What’s in a Quartile Anyway?
Private equity is an asset class where relative performance remains one of the biggest puzzles. In a world where benchmarks and universes and relative comparisons are so complicated, why have we as an asset class tied ourselves to a ‘quartile ranking’ as the primary indicator of good and not good?
First and foremost, think about a private equity manager and the compensation structures they operate under and we as investors accept. For a manager to earn coveted carried interest, what do we as LPs measure them against? Do we say, if you aren’t top quartile, you don’t get paid? Of course not. The manager themselves have no control over how their peers do. Private equity is not an asset class where everyone fishes in the same small pond. The sheer volume of companies available for ownership is vast, and nowadays, it’s rare for 2 GPs to own the same company at the same time. Because of that, this is not a zero-sum asset class. Everyone can do something different and still generate their target performance. And make no mistake, most managers are incentivized by that simple fact. If they can generate performance over and above a pre-determined absolute level, they can share in the profits. That is the motivation for the managers, to generate those levels of returns. Of course, ego runs high in this asset class, so I’m not naively saying these groups don’t care about how they perform relative to peers, only that they don’t (and an’t) manage to that figure. It’s largely unknown and is a loser’s game.
So maybe the purpose of a quartile isn’t to measure the manager but the allocator. Were you able to pick the best manager relative to their (mysterious) peer set? It seems like a reasonable enough exercise, but this assumes that all investors have access to all opportunities at all times. We know that is not true and that the world is not created equal for whatever reason. Private markets are still largely inefficient, and the act of sourcing funds is even more inefficient than the collection of performance data. In addition, many funds are not open to new investors (but included in a universe). Other investors are simply excluded from participation due to characteristics outside their control, be it disclosure, structure, size, etc. It strikes me as an unfair assessment of someone’s relative skill if they don’t have the same access as others based on things they cannot control.
The other confounding component of our insistent reliance on quartiles is there is no single source of truth. As an industry, we’ve made tremendous strides to become less opaque, and there are now several data providers who claim to cover ‘most’ of the market. The challenge is the methodology, collection, and vetting of underlying returns. These are varied by provider, leading to each universe giving you different outcomes. Some are black box approaches, others rely on public investors, still others allow for self-submission, and in the end, none are right, and none are wrong. They are all valuable data points, but this leads to the opportunity for the GP to pick and choose where they rank themselves, obviously the one where they are ranked most favorably. Several research studies have pointed out that more than 75% of the market can claim top quartile status based on one of the major universes using one or more of the primary performance metrics, with some latitude to define a vintage year as well.
This gives us an imperfect assessment of performance, but something that many in the industry call a ‘benchmark.’ Unfortunately, these so-called benchmarks (I prefer to call them universe comparisons) fail nearly every test (suitability, pre-specification, investability, measurable, unambiguity, et al.) outlined by the CFA Institute in what makes up a valid benchmark.
I’m not saying throw out quartile rankings, but use them with extreme caution. For example, I recently reviewed a fund with very strong absolute returns but was in a vintage year with only 11 funds represented in a particular universe. This fund ranked it in the 3rd quartile — but the real question is, should I care? Does that 11-fund universe tell me anything about the manager in question and their ability to generate returns? The headline of ‘3rd quartile’ doesn’t paint an accurate depiction of this manager at all.
Over the last 15 years, I’ve spent countless hours trying to figure out a better mousetrap. The dawn of PME comparisons and all the variations we now have are very valid, especially in asset allocation models. Still, many don’t view public equities as their guidepost or alternative. The flaws in the universe models are apparent, so what does that leave us? In truth, it leaves us with a back-to-basics model, to the roots of how our managers are compensated. Private equity began as an absolute return asset class, and I think we need to remember that. We need to set realistic targets for our managers, have terms that reflect that expectation rather than an antiquated cost of capital assumption, and hold them accountable for the only thing they can control — their own performance.
The same goes for assessing an allocator. Select a realistic, target LONG-TERM return. The derivation of that number will likely be institution-specific, based on their own needs. Once that number is determined, stick with it. If desired, set interim checkpoints based on capital deployment or length of holding period. It is essential to measure progress towards that goal on a regular basis but to keep in mind the asset class is not meant to deliver that return in a short period.
Before I retire from the asset class, I have no doubts that we will achieve a suitable ‘benchmark,’ but until then, let’s focus on a more simplified and realistic measurement of performance to determine what is good or not good.
