Tuesday, December 8, 2015

How Dynamic Hedging Will Fail

I expect dynamic hedging strategies to fail badly sometime in the next couple years.

There are companies that sell options and use dynamic hedging.  This means they make frequent adjustments to their overall position so that on net it remains neutral for small movements in the market.   They have to keep adjusting their position all the time in order to sort of be hedged.   For example, if they sell some calls on a stock and just the right number of puts on the same stock then for small changes of the stock in either direction the gains and losses on the puts and calls can compensate each other.  But after the stock moves a bit it will take a different number of puts and calls to balance each other, so they must either change the quantity of puts/calls they hold or buy/sell some of the stock so that once again they are net neutral for small changes in the stock.

 It is not a true hedge though.    In the event of a crash, they can not adjust their position fast enough and so they are not net neutral during the crash.   In the call/put example above, the calls become nearly worthless and the puts very valuable.  What they lose by being short the puts is far more than the gain being short the calls.   As they try to adjust their positions, say by shorting the stock as it goes down, they will contribute to the crash.   In a crash, companies using dynamic hedging could go bust.  


It is kind of similar to the portfolio insurance that people were doing around the time of the 1987 crash. It all looks ok on computer simulations which assume nice continuous pricing changes and that their buying and selling does not change the price much. However, after an investment idea is popular, large numbers of people doing the same thing means things do not work like they did in the computer simulation.   The people using the idea can drive the price and the price can move so fast that the idea can not be implemented as planned.    So the very activity of "portfolio insurance" or "dynamic hedging" done by many people can cause a crash. When the hedging algorithm fails, companies based on it will go bust.

A company selling a put on the S&P but using dynamic hedging is like a company selling insurance against a crash while not really being in a position to pay off on the policy in the event of a crash.   If too many companies selling such options go under, it could be very messy.

Murphy's Law says that anything that can fail will eventually fail.  Dynamic hedging can fail.


4 comments:

  1. O/T: Vincent, John Cochrane has an interesting article up now, in which he writes the following:

    "We seem to have in front of us a pretty clear measurement that long run dynamics are stable.

    "Nothing" is astounding. This dog that did not bark has demolished a lot of macroeconomic beliefs:

    MV = PY. Sorry, we loved you. But when reserves go from $50 billion to $3 trillion and nothing at all happens to inflation -- or at most we're arguing about percentage points -- it has to go out the window. "

    http://johnhcochrane.blogspot.com/2014/11/the-neo-fisherian-question.html

    Jason Smith is very excited about the rate raise at the Fed yesterday in that it offers a natural experiment for his model. Naturally he's made a quantitative forecast:

    http://informationtransfereconomics.blogspot.com/2015/12/zirp-is-over-let-experiment-begin.html#comment-form

    Jason left a comment for Cochrane too.


    So what do you think is in store? Is Cochrane right? Is MV=YP effectively dead? Jason's view is more nuanced: it's depends on the number of "microstates" in the economy: For large economies, yes, it's probably dead, but for small economies it's probably still in effect.

    ReplyDelete
    Replies
    1. Thanks much! I replied to both of them. I think interest rates going up will make velocity go up and so inflation go up. We are now testing theories!

      Delete
    2. Vincent, I didn't realize at first that Cochrane's post was not new: I first saw Jason's new post, followed by a new post by David Andolfatto (that Jason referenced) which in turn referenced Cochrane's older post. Just so you know.

      Delete
    3. Vincent I left a question for you here:
      http://informationtransfereconomics.blogspot.com/2015/12/zirp-is-over-let-experiment-begin.html?showComment=1450457983048#c2047841307519544727

      Delete

Looking for polite debate on ideas. Never attack a person. Be nice.