We are all familiar with the claim that private equity significantly outperforms public markets. Any survey coming out these days indicates that investors expect private equity to trounce public equity this year and in the years to come. Using data from Burgiss, widely regarded as the most comprehensive and accurate database of private equity fund cash flows, we look at recent PE performance and the bill that came with it.
We first need to choose a performance measure. We opt for the Public Market Equivalent (PME), currently the best tool for comparing the performance of private equity and public equity. This measure compares an investment in private equity to an equivalently timed investment in the public market.
Second, we need to choose a public market index for comparisons. At its origin, PME came with the S&P 500 Index as the benchmark. For now, let’s keep with tradition, and use the S&P 500 as our public market comparator.
Third, we need to select a sample. There are many types of private equity funds and, therefore, many samples one can choose. We want a recent and relatively homogenous sample that represents most of the capital and, if possible, performs better than the rest. The winner is: North American and Western European funds, classified as equity or real-asset funds by Burgiss. These funds basically do either leveraged buyouts (LBOs) – applied to real estate, infrastructure, etc – or venture capital. In both cases, the payoff is expected to be like a long call option on equity value (due to leverage, or the staging of venture-capital investments). Funds from the rest of the world don’t perform as well and neither do debt funds. We also exclude the most recent vintage years (2016-18) because these funds are young and underperform but they have an excuse (lots of fees are paid upfront). Finally, we choose the 10 vintage years before 2016: 2006-2015. Nice and round.
Our sample includes 2424 funds with a total size of $2.2 trillion. We find that PME as of March 2018 is 0.97 on average (median is 0.98) and total value to paid in multiple (TVPI) or total value distributed, compared to invested is 1.46. Verdict: performance of these funds is slightly below that of the S&P 500 Index.
What is the bill for this performance? Carried interest charged and latent (i.e., carry due on what is not exited yet) is unknown, but we can proxy for it by observing that 1518 funds are above the usual 8 per cent internal rate of return (IRR) hurdle. These funds have a total size of $1.52 trillion and TVPI of 1.64. Given their TVPI and a carry of 20 per cent of profits, it should amount to $195 billion. In short, we find carry is about 10 per cent of the total amount of money raised.
This computation makes some assumptions but nothing substantial. For example, it conservatively assumes that the funds that are not in-the-money have never been above an 8 per cent hurdle, (hence have not charged carry) and never will be above that hurdle. It uses net TVPI instead of gross TVPI to compute carry and assumes an American style deal-by-deal carry structure. We would expect European-style whole-of-fund carry to generate a similar amount but more spread through time.
Of course, there are more fee sources than carry. Management fees, portfolio company fees, fund and portfolio company expenses all would contribute to the bill – and these are much more difficult to estimate. Let’s keep it all on the low side.
A management fee of 1.5 per cent of fund size for the first five years of a fund’s life is a lower bound. Applied to our sample of funds, this is about $165 billion. Assume one-third of that amount for the following five years of a fund’s life (very much a lower bound), and management fees are above $200 billion.
To this, you would then add all other fees and expenses but let’s ignore all these and stick to the total fee we have counted so far. We have already reached a fee bill of $400 billion for net-of-fees returns slightly below those of the S&P 500 Index.
Change the sample
Rest assured that this estimate is robust with respect to changes in sample construction and assumptions. For example, if we include only US LBO funds (which are the best performers), PME is higher, at 1.04 (about 1 per cent outperformance of the S&P 500 per year). Total size is $773 billion. Funds in the money have a total size of $567 billion and a TVPI of 1.67. Total carry is about $76 billion – again 10 per cent of the money raised. Fees would be at least as much – and performance is close to that of the S&P 500.
Note that any of these investments would have a beta above 1, hence the benchmark should be higher than the sharemarket. We have also ignored any liquidity premia to compensate for ex-ante capital commitments and the illiquidity of the funds. Yet, private equity might bring diversification benefits that we have not accounted for here, and there may be positive externalities such as access to co-investment opportunities, learning about private markets, etc. Finally, these are just averages (if you select better funds, you do better).
Debunking the myth
Where does the myth of private equity outperformance come from then? First, recent history. Private equity funds outperformed the S&P 500 in most years from the 1990s until the mid-2000s. However, this might be due partly to the choice of benchmark. The average US stock outperformed the S&P 500 over this period by as much as private equity outperformed the S&P 500. Hence, private equity did as well as the average US stock before the mid-2000s. As it happens, since then, the average US stock has the same performance as the S&P 500 Index. This means the average private equity fund has returns similar to those of the average US stocks at any horizon.
It is important to bear in mind that each index is an active trading strategy. Thus, different indices perform differently. The S&P 500 is the most famous index but the types of stocks that it holds and trades are not comparable to the types of companies in which PE funds invest. In fact, now that the S&P 500 has been doing well, and private equity returns are in line with those of the S&P 500, we increasingly hear that the S&P 500 returns are driven by a dozen stocks that have nothing to do with PE-type companies. But instead of choosing an index that better resembles PE, most people now choose to use the MSCI World Index for comparisons. This index has had poor performance at any horizon and has lagged the S&P 500 Index, in particular, over the last decade. Hence, most presentations nowadays show that, compared with the MSCI World Index, PE is still outperforming.
Other data providers compute performance using end-to-end net asset value (NAV) internal rates of return. This method takes the NAV of all funds in the sample at the starting date as the initial investment, then uses aggregate cash flows each quarter, and the NAV of all funds at the end date, to calculate an IRR. But this measure is positively biased, as PE investments used to be more often held at cost, while post-2008 NAVs are closer to market value. This methodology also includes funds raised outside the sample period; for example, the performance for the last 10 years includes funds raised in 2003-07. By contrast, we used funds raised after 2006 and thus simulate the experience of an investor starting in 2006 who bought a representative set of funds. Furthermore, an internal rate of return is not an actual rate of return. Consequently, other data providers’ results will be slightly different to those discussed above.
More transparency please
Of course, the above analysis could be criticised for relying on back-of-the-envelope calculations using generic assumptions about fund terms to compute the fee bill. But lack of transparency regarding fees remains a great hurdle for academics and practitioners. Academics would welcome the details necessary to make a more precise estimate.
Ludovic Phalippou is the author of the bestseller Private Equity Laid Bareand is a tenured faculty member of Saïd Business School, University of Oxford.
Neroli Austin is a PhD candidate at Saïd. Prior to commencing her studies, Austin worked as an economist at the Reserve Bank of New Zealand for three years.