When the market tide goes out, investment managers must rely on true investment skill, rather than general buoyancy, to keep them afloat. The question is (and has always been) how.
By Clare Flynn Levy
Clare Flynn Levy is CEO & Founder of Essentia Analytics. Prior to setting up Essentia, she spent 10 years as a fund manager, in both active equity (running over $1B of pension funds for Deutsche Asset Management), and hedge (as founder and CIO of Avocet Capital Management, a specialist tech fund manager).
Twenty-two years ago, in a land not that far away, I was a tech fund manager living through a market not dissimilar to this one. The stock market was still reeling from the 2000 tech bust, geopolitical concerns were raging, and a gloomy uncertainty loomed over Wall Street.
There are significant differences between then and now, of course, and no one knows whether we’ll see the 78% plunge the NASDAQ suffered during that time (we’re almost halfway there). But it does feel a lot like déjà vu.
A wise person* once said, “history doesn’t repeat itself, but it often rhymes.” A clear common thread between then and now is the fact that when the market tide goes out, investment managers must rely on true investment skill, rather than general buoyancy, to keep them afloat. The question is (and was) how to do that.
A fund manager’s job is to make decisions. All day, every day. Some of those decisions result in trades, and even more do not. The question is, which of your decisions are helping and which are consistently hurting performance? Which types of decisions are you actually skilled at making, and which would be better made by someone, or something, else? Could you be using your own energy more efficiently by making fewer, better decisions?
Two decades ago, these questions were nearly impossible to answer. The best performance attribution analysis — still the primary evaluative tool for many investors and managers today — starts with the outcome and works backward to explain it by comparing it to the performance of an index alternative. But that didn’t really help the manager: while it was (and still is) useful for explaining why the portfolio performed the way it did during a certain period, it could not (and cannot) identify what the fund manager could do differently, in the future, to get a better result.
Enter decision attribution analysis, the largest and, for investors, most consequential area of behavioral analytics. Becoming increasingly refined in recent years with the exponential growth in machine learning capabilities, decision attribution looks at the actual, individual decisions the manager made in the period being analyzed, along with the context surrounding them. It assesses the value those decisions generated and destroyed, and identifies the patterns — the evidence of skill and/or bias — among them.
Decision attribution is a bottom-up approach, compared to the top-down approach provided by performance attribution analysis.
“None of us is a perfect decision-maker — sophisticated allocators of capital harbor no illusions about that.”
If you’re thinking “but surely people make decisions differently in different market environments,” you’d be partially correct. Fund managers do pick different stocks at different points in the economic cycle — of course they do. But the picking decision is only one of many choices that a fund manager makes during the life of a position. There are also decisions about when to enter, how quickly to get up to size, how big to go, and whether to add and trim the position as time goes on. Finally, managers make decisions about when to get out, and how quickly to do so.
These decisions are less conspicuous, less analyzed, and, it turns out, a lot less variable. Having studied equity portfolio manager behavior for the better part of a decade, I’ve seen proof, time and again, that while we change our picking behavior as the market environment changes, the rest of our “moves” are more habitual.
Anyone who has historical daily holdings data on their portfolio has the raw material required to see where they are skilled as investment decision-makers, and where they are making consistent errors. I wouldn’t want to mislead: decision attribution is a complex endeavor. Any investor who has attempted to do it themself can attest to that. And while it’s interesting to do as a one-off exercise, it’s only really useful if it can be done on an ongoing basis; otherwise, how can you tell if your skill (and not just your luck) is improving?
It’s only recently that technology has made it possible to do decision attribution analysis on an ongoing basis in a reliable and useful way. When the last market bubble burst, it simply wasn’t an option. This time around, more and more managers are doing it — not only to understand what they can do to get a better performance result, but also to prove their skills to investors when their performance is negative.
None of us is a perfect decision-maker — sophisticated allocators of capital harbor no illusions about that. But as a portfolio manager, being able to show your investors, with data-driven evidence, that you know exactly what you’re good at and what you’re doing to improve goes a long way. And given the availability of the underlying data and, now, the analytical toolset, there’s really no good excuse not to do it.
A market environment like this one has a sobering effect on every investor. The upshot is that, at moments like these, we’re forced to look in the mirror, whether of our own volition, or, eventually, at someone else’s behest.
Instinctively we may shy away, for fear of what we may find out. But the clarity provided by decision attribution also provides a roadmap for improvement — a way to weather this challenging environment that didn’t exist in the past.
In other words, our current sense of déjà vu does not mean history has to repeat itself — or even rhyme — this time around.
* Mark Twain is typically credited with having said this phrase, but the evidence is more in favor of it having been written by the psychoanalyst Theodor Reik, in 1965.
