Martin Zweig’s Foresight on the 1987 Stockmarket Crash

 

This past week I read Josh Brown and Jeff Macke’s new book Clash of the Financial Pundits (New York: McGraw-Hill, 2014).

 

One of the most interesting chapters was about when investment manager Martin Zweig appeared on Louis Rukseyser’s Wall Street Week on Friday, 16th October 1987, and predicted a stockmarket crash. You can see Zweig’s prediction in the above clip, 6:38 to 8:36 minutes. In the clip Zweig uses a combination of analogical reasoning about past market crashes, and observational studies on current market events and money flows.

 

Black Monday occurred on Monday, 19th October 1987.

 

Zweig was a PhD graduate and econometrics expert who also influenced the trend-following trading subculture. Zweig coined two phrases — “Don’t fight the Fed” (Federal Reserve) and “The trend is your friend” — which influenced momentum and trend-followers. George Soros trader Victor Sperandeo adopted the first phrase; popular trading culture adopted the second phrase — often without original attribution. This is how rumours work through virality and social contagion.

 

The difference is that Zweig had expertise and skills that set him apart from retail traders, and from momentum / trend-followers.

 

A few years ago I found a copy of Zweig’s book Winning on Wall Street (New York: Grand Central, 1986 / 1990). It’s an interesting artifact from the Masters of the Universe period (Tom Wolfe) of 1980s Wall Street. I rank it alongside George Soros (The Alchemy of Finance) and Peter Lynch (One Up on Wall Street) as a personal theory of financial markets and a codified trading methodology.

Hedge Fund Secret Source

The New Yorker‘s John Cassidy recently asked why some hedge funds make so much money.

 

Cassidy like hedge fund critic Les Leopold focuses on two primary reasons: (1) the ‘2 and 20’ fees that hedge fund managers charge investors where the funds charge a 2% administration fee and take 20% of the profits; and (2) carried interest loopholes in United States tax laws that hedge funds are structured to take advantage of.

 

Yet the other reason Cassidy does not explore is the hedge fund secret source: their trading strategies and transaction execution capabilities.

 

The vanilla version of hedge fund strategies is well known. For instance, ‘long / short’ funds take a long (upside) position in financial securities whilst ‘shorting’ (downside) others. Global macro funds profit from geopolitical risk and central bank monetary policy. Distressed debt and special event funds make profits from turnarounds or from creating situations where there are crowded trades and rational herding among investors.

 

The secret source is how a vanilla strategy is transformed into one where there is an edge or positive expectancy that is in the hedge fund’s favour. Some pre-quant hedge fund managers learned this from formative childhood experiences playing backgammon and poker. The quants studied Andrey Kolmogorov‘s probability work, and applied it to market microstructure patterns of the order book, and price / volume dynamics. Others benefited from geopolitical events: the 1973-74 growth of offshore Eurodollar markets (Paul Tudor Jones); the European Union’s Exchange Rate Mechanism and Black Wednesday (George Soros); or understanding bubble dynamics in the 1995-2000 dotcom bubble (trader Dan Zanger) and 2007-09 global financial crisis (John Paulson).

 

One of the keys to this is having a transaction execution capability. It means having a prime broker relationship with more favourable terms than retail traders get. It means having the complex event processing / stream processing capabilities to identify edges / positive expectancy and to trade them in many different financial instruments, markets, and timeframes. This is why some retail traders look at Edwards & Magee-style technical analysis and signals software; successful proprietary traders use trading psychology and market microstructure theory; and quantitative hedge funds use computational intelligence, machine learning, and software agents.

 

Cassidy and Leopold rail against hedge fund managers as a financialisation symbol of extreme income inequality. Their arguments resonate with many people who are legitimately angry about how much money some hedge funds make – even though there is survivorship bias. But what Cassidy and Leopold may obscure is the fact that – to quote the mid-1990s television show The X-Files – the information to create hedge fund-like capabilities is out there, scattered, waiting to be identified and reassembled into new forms. William Gibson, Bruce Sterling, and Charles Stross have already given fictional hints in their novels about what this proto-cyberpunk world might resemble.

 

When these hedge fund capabilities ‘cross the chasm’ from the hedge fund managers (1%) to the multitudes (90%) then things will get even more interesting.

We Are All Traders Now?

