Do We Get the Full Picture of PE?

Are PE GPs telling the full truth, the whole truth, and nothing but the truth? Do LPs want them to? I shakily stand on the shoulders of giants to find out.

Image Credit: FT Alphaville

Picture this: I was sitting in my room in Jouy-en-Josas, working on a case for my LBO class and mulling the PE industry’s vaunted outperformance of public markets since the dawn of time, when I discovered[1] a 2019 article by Clifford Asness called “The Illiquidity Discount?” that picked away at the conventional thinking about PE. In the article Cliff argued that PE LPs were arguably accepting lower expected returns in exchange for illiquid assets and the understated volatility implicit in that illiquidity. Colour me intrigued. Part of his argument rested on the PE industry measuring itself against inappropriate benchmarks.[2] So I figured I would investigate this and went off to talk about it with my friend and professor, Denis Gromb.[3] When I told him about the article he leapt up[4] to his white board to draw a curve from a paper he had recently read about revisions to prospect theory,[5] showing his characteristic enthusiasm. We lamented that it was too late for me to turn this into an MBA project but I resolved to undertake some research to really look into this. Well long story short I hit a wall in this project[6]—I am no economist and my knowledge of quantitative finance is more suited to its consumption than its production.[7] Jump-cut to the present day when I came across an FT Alphaville post about the phenomenon using the provocative term “Volatility Laundering” featuring the above funny meme-esque image.

Image Credit: Bain & Co. 2022 Global Private Equity Report

But, before we dive into Volatility Laundering, let’s talk a bit about PE returns and how great they are and how they clobber their public comps. Man, just look at how much better those top quartile performers are. Furthermore, the folks at Bain who put this together for their 2022 Global Private Equity Report would like to remind you that most of the S&P appreciation lately has been driven by exposure to giants like Microsoft, Apple, and Amazon—I suppose implying that the buyout funds are actually doing even better than you think! Are they? Depends on how you measure. The Cliff Asness post that got me on this road drew heavy inspiration from and highlighted a paper produced by his ACQ colleagues, “Demystifying Illiquid Assets: Expected Returns for Private Equity” published in The Journal of Alternative Investments. It’s a good technical read—but here are a few things I took away. First, using a large cap index designed to capture the entire equity market (i.e. with a β of 1) is not appropriate for buyout funds who are very clearly not investing in the whole market but rather generally in companies that have greater exposure to the size factor. Furthermore they don’t account for the L word.[8] Accordingly they propose using a leveraged small-cap portfolio as a benchmark. When you measure using that yardstick—the outperformance seems less… outperform-y. They note that from 1986-2017 while PE outperformed large caps by 2.3% p.a. (on an arithmetic mean basis) that gap shrunk to 0.7% against a1.2x levered small cap index. and compared to a Fama French small-value factor portfolio,[9] PE underperformed by 1.6%. Take a look-see at this graph comparing PME vs the S&P 500 and a leverage and factor adjusted S&P 600. Not so rosy. Now granted this stops in 2014 and PE has been on a absolute rip of late.[10]

Image Credit: “Demystifying Illiquid Assets: Expected Returns for Private Equity”

So, that’s returns. Look, they are good. Nobody can argue that. Are they as good as they should be though? Unclear. And that’s why we now turn our attention to return’s good buddy and long-time denominator co-star, risk. Earlier, I pointed out that the ACQ folks identified a mismatch in factor tilt of PE benchmarks and, implicitly, questioned the way we think about PE portfolios. Choosing the S&P 500 as benchmark suggests that PE funds have a β of 1. To put it succinctly, that beggars belief. A priori, it is easy to doubt this, as the whole animating principle of PE and especially good PE is being selective and investing in good companies and not the whole market—to say nothing of investing in smaller companies that generally have higher βs. Don’t take my word for it, the folks at ACQ note that “many empirical studies have concluded that a beta estimate of 1.2–1.5 is more realistic.” The corollary, PE’s volatility is higher than implied and hence risk adjusted returns are overstated.

This leads me to the next issue and the one that Asness originally got me thinking about which is how does illiquidity impact volatility. Let’s engage in a thought experiment:[11]

Image Credit: NOAA

Above we have a week’s worth of tide data for Charles Pierson Dot Com’s Global HQ in Rockland, ME. As we can see the tides vary about 2 metres. We can see that there is a lot of variation in how high the water is and were we to constantly record its height we could get a very accurate picture of its variation. Let’s call this, I don’t know, the public market approach. What if I decided[12] to see this as a much more stable level of water. I could easily do that by only going and looking at the water level at high tide or at low tide. Hey presto, my σ is now close to 0. For symmetry, let’s call this the private market approach. Point is, if you only look some of the times at something that moves a lot instead of all the time, you will miss a lot of the volatility. Furthermore, if I want to show you that the water level is always high, I can just measure it at high tide.[13] Public markets are naturally going to show you all the movement and even if you are really, really assiduous in private markets you will miss much of that movement.

