Academic

There has been discussion on this blog, as well as more broadly, about whether the recent financial crisis was a ?black swan? (i.e., a transformative event that was unforeseeable a priori).  On the one hand, you could point to low interest rates, double-digit home price appreciation, and other signs of frothiness to argue that the collapse was predictable.  There was also a handful of commentators who predicted aspects of the crisis.  On the other, you could point to the numerous experts, who at least outwardly appeared to see nothing dire on the horizon.  (Nassim Taleb, by the way, has chimed in; he calls the collapse a ?white swan.?)

I think the debate about how to properly categorize recent events is ultimately unwinnable, however, because the contours of what constitutes a black swan haven?t been sufficiently well-defined.

To identify an event as a ?black swan,? we must determine that the event?s occurrence was ?unforeseeable.?  But how are we to tell?  We could search for evidence existing prior to the event on which a prediction of the event?s occurrence could have been based.  But how much evidence would we need to see before we conclude that an event was foreseeable?  No obvious threshold comes to mind.

We could also look to see whether anyone predicted the event.  The logic being that an event isn?t unforeseeable if people actually foresaw it.  But this is also problematic.  If a science-fiction writer foresaw the Internet, does this mean it was foreseeable?  The invention of the laser is also pointed to as a black swan.  But Einstein predicted it, and scientists working on the invention had to see it coming.  Black swan no longer?  The underlying uncertainty here appears to be in how many people need to predict something before we call it predictable.  We could argue that if something happens that is contrary to prevailing sentiment, it is a black swan.  But this appears far too broad.

Further problems involve how we define the actual ?event? that we are seeking to label as a black swan.  The more specifically we define it, the less foreseeable it becomes.  People may be able to broadly predict future occurrences, but be unable to foresee exactly how they will unfold.  Again consider the Internet.  Could we argue that it was foreseeable if someone foresaw networked computers?  Or would someone have to have foreseen the Internet?s utterly transformative impact? 

The underlying uncertainty of the black-swan concept renders the debate about our current financial collapse rudderless.  Yes, there was evidence that could lead one to suspect a coming collapse.  Was it enough to make it foreseeable?  Some pundits did predict aspects of the financial crisis.  But were those predictions sufficiently precise for us to conclude that they predicted it? 

I think ultimately it?s not very useful to engage in this debate.  Nor do I think we need to crystallize the exact parameters of black swanness.  Rather we should look at how the insights that make the black-swan metaphor so appealing can help us understand the recent collapse and society in general.  I?ll look at this topic in my next post.