Lessons From a Hedge Fund Manager

In 2020, Paul Marshall, who co-founded the hedge fund firm Marshall Wace, wrote 10½ Lessons from Experience: Perspectives on Fund Management.  He called it a “small publication.”

Indeed, the book is slight in form and a quick read, but it sets forth a statement of beliefs that touches on many important concepts.  Therefore, you can use it to compare and contrast the views espoused to your own beliefs, as a jumping-off point for more in-depth examination of those concepts, or as the narrative frame against which you could do due diligence on the firm.

In the introduction, “The Great Disconnect,” Marshall says that “axiomatic thinking is inherently dangerous in the social sciences — basically in any domain which involves human agency in the systems you are trying to model and predict.”  Yet, the rationalist train of thought “led by the Chicago School . . . remains to this day remarkably disconnected from financial market practice.”  Not a surprising view coming from a practitioner of active management.

(The section headings below, other than the last one, are the lessons that serve as the titles of the book’s chapters.  None of the links used come from the book.)

Markets are inefficient

The biggest battle regarding investment beliefs concerns the efficiency of markets.  For Marshall, “Price formation happens with varying degrees of inefficiency, and price formation is typically volatile and non-linear.”  Not at all what the theorists of efficiency believe.

In contrast, Marshall cites the work of Benoit Mandelbrot, who shone a light on the normal distribution as an unrealistic representation of how markets actually work — even though it still underlies most tools of investment analysis.  The reflexivity theory of George Soros illuminated that disconnect, positing that “markets not only anticipate economic developments but actually drive them and are in turn driven by them, because ‘human beings are not merely scientific observers but also active participants in the system.’ ”

Generally, markets grow more efficient over time, but new inefficiencies develop and market crises spawn those reflexive reactions that belie the notion of efficiency as a governing force.

Humans are irrational

Marshall Wace tracks its portfolio managers and outside contributors in reference to the behavioral biases identified by Daniel Kahneman, Amos Tversky, and others.

In his review of some of the more common biases, Marshall admits that he is most prone to optimism bias (“I am too easily romanced by new stories and new opportunities”), while his partner Ian Wace tends toward mean reversion bias, a form of the “gambler’s fallacy.”  As he notes, “Taken together our style biases were a good hedge to each other,” in contrast to organizations where similarity is prized over balance.

He attacks the problem of overconfidence, writing, “When someone is doing well their portfolio should be subject to extra scrutiny and their ego deflated.”  (That is in stark opposition to the normal fawning treatment of them during periods like that.)  And he warns of a “more subtle form of sunk cost fallacy, relating to the irrational weight given to the time an investor has spent analyzing an opportunity.”

Investment skill is measurable and persistent

This lesson is among the most controversial, especially the “persistent” part of it.  To dive into the definitions of persistency and the measurement of skill is beyond this posting — and beyond Marshall’s short book.  But with more than twenty years of data, he believes “we are able to identify persistent skill with a high degree of confidence.”

But the percentage of successful trades for those with skill is a mere 52-53%, and the average for “a truly great manager” is only 55%.  Not what most outsiders would expect.

Of note, “the main threats to persistent performance are all character related” and “the reddest flags for underperformance . . . are problems in people’s personal lives — the three Ds of death, divorce and disease.”

In the short term the market is a voting machine, in the long term it is a weighing machine

This quote from Benjamin Graham gets thrown around a lot.  Within this lesson, Marshall also looks at John Maynard Keynes, who “likened investing to choosing the winner of a beauty contest.”  Keynes wrote of the process as “anticipating what average opinion expects the average opinion to be.  And there are some, I believe, who practice the fourth, fifth and higher degrees [of reading those layers of opinion].”  You might see a connection to Soros’ reflexivity there.

Marshall says that his firm tries to combine the two kinds of machines, with its quant trading operation being in the voting business, while its fundamental managers try to weigh what will win out over longer time horizons.

Seek change

While Marshall emphasizes the importance of catalysts in causing the revaluation of securities, he points out that you have to be nimble and willing to get into an idea early to really gain an advantage:

By the time a story becomes so well packaged that it can be pumped on CNBC or at Breakers or the Sohn Conference it is probably too late.  The interesting moment is when the idea is just in its dawn, half-glimpsed and half-understood.

