Are You Holding Out On Me?
Investors want to know everything about a fund, and most quantitative funds will not share their models. This alone is often viewed as a lack of transparency and many investors see it as a key point of discomfort relative to traditional equity and fixed income funds. For contrast, when a traditional manager shares their portfolio holdings investors call it “full transparency”. This is belief seems intuitive but is actually pure fallacy.
For a quant, sharing their model is akin to sharing the blueprint for (a) replicating their future trades and (b) trading against them.
So to be fair, if an investor insists on requiring all the factors and models before investing in a quant they should also insist that traditional managers share all of their trades and the full rationale for those trades in advance. That is the analogous level of transparency, and of course it is completely unrealistic to expect.
This common double-standard serves to underscore an important point; quants simply have more concrete information they could share with investors; namely their precise investment process.
Great Stories ≠ Great Returns
For a typical long/short manager, the closest we get to understanding the investment process is when the manager describes the rationale for their positions. The manager presents a (hopefully) clear and articulate thesis, and the investor usually ends up allocating to managers with good historical performance and who give the most compelling speeches about their positions in the hope that good storytelling acts as a proxy for reality.
If the average performance of a long/short fund is any guide, there isn’t enough non-fiction going around to justify the fees. When looking past all the pretty stories, the real factors that go into any individual buy/sell decision for a discretionary manager can vary far and wide. Managers may weight information based on:
- Past investing or life experiences (“I made money this way once before…”)
- Their constantly evolving views on the importance of certain financial, momentum, valuation or economic metrics.
- Their mood (“I feel lousy today therefore my portfolio positions feel lousy.”)
- Their team dynamic (“This team member is smart and presents well, therefore the idea seems like a good one.”)
- An immeasurable number of human and non-human factors.
The portfolio holdings may be available to the investor at any given time but the ‘process’ is only as clear as the eternal labyrinth of human psychology.
Science? Yes Please.
On the contrary, quants know exactly which factors they value, how they weight them, and the precise circumstances under which they make buy & sell decisions. To assess a quant (or any manager for that matter) we must go one level deeper and assess their methods for coming up with their investment factors. That is the secret sauce and the most effective way to diligence the manager.
The key here is not the factor itself, rather:
- Why a factor is important?
- How does the manager select & vet factors?
- How do they decide to modify or eliminate a factor?
- How do they anticipate changes in the market that may make a factor less relevant?
- What restrictions and risk parameters do they place around the factor(s) and model(s) to maintain a stable portfolio?
The answers should be rooted in a scientific process. There should be a specific, clear hypothesis, a test that avoids biases and data mining, and a result that supports or refutes the theory with supporting statistics.
What Else Can You Ask For?
Aside from assessing the development of factors, other items that can be helpful:
- Holdings. Some quants will share all their portfolio holdings, though others may be concerned that the portfolio moves can be reverse-engineered to ascertain the model.
- Factor breakdown. Quants often invest in factors, so understanding the exposure of those factors is often more valuable than having a list of the 500 securities they hold at any given time. What (i) country (ii) industry (iii) sector (iv) instrument or (v) risk factors are they exposed to? Most produce risk reports that break down these exposures in detail (internally & also for investors).
- Examples. Quants may also provide several examples of models that have been discarded or are no longer relevant in order to demonstrate how they develop a thesis, how they modify it, and how they adapt to changing market conditions. These can be extremely helpful in assessing the process of coming up with investment decisions.
Thank you for reading! Feel free to contact us with your thoughts & questions (or leave a comment below).