Mark Pesce pointed me to Bernard Lunn’s article which contends netizens now live in a real-time Web. Lunn suggests that journalists and traders are two models for information filtering in this environment, and that potential applications include real-time markets for digital goods, supply chain management and location-based service delivery.

Lunn’s analogy to journalists and traders has interested me for over a decade. In the mid-1990s I read the Australian theorist McKenzie Wark muse about CNN and how coverage of real-time events can reflexively affect the journalists who cover them. As the one-time editor for an Internet news site I wrote an undergraduate essay to reflect on its editorial process for decisions. I then looked at the case studies on analytic misperception during crisis diplomacy, intelligence, and policymaker decisions under uncertainty. For the past year, I’ve read and re-read work in behavioural finance, information markets and the sociology of traders: how the financial media outlets create noise which serious traders do not pay attention to (here and here), what traders actually do (here, here, and perhaps here on the novice-to-journeyman transition), and the information strategies of hedge fund mavens such as George Soros, Victor Niederhoffer, David Einhorn, Paul Tudor Jones II and Barton Biggs. This body of research is not so much about financial trading systems, as it is about the individual routines and strategies which journalists and traders have developed to cope with a real-time world. (Of course, technology can trump judgment, such as Wall Street’s current debate about high-frequency trade systems which leaves many traders’ expertise and strategies redundant.)

Lunn raises an interesting analogy: How are journalists and financial traders the potential models for living in a real-time world? He raises some useful knowledge gaps: “. . . we also need to master the ability to deal with a lot of real-time
information in a mode of relaxed concentration. In other words, we need
to study how great traders work.” The sources cited above indicate how some ‘great traders work’, at least in terms of what they explicitly espouse as their routines. To this body of work, we can add research on human factors and decision environments such as critical infrastructure, disaster and emergency management, and high-stress jobs such as air traffic control.

Making the wrong decisions in a crisis or real-time environment can cost lives.

It would be helpful if Lunn and others who use this analogy are informed about what good journalists and financial traders actually do. As it stands Lunn mixes his analogy with inferences and marketing copy that really do not convey the expertise he is trying to model. For instance, the traders above do not generally rely on Bloomberg or Reuters, which as information sources are more relevant to event-based arbitrage or technical analysts. (They might subscribe to Barron’s or the Wall Street Journal, as although the information in these outlets is public knowledge, there is still an attention-decision premia compared to other outlets.) Some traders don’t ‘turn off’ when they leave the trading room (now actually an electronic communication network), which leaves their spouses and families to question why anyone would want to live in a 24-7 real-time world. Investigative journalists do not generally write their scoops on Twitter. ‘Traditional’ journalists invest significant human capital in sources and confidential relationships which also do not show up on Facebook or Twitter. These are ‘tacit’ knowledge and routines which a Web 2.0 platform or another technology solution will not be the silver bullet for, anytime soon.

You might feel that I’m missing Lunn’s point, and that’s fine. In a way, I’m using his article to raise some more general concerns about sell-side analysts who have a  ‘long’ position on Web 2.0. But if you want to truly understand and model expertise such as that of journalists and financial traders, then a few strategies may prove helpful. Step out of the headspace of advocacy and predetermined solutions — particularly if your analogy relies on a knowledge domain or field of expertise which is not your own. Be more like an anthropologist than a Web 2.0 evangelist or consultant: Understand (verstehen) and have empathy for the people and their expertise on its own terms, not what you may want to portray it as. Otherwise, you may miss the routines and practices which you are trying to model. And, rather than commentary informed by experiential insight, you may end up promoting some myths and hype cycles of your own.

Are Financialistas Over Hedge Fund Chic?

You can blame George Soros for making hedge funds the dark horse of the irrationally exuberant 1990s.

As the public face of the Quantum Group of Funds, Soros gained notoriety for short selling the English pound in September 1992 and allegedly making $1 billion in profits.  Adam Curtis observes in his riveting documentary The Mayfair Set (BBC, 1999) that Soros’ victory signalled the first time that market speculators had beaten a country’s central bank.  In the aftermath Soros cultivated a master trader persona based on his personal ‘theory of reflexivity’ or how ‘participant’s bias’ can shape our actions in and perceptions of market events.  Hedge fund chic arose in Wall Street as investment banks rushed to found hedge funds, which use leverage and pooled capital to manage assets, derivatives and securities for an investor group.

