There is a popular narrative these days that human fundamental investors are no match for quants and machine learning algorithms. They will be wiped out in the quantpocalypse. I only half believe this. My reasoning goes back to the sources of investing “edge.” We will approach the quantpocalypse through the lens of investment edge and by the end you will understand the relevance of the photo above.
First thing first. If you have no edge, and you know it, then by definition you should index your investments. That’s not meant as an insult to index investors. It’s a perfectly rational strategy. Beyond that there are three broad sources of investment edge (I do not recall exactly who came up with this but for the record it was not me):
Informational Edge: You know something others don’t about a security.
Analytical Edge: You process information in such a way as to arrive at a unique insight that indicates a security is mispriced.
Behavioral Edge: You are able to exploit other investors’ bad behavior and behavioral constraints imposed on them by others.
Qaunts are very, very good at gaining and maintaining informational and analytical edges. They are engaged in an ever-accelerating arms race to acquire and process data sets before information becomes widely disseminated and their edge is arbitraged away. This data includes things like cell phone location data and satellite photos of retail parking lots. Heard of people combing credit card data for insights? That is old news. I recently listened to an excellent podcast on this subject where credit card data was described as “table stakes.”
I will tell you right now that you as an individual have neither an informational nor analytical edge versus a decent quant shop. Most active fundamental managers don’t, for that matter. They are competing with the quants to call quarters–to identify which companies are going to beat consensus earnings estimates and which are going to miss them.
This is a fool’s errand. You will not beat a quant at the game of his choosing. That computer from Wargames said it best, “the only way to win is not to play.”
In my view, the type of edge that is still available to traditional fundamental investors is behavioral. There are two reasons for this. The first is that many quants are shorter term investors with a quarterly time horizon or less. The second is that many sophisticated investors, quant or otherwise, are constrained by their investors.
Wes Gray of Alpha Architect explains this using a poker analogy:
On one end of the table, we have our irrational investors. They drop their cards, they giggle when they get an Ace, and they ask people next to them “Is it a good thing if all of my cards have the people’s faces on them?”
On the other side of the table is an institutional poker player, hired by wealthy investors, to play poker as best as possible. This poker player is a pure genius, mathematically calculates all probabilities in her head, and knows her odds better than anyone. Now imagine that our super player, as a hired gun, has a few limits. “We need you to maintain good diversification across low numbers and high numbers. We also want to see a sector rotation between spades, aces, and clubs. Don’t take on too much risk with straights and flushes, stick to pairs like the market does…” No one would ever play poker like this. But in finance, this is how people play.
Now the cards are dealt. Super Player sees a great opportunity with a high chance of success, but it violates all the requirements of her investors. She doesn’t bet, and sure enough, she could have won big.
Thus the “easiest” edge for today’s investment professionals to exploit is behavioral. Specifically, you have a behavioral edge if you can arbitrage the time horizons and constraints imposed on other sophisticated investors.
I say “easiest” here because this doesn’t require investing millions of dollars in data sets, machine learning and top flight data scientists. It does, however, require a diligent and disciplined investment process, along with either a permanent capital base or highly aligned investors who are not going to bug out if a core position draws down 50% on a bad quarterly result.
Importantly, this is not something that can easily be arbitraged away by quants. It is a structural market inefficiency and for an investor that is a beautiful thing.
In my experience most non-practitioners have a limited understanding of the investment ecosystem, its participants and their varying motivations, constraints and time horizons. This leads to a monolithic classification of market participants. From the layperson’s perspective Bridgewater “competes” directly with Baupost in every position. This Level 0 view of the market is reinforced by media coverage that tends to compare apples to durian. Reality, as usual, is more complex (#notallhedgefunds?).
There is no “magical investment strategy” that will outperform in continuous time. Every strategy has strengths and weaknesses. The trick is to employ the right strategies in the right asset classes at the right time horizon based on your objectives.
To illustrate this another way, consider the example of American pilots flying F4F Wildcats against Japanese A6M Zeros (pictured at top) during the Second World War. The Wildcat was slower and less maneuverable than the Zero. But it was much better armored and also faster in a dive. The trick to staying alive as a Wildcat pilot was not to fight the Zero’s fight. Wildcat pilots were generally quick studies on that front (turns out staying alive is a pretty powerful incentive). If the statistics on Wikipedia are to be believed, the resulting kill ratio versus the Zero was an incredible 7 to 1.
I pre-ordered this book on Amazon after seeing it mentioned on Josh Brown’s blog,
