far from those reported by Hamilton Lane, Cambridge Associates, the American Investment Council, and others. They tell a completely different story about private equity as an asset class. Let’s investigate how Phalippou and Gottschlag assert that private equity underperforms public equities by –6%. Their study provides a good example of meticulous, truth-seeking academic research.
First, as in other studies, they use PMEs (which they call profitability indexes). Their PME methodology also adjusts for biases in how fund returns are aggregated. Traditional methodologies use total capital committed at inception to determine the weights used to calculate averages. However, these weights don’t reflect the present value of money invested. It’s a technical adjustment, but the authors find that “standard aggregation choices bias performance estimates upward.”
Next, they control for the “self-reported” bias. Net asset values (NAVs), which are part of private equity return calculations, tend to be smoothed and inflated. To correct for this bias, Phalippou and Gottschlag only select funds that have been terminated. Curiously, many terminated funds report a residual NAV in private equity databases, even though cash inflows and outflows have stopped and the funds have been closed. The authors write off those “living deads,” residual NAVs, because they don’t represent actual cash returned to the investor. In a sense, living deads may represent the aggregate effect of the overvaluation of NAVs over the life of the investment. In the words of the authors, “Funds that have reached their normal liquidation stage still report large accounting valuations for ongoing investments, which biases performance estimates upward.”
To adjust for the sample bias—the fact that only funds with good returns report to the main databases—they add funds that aren’t available in commercial databases. These funds underperform the broader universe.
Then they adjust for “levered” bias. They account for systematic risks, which include leverage in leveraged buyouts and small/growth factors in venture capital. Other studies have shown that these types of systematic risk adjustments alone can equalize the returns of private and public equity.5 One study (Phalippou, 2014) has shown that leveraged buyout funds underperform a levered-up benchmark composed of small and value public market indexes by –3.1% per year. Moreover, even without leverage, some studies (Stafford, 2017; Ilmanen, Chandra, and McQuinn, 2019) have shown that private equity underperforms small value stocks quite significantly after private equity fees are considered. Other studies (Kinlaw, Kritzman, and Mao, 2015) have shown that sector effects matter.
Last, Phalippou and Gottschlag estimate that fees represent more than 25% of the value invested in private equity (–6% per year, which matches their reported level of underperformance for private versus public equities). To the extent that marketing presentations for private equity use gross-of-fee returns, this bias may be the most worrisome. The first question any investor should ask is whether returns are gross or net of fees.
Some Practitioners Don’t Buy the Private Equity Hype
Many sophisticated investors are also skeptical of private equity’s supposed superiority as an asset class. In the 2006 article “You Can’t Eat IRR,” Howard Marks, cofounder of Oaktree Capital, first explains that the IRR methodology has some advantages. With conventional time-weighted returns, large percentage gains on a small amount of capital (“big percentage gains on small dollars”) inflate returns. In contrast, IRRs adjust returns for the amount of capital deployed.
However, Marks also points out a major issue with IRRs: they adjust for when the capital was called, not actually deployed, by the private equity manager. He concludes that “a high internal rate of return does not in and of itself put money in one’s pocket.” To illustrate, he uses a real-life example. An investor received a report that showed his private equity manager had delivered an IRR of +27.1%. But the investor only grew his $750,000 investment into $1,023,000 in 4.5 years, for an annualized rate of return of +7.3%—a staggering difference.
Cliff Asness of AQR recently introduced a satirical private equity fund called S.M.O.O.T.H.6 The fund will be invested in liquid assets, but it will deliver remarkably low volatility and great diversification. The small print reveals that the fund will be afforded extra leverage, lockups, and flexible NAV valuations and reporting. Asness jokes that “due to our new proprietary S.M.O.O.T.H. process, even if we fail to deliver this lack of correlation, the S.M.O.O.T.H. Fund will still report returns mostly unrelated to normal markets.” He goes on to say:
This proprietary process involves implementing our normal hedged liquid alternative process, but only marking to market occasionally, and then reporting some combination of the weighted average of the prior few years’ prices, with a healthy weight also given to our own unaudited estimates of what the Fund is likely worth (based on how we think it should’ve performed).
Satire aside, a 2019 article by my former colleagues Megan Czasonis, Mark Kritzman, and David Turkington shows significant biases in how private equity managers report performance. Private equity valuations are reported after the fact, over the course of the following two or three months. Returns for 2018 are calculated in Q1 2019, for example. This differs from mutual funds, which publish returns daily. This study reveals an asymmetry in how public market returns affect private equity valuations: private equity managers tend to mark up their investments when public market returns are positive, while they don’t mark them down as much when public market returns are negative.
