Private markets
13 June 2022
By Lukas Gehrig, Zurich Switzerland, Quantitative Strategist
Holding private equity investments can allow an illiquidity premium to be harvested, potentially boosting portfolio performance. But what exactly is the illiquidity premium and how can it be exploited? We explore historical estimates for this premium, analyse its trends over time, and discuss how best investors can incorporate private equity among their holdings.
The liquidity of a financial market helps oil the pace at which deals are made, the volumes that can be digested, and the spreads that are paid. Market liquidity is not a binary distinction. While public equities are traded every second, corporate bonds may be traded only daily, municipal bonds usually twice a year, and institutional real estate around every decade1.
The value of liquidity is most apparent when it comes to buying or selling an asset. For instance, in illiquid markets, a sale may incur high search costs (to find a buyer), transaction costs, and the trade itself could hit the price. A tightening of financial conditions would exacerbate all these issues and make selling illiquid assets in distressed markets a very poor choice. It is for these reasons that investors can demand an illiquidity premium for holding such assets. In other words: they get rewarded for providing liquidity to an illiquid market.
There are several ways to obtain the premium. You could invest in inherently illiquid assets, like real estate or private market funds; focus on illiquid parts of broader markets, such as less liquid bonds; act as a market maker; or simply rebalance your diversified portfolio frequently, which provides liquidity to sought-after assets and absorbs liquidity for assets that recently sold off.
In the remainder of this article we focus on the illiquidity premia harvested from private equity (PE) funds. This includes looking at how to measure such premia, analyse changes in them over time, and address volatility in conjunction with liquidity.
Private equity funds, by virtue of their complex capital commitment and distribution structure, are difficult to trade and may merit an illiquidity premium. Viewing them as an illiquid alternative to publicly-traded equity indices would, however, be flawed. Private equity funds also require closer attention to managing portfolio liquidity and pose more idiosyncratic risk than broad equity indices – making them suitable for sophisticated investors only.
Unfortunately, measuring liquidity premia from private equity is not straightforward. The difficulty arises from the irregularity of capital calls and then profit distributions that require a portfolio around any private equity investment to first provide, and later absorb, liquidity.
Traditional metrics, such as total return (price return and dividends), are not appropriate when applied to private equity in this instance. Instead, the world of private markets has created return-like measures, such as the internal rate of return, distribution versus called-capital multiples, as well as so-called public market equivalents. These measures, (see box-out) can be applied when comparing funds. That said, they are less useful when comparing private- and publicly-listed assets.
The closest metric to a total return index would be quarterly appraisal-based indices, such as the one calculated by data provider Preqin. Their quarterly indices combine data on actual capital flows with periodic fund manager appraisals of the remaining value in their fund. This allows the quarterly changes in the value of a fund to be computed. Aggregating this across PE fund vintages, yields one single total-return mimicking index.
In comparing the Preqin index with total return indices from public markets, two implicit assumptions are needed: private equity money is never idle and fund manager appraisals always reflect the “fair” value of the fund. This highlights a couple of important caveats to our findings: first, in order for the no-idle-money assumption to become less problematic, investments need to be staggered across several vintages, such that capital calls and distributions can start to cancel each other out.
Second, agency issues are likely to be larger than in public markets, where reporting standards are higher and investments are potentially open-ended, instead of time-limited to ten to fifteen years.
By comparing annualised total returns from liquid benchmarks with constructed return-like indices from Preqin, we arrive at an initial estimate of the average liquidity premium of 2% to 4% for buyout funds and 3% to 5% for the riskier early-stage venture capital funds, between 2007 and 2021 (see table).
|
Buyout funds |
Venture capital funds (early stage) |
MSCI World Small Cap |
Russell 3000 |
|
|---|---|---|---|---|
|
Annualised "return" |
12% |
13.5% |
8.2% |
10.3% |
|
Premium vs Russell 3000 |
2% |
3.2% |
|
|
|
Premium vs MSCI World Small Cap |
4% |
5.3% |
|
|
Source: Preqin, Bloomberg, Barclays Private Bank, September 2021
These figures are in line with several academic and financial industry reports2 and also happen to be similar to estimates from Ang in 2014 that an investor should require a return of 4% to 6%, in order to lock up capital for a maximum of 10 years.
The choice of benchmark is important, as not all return differences can be attributed to the illiquid nature of PE vehicles. For buyout funds, which make up the largest portion of the private equity universe, their target firms tend to have smaller capitalisations. Therefore, benchmarking them against the S&P 500 would be inappropriate. Also, geographical matching matters, as returns have tended to be higher in US equity markets than in European ones this century.