That’s if you are assiduous. But what if you are incentivized to be less assiduous. Well you get this:

What are we looking at? Well, it’s private companies over a billion in valuation and the latest greatest IPO stocks like Airbnb, Palantir and Rivian. This doesn’t perfectly isolate public vs private as a factor, but it’s close. They track closely throughout time as they probably should. But let’s zoom in on early 2020 when we can see the public companies tank and the private companies… don’t. This break in correlation ends once things start ripping again with a bit of a lag for the privates. Now look at the beginning 2022, when things started going pear shaped for markets. After a bit of a drop the privates just… stop moving. A very compelling theory here is that the people who mark the value of these assets just stop when things get rough and then wait until the all-clear has been sounded and they can start credibly marking their assets up again.[14] Generally speaking, asset correlations increase in times of market stress, the opposite appears to be the case here. And it is hard not to draw the conclusion that there is something afoot.

But are PE LP’s just the marks of GP’s here? Let’s see what Bob Maynard, the CIO of Idaho’s Public Employee Retirement System said at a CalPERS conference in 2015:

We did know that our actuaries and accountants would accept the smoothing that the [Private Equity] accounting would do. It may be phony happiness, but we just want to think we are happy and they actually do have consequences for actual contribution rates we are going to be able to put in place[.] Even if [Private Equity] just gave public market returns, we’d be in favor of it because it has some smoothing effects on both reported and actual risks.

So I guess if Bob here is representative, the LPs want to close their eyes and think of a happy place while their GPs tell them stories? Yeah, that’s pretty much what some researchers at University of Florida argue in their paper “Catering and Return Manipulation in Private Equity.” They compare the performance of REITs and PERE[15] returns to show that not only do the PE fund manipulate return and volatility characteristics but that funds that do so are rewarded by LPs. This passage sticks out and highlights how agency theory could explain this[16]:

If a GP boosts or smooths returns, perhaps by strategically timing asset acquisitions and dispositions or by misstating the values of underlying assets, investment managers within LP organizations can report artificially higher Sharpe ratios, alphas, and top-line returns, such as IRRs, to their trustees or other overseers. In doing so, these investment managers, whose median tenure of four years often expires years before the ultimate returns of a PE fund are realized, might improve their internal job security or potential labor market outcomes.

Oof. Now they largely address the issue of IRR manipulation, which is pretty easy to manipulate when you exercise some control over the timing of cashflows. Especially noteworthy is how IRRs seem to peak around times that they are raising money. Viz:

Image Credit: “Catering and Return Manipulation in Private Equity”

They do point out that GPs do not engage in direct overstatement of fund NAVs. And that funds that do are often punished for doing so. But I think the tidal thought experiment is still instructive here—if you measure less often you will observe less volatility. And then it’s hard to look at the IPO vs private unicorn graph and not have some questions about NAV accuracy leap out at you.  

Fundamentally I think the questions raised, and to some extent addressed, here lead me to further question what we we are doing with PE as an asset class. I am not saying PE is bad, far from it, I think PE and especially PE done well is an important activity. At the core, the idea of buying businesses to make them better and then sell them for more than you paid is a good one. But a capital A Alternative is meant to provide uncorrelated returns and enhanced risk/return characteristics. And I am less than convinced that PE actually does that now, despite what people may say. I think at one point when PE and LBOs were new and value abounded it was much easier and less risky for smart people to do this—but that that is increasingly difficult to do in a climate where so much money is chasing a finite number of good deals. Furthermore, I think we are coming out of a low rate environment where equity and debt financing was plentiful, prices always went up, and Alternatives were less and less an alternative. There appear to be structural issues that allow PE to be used as a tool to dress up performance in a host of ways. So we shall see where PE goes from here. And look—there are still tonnes of Mittelstand businesses to roll up in Germany. So with that in mind, who wants to give me some money? I’ll do 1&15 and I promise to not mark NAV too often.  

[1] Read: I was distracting myself on Twitter  

[2] More on that later

[3] It pains me to even think about this, as we have lost tragically lost Denis. Denis meant and still means a lot to me and to everyone in the HEC community. Please read more about Denis and his immense academic, pedagogical, and personal contributions here or here  

[4] No doubt fuelled by the coffee and Auchan escargots de chocolat he had been forcing on me and consuming himself. This was just one of the many memories that people who knew and worked with Denis will always cherish It was a pleasure to know a person who was so comprehensively generous and it breaks my heart that he’s gone. But his memory will live on in all the lives he touched and the work he inspired   

[5] I think; this was all very deep economics that is frankly beyond me. Clearly I should have taken notes—I could have closed this lacuna

[6] That is to say, the wall of my lack of experiment design ability and rudimentary R and Python skills

[7] I can read a paper about it—but I sure cannot write one

[8] No, not the Showtime series—leverage. It’s leverage.

[9] And what is a good buyout target if not a smaller undervalued company??

[10] But so has everything, thanks ZIRP! (Probably)

[11] Long-time readers rejoice: it’s a nautical metaphor

[12] Or was financially incentivized

[13] Caveat time: yeah I know that this metaphor isn’t perfect because tides are totally predictable unlike markets… look, you get it though

[14] I am not saying definitively that is what they are doing, but there is strong circumstantial evidence supported by a priori reasoning to think that they may be.

[15] Private Equity Real Estate

[16] Sorry, another block quote.