Last week, we covered a Spring 2007 research paper on hedge fund “contagion”. The paper, by Nicole Boyson of Northeastern University, Christof Stahel of George Mason University and Rene Stulz of Ohio State, showed the interconnectedness of various hedge fund strategy indexes. Today, AllAboutAlpha contributor Timothy Laing reports on a speech given by Christof Stahel last Friday in Zurich.
The Swiss chapter of the CAIA Association showed remarkable foresight earlier this year when they scheduled Christof Stahel to speak at their quarterly meeting on August 31st. Stahel and his co-authors released a paper in March 2007 that has seen a fair bit of press coverage since the recent subprime crisis led to a flurry of hedge fund losses (or was it the other way around?). Their paper “Is there hedge fund contagion?” has been quoted in industry journals, the mainstream press, and a number of financial blogs, assuring that Professor Stahel´s presentation last Friday in Zurich was well attended by a cross section of the traditional and alternative investment community.
As Stahel describes it, the research is only a first step. In fact, it raises as many questions as it answers. Stahel indicated during his presentation that his team plans to continue the research, which will attempt to drill down into the results of its first paper – exploring individual funds in an attempt to discover more precisely the origins of the “contagion”.
Recent events raise the question of how the behavior of hedge funds might impact traditional markets, and this discussion is notably absent from the paper. Stahel admits that, when they extend the data to include August of 2007, and in consideration of recent events, that they might have to “modify their conclusions”. In spite of this, and considering the severity of the recent shocks to the credit markets, the research is timely.
Unfortunately, the source of contagion in the financial markets is often impossible to trace and a two dimensional model cannot hope to encompass all the various levels of (inter-)action and causality. Essentially, Stahel argues that any complex, quantitatively driven model, be it a complicated hedge fund strategy or the system of traffic lights governing our daily commute, can break down due to an unpredicted or unpredictable event – a revaluation of an illiquid portfolio or a twelve-car-pile-up.
Since hedge funds invest in various traditional markets, they are both influenced by and have influence over these markets. So any attempt to quantify this behavior or interrelationship is bound to be frustrated by the ever-changing nature of those markets.
Following his presentation, Stahel was quick to note the shortcomings of the research, and cautions the market and practitioners on the “mixed results”. According to Stahel, these shortcomings – like those of other hedge fund studies – mainly stem from a lack of good data, particularly detailed daily returns. Stahel says that his colleagues and he are somewhat hampered by the limitations of commercially available databases (such as survivorship bias).
Still, this research is useful and it provides the basis for further academic investigation into the nature of hedge fund/traditional market contagion. Given the chaotic nature of the financial markets and their evolution into an increasingly complex system, what practical application does this paper offer? While Stahel’s model can predict how different hedge fund strategies will perform given the concurrent results of other hedge fund strategies, it does not address causality, timing or the “paths of contagion”.
Which leads us back to the central question on everyone’s mind: can hedge funds infect the rest of us? While many would argue that so-called hedge fund contagion has already spread to the broader capital markets, Stahel’s research has yet to pin-point the exact mechanisms for this. In addition, the important questions “who infected whom?”, and “how we can predict the next outbreak?” also remain unanswered at this point.
So stay tuned. And keep taking those Vitamin C’s just in case.