Financialistas however are showing signs of buyers’ remorse as subprime turbulence brings an end to Soros-inspired hedge fund chic.  The high-profile collapse of Bear Stearns‘ two hedge funds in mid 2007 was only a precursor, Hedge Fund Research notes, of 170 liquidated in early 2008.  The survivors have adopted Soros’ global macro strategy which relies on computational finance and dynamical models of currencies, interest rates and other macroeconomic factors to achieve returns.

Global macro is a risky strategy for several reasons: it requires forecasting models of complex interactions, computing power and fund mangers with impeccable judgment for asset allocation.  In fact global macro deals with a specific risk class known as systemic risk that results from business cycles and macroeconomic movements, thus it cannot be diversified away.  Add funds’ massive leverage of pooled securities, industry secrecy, little government regulation and hypercompetition between different funds and managers, and an accurate calculation of risk-return is difficult.  These challenges overshadow the potential of applied research solutions, such as Fritz Zwicky‘s morphological analysis, a problem-solving method which deals with ‘multi-dimensional, non-quantifiable problems’ – relevant to the macroeconomic factors and systemic risk in global macro strategies.

Hedge fund chic faces several other problems.  As an investment category hedge funds have matured and their combination of high leverage and high management fees are unsuitable for many non-institutional investors.  Subprime fallout is triggering change in US financial and regulatory institutions which will inevitably lead to more rules and regulatory oversight of edge funds and managers.  Internally, hedge funds also need to separate managerial processes (principal management, portfolio execution) from financial reporting (mark to market book) and governance (board, corporate and policies & procedures).

Which means despite Soros’ alchemical touch hedge fund chic may now be a fad.

Bryan Burrough on Bear Stearns’ Demise: A Dark Possibility

Bryan Burrough is legendary in M&A circles for co-writing Barbarians at the Gate (Harper & Row, New York, 1990) with John Helyar, the cautionary tale of RJR Nabisco’s leveraged buyout and the winner’s curse faced by deal-maker Henry Kravis.

Burrough’s latest investigation for Vanity Fair contends that short sellers used CNBC and other media outlets to spread rumours that destabilised Bear Stearns and sparked a liquidity run on the investment bank’s capital.  Burrough’s thesis has sparked debate that overshadows his investigation’s strengths: a strong narrative and character portraits, new details of the negotiations with JPMorgan Chase and the Federal Reserve, and a cause-effect arc that shifts from CNBC’s internal editorial debate to the effects its coverage has on the marketplace and the subjective perceptions of individual investors and senior decision-makers.

In the absence of a ‘secret team’ or a ‘smoking gun’ how could Burrough’s thesis be tested?

Theoretically, Burrough’s hypothesis fits with: (1) a broad pattern over two decades of how media outlets respond to media vectors, systemic crises and geostrategic surprises; (2) the causal loop dynamics and leverage points in systems modelling; (3) the impact that effective agitative propaganda can have in psychological operations; and (4) the complex dynamics and ‘strange loops’ in rumour markets (behavioural finance) and rumour panics (sociology), notably ‘information cascade’ effects on ‘rational herds’.

This is likely a ‘correlation-not-cause’ error although it does suggest a dark possibility for strategic intervention in financial markets: could this illustrative/theoretical knowledge be codified to create an institutional capability, deployed operantly, and which uses investor fears of bubbles, crashes, manias and various risk types as a pretext for misdirection?  Behavioural finance views on groups and panics, and George Soros‘ currency speculation against the Bank of England’s pound on Black Wednesday suggest the potential and trigger conditions may lie in the global currency/forex markets (using stochastic models like Markov Chain Monte Carlo for dynamic leverage in hedge funds) and money markets (using tactical asset allocation).  If possible, this capability could also create second- and third-order effects for regulators, the global financial system and macroeconomic structures, and volatility in interconnected markets, which may actually be more dynamic and resilient than this initial sketch indicates.

To meet quantitative standards and validate Burrough’s hypothesis a significant forensic and data analytics capability with error estimates would also be required.  ‘Strong’ proof may not be possible: Burrough’s hypothesis is probably an unsolvable ‘mystery’ rather than a solvable ‘puzzle’ (a distinction by intelligence expert Gregory Treverton that The New Yorker‘s Malcolm Gladwell later popularised).

Ironically, several CNBC analysts have already decided: they used parts of Burrough’s hypothesis to explain the subsequent short-selling driven volatility of Fannie Mae and Freddie Mac‘s stock prices in mid-July 2008.