Questions You Never Thought You'd Ask

Hold on to your hats:  Are more efficient markets really that good for us?

I know, I cannot believe I wrote it either.

Let's be precise.  If we define "efficient markets" as a full complement of Arrow-Debreu securities, (these are securities that have a payoff of one in a single, particular state in the future, and pay off zero otherwise..pause and think about this...that's a whole heck of a lot of securities!) then we can accept the fact that "efficient markets" is a theoretical construct.  All we can strive for are "more efficient" markets by constructing more an more useful securities.  Useful being the operative term.

Why are a complete set of Arrow-Debreu securities useful?  If we had them, we could hedge anything.  Because we don't, we can't always hedge.  This means messy states of the world will exist.  Think May's "Flash Crash".


In this story, we learn about one after the fact solution to the Flash Crash.  A husband and wife pair of market micro-structure experts, working with a high frequency trading practitioner at Tudor, have developed a measure called VPIN, for "volume-synchronized probability of informed trading."  Apparently this measure predicts movements in volatility, and could have predicted the Flash Crash. 

The researcher at Tudor has applied for a patent on the measure, with the logical next step of a futures contract on it.  We can only assume options on VPIN follow shortly thereafter.  The VIX (volatility index) followed a similar securitization path.

So, how could I be opposed to such a securitization, which only serves to reduce market inefficiencies?  Safety measures have consequences.  Sam Peltzman's research on seat belts tells us everything: Drivers who wear seat belts drive more recklessly. 

Since the Flash Crash likely had roots in lousy liquidity, I find the situation ironic.  Design securities with lousy liquidity to hedge a liquidity driven crash!  

How do I know VPIN based securities will have lousy liquidity?  No one wants to be long VPIN risk.  In a very simple sense, this means you lose a whole bunch of money when a disaster hits the equity market.  (I wrote here about hedging catastrophic market events.  Providing insurance for truly catastrophic events requires massive amounts of capital and very high fees: you have to be certain you'll have capital after the event happens, but at the same time you have to be paid enough to have your capital doing nothing at other times so you don't lose it in the crash.) 

There are no natural buyers of VPIN risk.  (Who wants to agree to lose money in a liquidity driven crash?)  Lot's of sellers means some combination of lousy liquidity and horrible volatility imbalances in options of VPIN...until someone constructs futures (and options!) on VPIN volatility!  That way market participants magically hedge the VPIN illiquidity!  You see where this goes?  Rinse and Repeat.

We'll keep hiding the risk in more and more esoteric derivatives, but the risk doesn't go away.  As much as it pains me to say this, I can't quite see how more securities really makes us better off.