During bear markets, or more generally, when investors rush for the door, the fact that private assets’ reported valuations can levitate above their true market value becomes painfully obvious. Illiquid assets embed a short option position. When liquidity abounds, the “liquidity option” pays a premium, but when liquidity disappears, the investor must pay up. In 2009, Harvard faced a liquidity crisis and received bids at 60 cents on the dollar on its private investments. A Forbes article (2009), “How Harvard’s Investing Superstars Crashed,” on the topic quotes the investment bank involved in the transactions: “The big discounts are due to ‘unrealistic pricing levels at which funds continued to hold their investments’ and ‘fantasy valuations.’”
In 2019, in the United Kingdom, investors in the Woodford Equity Income Fund were denied redemptions. The fund froze more than GBP 3.7 billion in investors’ money. This fund was marketed to be as liquid as a US mutual fund. The issue, according to the Financial Times, was excessive investments in unlisted and generally illiquid assets.7
In a 2016 article published on Alpha Architect, David Foulke sums up why many practitioners have started to take the side of academics on the private versus public equity debate:
Indeed, how meaningful is it to claim some performance advantage over public markets when you use leverage to buy small, cheap private companies, and exercise substantial discretion in measuring NAV? And then to say this “advantage” justifies an all-in ~7% fee?
Can Public Equity Fund Managers Deliver Private Equity–Like Returns?
Unlike private equity fund managers, public equity managers must provide daily liquidity. They must always be invested (plus or minus a small cash buffer). They must buy securities as investors give them cash to invest, even though markets may be overvalued, and sell securities as their clients redeem cash, even though it may be the worst time to sell. In other words, several of their buy-and-sell decisions are nondiscretionary. Unlike private equity managers, they can’t wait for the best opportunities to deploy or return capital.
In a 2019 study that raises questions about the supposed superiority of index strategies over active management, titled “Do Mutual Funds Trade on Earnings News? The Information Content of Large Active Trades,” Linda Chen, Wei Huang, and George Jiang analyze a large sample of nearly 2,000 US public equity mutual fund managers’ trades from January 1998 to March 2015. They show that discretionary trades (large active trades) by skilled mutual fund managers add significant value over passive benchmarks.
In contrast, “forced” trades—those generated by fund flows (subscriptions and redemptions)—detract from returns. The authors explain that public equity managers offer two benefits to investors: professional portfolio management and investment liquidity. The latter comes at a cost, because fund flows are typically “uninformative.”8 If it weren’t for fund flows, the study shows, skilled public equity managers would outperform the bottom quintile of managers by +1.27% per year, after adjustments for market beta, size, style (value versus growth), and momentum factors.
In the 1999 article “Investor Flows and the Assessed Performance of Open-End Mutual Funds,” Roger M. Edelen explains this effect from a theoretical perspective:
The conventional analysis [which shows that mutual fund managers underperform passive benchmarks on average and have no market timing skill] gives no consideration to the fact that fund managers provide a great deal of liquidity to investors and thus engage in a material volume of uninformed, liquidity-motivated trading. . . . In an asymmetrically informed market with costly information production, equilibrium is attained only when liquidity-motivated traders sustain losses to informed traders. . . . Thus, any trader forced to engage in a material volume of liquidity-motivated trading in a financial market that is in informational equilibrium will be unable to avoid below-average performance, ceteris paribus.
Academic theories that rely on assumptions of “an asymmetrically informed market with costly information production” and “informational equilibrium” may seem opaque to investors, but the idea is simple: if we believe in a zero-sum game, stock pickers who trade just to provide liquidity to their investors are at a disadvantage to those who trade based on information. Edelen provides an intuitive example:
Consider the performance of an open-end fund manager who occasionally has private information that leads to positive risk-adjusted returns, but who also satisfies investors’ liquidity demands. A well-functioning performance measure should identify this manager as being informed. Yet fund flows force the manager to engage in liquidity-motivated trading. Depending on the timing and relative magnitude of information arrival and investor flows, the fund’s average risk-adjusted return could very well be negative even though the manager is informed. Thus, the very act of providing a liquid equity position to investors at low cost, arguably the primary service of an open-end mutual fund, can cause an informed fund manager to have negative abnormal returns.
Basing his findings on a time period that varies across funds (with a mean start date of May 1985 and mean end date of July 1989), Edelen calculates that for every unit of flows-based trading, defined as 100% of the fund’s assets in a year, performance deteriorates by –1.5 to –2%. Based on his analysis of a sample of 166 mutual funds, he concludes that “controlling for this indirect cost of liquidity changes the average fund’s abnormal return (net of expenses) from a statistically significant –1.6% per year to a statistically insignificant –0.2%.”
This result raises an important question: What if we allowed lockups in public market portfolios? Edelen estimates that for the median fund, about half of all trading volume is due to liquidity provision, as opposed to informed trading.
What if we went one step further and completely leveled the playing field? What if we allowed public market fund managers the same flexibility across the board (use of leverage; when to call, deploy, and return capital; etc.) as private market fund managers? If we set aside return-smoothing shenanigans, there is no reason why we must limit private equity–like structures to private markets.