To understand more about how changes in illiquidity premia behave over time, we want to control for size, geographical, and sectoral exposures as much as possible. However, appraisal-based valuations are much less volatile than market-based ones, which makes a quarter-by-quarter comparison futile.
As such, we now turn to public market equivalents (PMEs). These discount all capital calls and profit distributions that occur during a fund’s lifetime to the inception date, using a chosen public market benchmark. PMEs larger than unity (see chart) suggest that investing in the fund would have been more profitable than investing in the public asset. While not allowing an absolute comparison of returns to be made, some lessons can still be learnt about the changes in the size of the premium.
In the left panel we display PMEs by vintage, where we run each vintage for a maximum of 15 years (not all funds are closed after 15 years). The best vintages in the last two decades have been those with inceptions early in the millennium, which ended up with PMEs that were mostly between 1.25 and 1.50. Only more recently have young vintages emerged with a PME of above 1.25 again. Looking at the simple average across active vintages, the illiquidity premium appears to have come down since the global financial crisis, in 2008, and could be growing again with the latest vintages.
Looking at the average PME (dashed line) we find the typical outperformance of private markets occurs during abrupt market sell-offs. If nobody can sell their investments easily, a self-fulfilling sell-off is much less likely. Moreover, the time right after such a crisis usually offers juicy acquisition targets for buyout funds.
As final PMEs are solely based on public market data and actual cash flow transactions, there is less to worry about when it comes to manager appraisal methods diluting our results. However, we assume that capital calls, distributions, and the public benchmark share the same risk.
This is problematic, as returns from a fund with 10 investments are likely to be more volatile than those from an index consisting of 3,000 constituents. Furthermore, part of the capital called is management fees, which represents risk-free liabilities. Taking this into account, Sorensen and others3 notch up the discount factor for capital distributions, given its increased riskiness, and find that depending on the co-movement of the fund with the index, PMEs should be at least 1.2, if not higher, to breakeven versus public stock markets.
With this in mind, the median performance of North American buyout funds over the last decade looks much less appealing. This brings us to the enigma of fund manager performance in private equity, where an often cited study by Kaplan and Schoar suggests that “top quartile” managers outperform the rest with some persistence, and every fund manager wants to be in the top quartile. Likewise, investors want to fill their portfolio with top-quartile funds.
The less regulated nature of private market reporting complicates sorting the wheat from the chaff in fund manager selection. So we rely on Preqin’s own quartile ranking, which assigns funds, by vintage and category, a rank for every reporting period. We then source all reported investments into buyout funds by the 20 most experienced fund investors across all investor groups, as well as the five most experienced for each of the following: public pension fund, private pension fund, sovereign wealth fund, fund of funds, and single family office.
By experience, we mean the number of active and past fund investments. Public pension funds make up most of the top 20 most experienced fund investors. Between them they have reported investments in around 5,400 funds since 1980. As such, we deem it less likely that they selectively reported investments which could bias the results.
Looking at the most recent quartile ranking for each fund, we can then gauge the ability of very experienced investors to cherry-pick first-quartile funds. The good news is that all investor types have invested in what turned out to be first-quartile funds at least 25% of the time (see chart). The bad news is that those who picked first-quartile funds most often, also pick fourth-quartile funds more than 20% of the time.
Ideally, we would report PMEs for each quartile in this sample of experienced investors. Unfortunately, the data are not available, which may be down to reporting being mostly voluntary. Therefore, we have to rely on less robust metrics to gauge the difference in the performance across quartiles.
The next chart depicts the internal rate of return net of fees and raw multiples for the (global) buyout funds reported by the 20 most experienced investors. Imagining the median fund to lie somewhere between the mean of the second and the third quartile, it is safe to say that first-quartile funds have – across all vintages – bested the median quite significantly.
For the nineties’ vintages, second-quartile funds also outperformed third-quartile funds significantly enough that it is safe to assume that they harvested an illiquidity premium. During the 2010s the case is less clear. For fourth-quartile funds, it is already transparent from raw multiples that investing in a public index would have made the investor better off.
The big dispersion in PE fund performance reminds us that the illiquidity premium is a theoretical concept, not a guaranteed payoff. This has two practical implications for investors, the need to diversify across vintages and focus on the fund selection process.
That said, for an investor considering a PE investment, harvesting an illiquidity premium is only one out of many reasons to consider the asset class. For instance, as we touched on in May’s Market Perspectives, from a behavioural standpoint: the illiquidity in PE may prevent panic-selling decisions.
This month we take a look at the major trends at play in the financial markets and the implications of surging inflation, central bank rate tightening, and slowing economic growth prospects. We also look at how private markets might be needed to bolster diversification, and what investors can do to limit the effect of what may continue to be a volatile end to the year.