And he highlights the importance of the narrative — and how typical due diligence approaches fall victim to manager narratives:

What everyone should know is that it is very easy to tell a story about a stock.  Your ability to tell a story has almost nothing to do with your ability to pick stocks.  In the case of some successful managers, it is almost inversely correlated.  Yet it is the staple, still, of many due diligence processes.  By all means ask questions about stocks for entertainment and to illustrate the process of the manager.  But don’t give it much weight in your due diligence.

Here, uncharacteristically for an asset manager, Marshall encourages those doing due diligence to resist the easy narratives that managers put forth.  (Instead, the goal should be to crack the narratives.)

The best portfolio construction combines concentration with diversification

That sounds like a contradiction.

How can these two perspectives be reconciled?

The answer is that you cannot fully reconcile them at the level of a single portfolio — there are trade-offs between concentration and diversification, between return and risk.  But you can reconcile them at the level of the business and at the level of the clients.

Marshall wants concentrated pools that come from different sources to be combined together at the allocation level.  Therefore, he favors “platform” hedge funds which feature pods with many discrete strategies, and presumably would argue for an asset owner taking a similar approach across its portfolio.

Shorts are different from longs

For starters, “the informational bias of the market is entirely structured against the short seller.”  Plus, the mechanics of implementation are daunting.  While Marshall outlines the characteristics that make for a good short, it’s not an area for the faint of heart or the inexperienced.

A machine beats a man, but a man plus a machine beats a machine

This speaks to one of the most important topics of the day.  Marshall thinks that fundamental, systematic, and quantamental investing will converge further — and that each requires technological prowess for success.  By 2023, he expects his firm to be processing twenty petabytes of data per day, ample evidence that the game has changed from years gone by.

Risk management — respect uncertainty

The concepts of “risk” and “uncertainty” get conflated, and the industry’s cavalier use of the terms doesn’t help.  Statistical measures of “risk” can be used to guide exposures, but reliance on them as the essence of “risk management” overlooks the real uncertainty that exists.  Leverage, illiquidity, and the presence of real actors can easily overwhelm modeled scenarios.  (See reflexivity again.)

Size matters

The “guilty secret of the fund management business”?

Beyond a certain level of AuM, size becomes an impediment to skill-based returns as it raises trading costs in a non-linear fashion and reduces the flexibility of trading and risk management.

Marshall believes that a fundamental equity strategy should be “capped anywhere from $1-3 billion in capital.”  (That said, according to its website, Marshall Wace has more than $55 billion in assets spread across its different kinds of approaches.  How well will the “concentration plus diversification” approach deliver returns going forward with that kind of scale?)

And now for that half lesson . . .

Most fund management careers end in failure

“When a fund manager is popular, investors rush to their fund, pumping it up to a size which makes it increasingly challenging to deliver the same return per unit of risk.”  Then follows the loss of the manager’s halo, redemptions, and forced liquidations of assets, sometimes little by little and sometimes all at once.

(Ironically, such “failures” leave the managers with very large nest eggs, yet even in some famous funds investors on balance lose money, since the strongest performance often occurs when a fund is small.)

The final paragraph of the book:

Ultimately, of course, it is all about character.  If you do not begin your fund management career with a sense of your fallibility, you are likely to learn it.  If you do not learn it, you are likely to fail.

Lessons and beliefs

Any attempt to summarize lessons learned confronts the problem of where to land between a simple list and a weighty tome.  Marshall strikes a good balance, conveying the essence of his thinking and his firm’s approach.

Given the brevity of the book, the natural response is to want more information.  Among the topics of interest is TOPS:

TOPS (“Trade-Optimised Portfolio System”) is the name given to Marshall Wace’s “alpha capture” strategy, in which we ask outside contributors from the brokerage community to input their best ideas into an online portal which allows us to track their performance and to identify those contributors who have reliable information to offer.

The data would be fascinating to parse — no doubt providing an insightful study in behavioral finance — as would learning more about the implementation process.  How do the choices of the contributors ebb and flow in response to market dynamics?  Are the more persistent and strongest signals coming from those who are wrong or those who are right?  How are their firms rewarded or penalized in terms of trading business?  How are signals used directly and indirectly by Marshall Wace?

Such questions and many others are prompted by Paul Marshall’s book, which provides a peek inside a successful asset management firm and how theory, practice, beliefs, and lessons have melded into its approach.