Back in 2015, when CalPERS held a one-day private equity workshop, which was a vehicle for justifying its loyalty to the strategy despite the SEC and major press stories revealing abuses and embezzlement, its star witness, Harvard Business School professor Josh Lerner, gave a surprisingly tepid defense. Lerner conceded that typical returns in fact didn’t justify investing in private equity, given its higher risks. It made sense only if you could get into top quartile funds.

A new study confirms that Professor Lerner’s, and most private equity industry investors’ aspirations aren’t worth pursuing since superior performance in private equity no longer persists. Or as a Bloomberg write-up of that paper put it, Private Equity Legends Are No Longer Living Up to Their Hype.

What is nevertheless surprising is that the academics act as if their finding is novel. It isn’t, as well demonstrate shortly. Yes, there was a time, before the 2000s, when top private equity fund managers could regularly out-do their competitors. The most likely explanation is that the private market was inefficient enough that there were arbitrage opportunities and some firms were better and finding and executing on them than others. Too many parties chasing too few deals have bid away a lot of the upside.

We’ll also remind readers shortly that even if you believed that there were private equity fund managers who could regularly beat the odds, it’s no different than saying, “It doesn’t make any sense to invest in stocks unless you can invest in the top 25% performers in the market.” Investors have accepted that they can’t all be the next Warren Buffett and expect to out-invest everyone else; it’s now widely held wisdom that you are better off to accept market returns rather than doing even worse via ovetrading.

But even back in 2015, we wrote that academic evidence was mounting that so-called persistence of top quartile performance in private equity was a thing of the past. A new, large-scale study puts another nail in that coffin. We’ve embedded this NBER paper by Robert S. Harris, Tim Jenkinson, Steven N. Kaplan, and Ruediger Stucke at the end of the post.

It is worth noting that the authors of this study include some of the top academic researchers into private equity. Sadly, private equity academics are almost entirely captured; they make far more money consulting to private equity firms than they do from their day jobs. And you can see signs of that in this paper. The authors repeatedly tout investing in private equity and venture capital by comparing returns to the S&P 500. But that’s bogus, as referring to their own past publications would reveal. Private equity is higher risk that public stocks due to its illiquidity and higher leverage; venture capital is even riskier. The authors never once acknowledge that private equity needs not only to out-do public stocks, but to do so by a meaningful margin, which historically was set at 300 basis points. Venture capital should be benchmarked against smaller-cap stocks and have a risk premium at least as high as private equity.

Implicitly, the repeated references to private equity and venture capital returns compared to the S&P 500 is an effort to counter the widely-reported findings by Oxford professor Ludovic Phalippou, that since 2006, private equity has performed pretty much on a par with the S&P 500, which means its performance is too poor to compensate for its outsized risks.

Their study is noteworthy for being more comprehensive than past studies on fund performance, and in particular, in relying on a non-cherry-picked data set, one from Burgiss, a data service for fund investors. The authors have also taken the critically important step of looking at the reported performance of fund managers’ most recent fund at the time they are raising a new fund, which is typically on a five-year cycle. That is the data investors have available to them when considering a new offering; past studies have looked at what ultimately happened to the predecessor fund, which is something no investor could know at the time when it makes a commitment.

Nevertheless, the study still has limits in that it relies on IRR, as opposed to the less-game-able PME (public market equivalent). But the latter can’t be properly calculated until a fund is wound up. Specifically, the authors fail to acknowledge the rise of subscription lines of credit and how they can render IRR computations utterly meaningless.

Before 2000, the authors found that a fund that looked like a top quartile performer at the time of a new fundraising would give you one-in-three odds of getting a winner again, but that vanished:

…using performance information available at the time of fundraising, the results differ. For buyout funds with post-2000 vintages, performance persistence based on fund quartiles disappears. When funds are sorted by the performance quartile of the GP’s previous fund at the time of fundraising, performance of the current buyout fund is statistically indistinguishable regardless of quartile. First-time funds perform at least as well as any of the groups based on prior fund quartile rankings.

The authors did find that if you looked at final performance, that did predict performance of the next fund, even after 2000. But that’s of no use to those deciding which horse to ride. The study also found, contrary to conventional wisdom, that new funds performed as well as established funds. They also determined that top venture capital funds are likely to deliver again.

What is disingenuous is the authors’ position that their finding is news. It isn’t. We been pointing to evidence of the lack of persistence of top fund performance for years. And we are hardly alone. From the transcript of public comments at the aforementioned 2015 CalPERS private equity workshop, by Rosemary Batt, co-author with Eileen Appelbaum of the landmark book Private Equity at Work:

And so I want to talk a little more about the performance data and just expand on Professor Lerner’s excellent presentation….

Professor Lerner then refers to another really important paper by Robinson and Sensoy who report, they actually report performance based on real returns, so liquidated funds. And here they find that the average fund does outperform the stock market by 1 to 1.5%, depending upon the index.

The median fund, however, just matches the stock market, which means that 50% of the funds do not perform as well as the stock market. It is the top quartile funds that outperform both indexes by 3 or 4%.

So, if we put this together, then, Professor Lerner also points out this problem of persistence of performance. And the studies he refers to show that prior to 2000, the private equity firm with a top performing fund had about that 50% chance of being in the top performing funds in the follow-on fund. But since 2000, that probability has dropped to 22%, and that is less than would be expected if the distribution were random.

In other words, all this new study has done is refine what ought to be conventional wisdom, save that private equity investors have staffs who have job incentives to believe that fund-picking is a productive exercise, and those staffs have industry consultants who also have strong financial incentives to perpetuate that fiction. The older studies found that private equity top quartile persistence was actually higher, 50% versus 1/3 now, but found that since 2000, you’d do better throwing darts than investing in a prior-period top quartile fund.

And that’s before getting to the elephant in the room, whether investing in top-quartile funds, even if they could be identified, is attainable in practice. As we wrote in 2014:

Rather than question the logic of investing in private equity at all, everyone in the industry has convinced themselves that it is reasonable to believe that they can be the Warren Buffett of private equity. The investment consultants go through the shooting-fish-in-a-barrel exercise of convincing their institutional clients that each of them is prettier, smarter, and more charming than average, and therefore capable of achieving sparking results. Needless to say, flattery is an easy sell….

Fundamentally, this is an intellectually dishonest exercise, and diametrically opposed to the way many public pension funds construct other parts of their investment portfolios. With public equity in particular, it’s almost certain that a significant majority of U.S. pension fund assets are invested in index funds. That’s because pension funds have recognized that, collectively, they cannot do better than average, and that after paying active management fees, actively managed public equity portfolios typically perform worse than the market average.

So it’s not as if these investors are so clueless that they can’t grasp the point that all of them cannot achieve above average results, let alone significantly above average results. Instead, with private equity, there is a desperate desire to be in the asset class for reasons that probably reflect a combination of intellectual capture by the PE managers, political corruption in legislatures that control public fund board appointees, and the need to have a strategy that could conceivably solve the pension underfunding problem over time.

So while it is gratifying to see our contentions about private equity reaffirmed, it’s discouraging to see investors desperately throwing even more money at it. Just like second marriages, it’s a triumph of hope over experience.

00 Nov 2020 NBER private equity returns paper