For an ongoing research project, my colleagues Rob Sharps, Rob Panariello, Camila Arbelaez, and I have surmised that many long-term investors in public markets would be prepared to accept similar terms as those afforded private equity managers in exchange for enhanced performance. Unlike with private market investments, the investors would benefit from the added confidence of transparent and unbiased valuations, as well as the confidence that in a crisis, the underlying investments would remain liquid.
Many institutional investors may want to go one step further and encourage public market fund managers to use appraisal accounting on such a product. These investors might think that appraisal-based returns help them manage their career risk. Appraisals often cushion losses, at least in the short run. Smoothed returns look great in asset liability studies. (One manager selection professional once told me that if we could deliver a smoothed version of the S&P 500 index, many asset owners would be prepared to pay very high fees for it, even though the underlying return stream can be obtained for about 5 bps in an index fund.) But to use self-reported appraisals when marked-to-market data are available would be bad practice from a risk management perspective, akin to how ostriches put their head in the sand.
In our research, we will estimate the benefit of leverage and lockups as applied to actively managed public market funds. Simple lockups don’t necessarily provide an advantage, especially if they lead to other agency and tax issues. For example, there’s a lack of evidence that closed-end funds (which eliminate liquidity provision) outperform open-end funds after leverage adjustments.9 These funds don’t have full discretion on when to return capital. In fact, most of them are focused on the provision of a high and predictable yield. However, private equity terms, which include the ability to call, deploy, and return capital at will, can magnify an investor’s ability to generate alpha. When combined with leverage, we surmise that such terms would let active public market fund managers go toe-to-toe with private markets fund managers. They would no longer be required to bring the proverbial knife to a gunfight.
Of course, if there is a liquidity risk premium, private assets may still outperform public markets. But with the amount of money deployed in private markets, which seems to have reached bubble-like levels, one could question whether these markets will deliver a liquidity risk premium going forward.
In prior studies focused on PMEs, authors have assumed that private equity fund flows would have been deployed into an index (usually the S&P 500). In our study, we will substitute index exposures for actively managed funds. We will also adjust for the liquidity provision component in public funds. Essentially, we will calculate PMEs on reverse-engineered, “locked-up” versions of a large sample of public fund managers’ track records. Depending on the outcome, we may also consider a product development effort to create private equity–like, actively managed public equity vehicles.
The bottom line is that private equity is not necessarily a free lunch. Still, private equity and other alternative asset classes can have their place in investors’ portfolios, if the asset allocator can identify good managers and model risk thoughtfully.
Portfolio Construction with Alternative Assets
In Chapter 17, we will review portfolios that broadly incorporate the body of research and best practices we have discussed throughout this book on return forecasting, risk forecasting, and portfolio construction. Many of these portfolios won’t include alternatives because most tax-deferred US retirement plan sponsors seek to minimize fees and maximize liquidity. Other model portfolios will include about 10%. For qualified investors who can forgo some liquidity in their portfolios, all things considered, a 10% allocation to alternatives seems reasonable. This weight is close to the market portfolio’s allocation (Doeswijk, Lam, and Swinkels, 2014); and as we have discussed in Chapter 1, when in doubt, the market portfolio provides an anchor to portfolio construction. Other investors, such as endowments and wealthy individuals with a long time horizon, may want to allocate more to alternatives. But they should remain mindful that alternatives are not a free lunch, and that the “alternatives” category remains ill-defined and includes a wide range of investments (hedge funds, private real estate, private equity, infrastructure, agriculture, etc.).
Notes
1. Hamilton Lane, Investor Conference, April 2, 2019, http://www.hamiltonlane.com.
2. Broadly defined, including private equity, venture capital, private credit, private real estate, and infrastructure.
3. To be fair, it could be argued that private equity fund managers have better market timing skill than public equity managers, but this “skill” would be hard to measure because public equity managers don’t have the same flexibility on how they deploy cash.
4. Harris, Jenkinson, and Kaplan (2016) found that private equity provided zero alpha versus public markets from 2005 to 2014. L’Her, Stoyanova, Shaw, Scott, and Lai (2016) adjusted for other factors and improved on the PME methodology and found zero alpha. Brown, Harris, Jenkinson, Kaplan, and Robinson (2015) compared four private equity datasets and found that leveraged buyouts (the main style of private equity investing) did not outperform public markets from 2006 to 2014.
5. See Franzoni, Nowak, and Phallipou (2012) and Gupta and Van Nieuwerburgh (2019) as examples.
6. See Asness (2019).
7. https://www.ft.com/content/82bab25c-972c-11e9-9573-ee5cbb98ed36.
8. See Edelen (1999).
9. See, for example, the article by Elton, Gruber, Blake, and Shacar (2013), which is quoted in an exhaustive review of the literature on closed-end funds by Cherkes (2012).