Sep

26

"Never let a crisis go to waste." That political statement can have implications for those of us who speculate. Specifically, we can learn from what has recently happened, and by knowing history can hopefully avoid repeating those lessons.

I noticed just today that the option volume attributed to market makers ("MMs") was particularly high relative to that of firms and customers. That is quite logical, as emotional demand in options is usually offset by the MMs providing liquidity. That of course is the function of MMs, for which they are usually rewarded. But has that been the case historically, and if so, can it teach something?

I looked at the volume of market makers relative to their counterparties in two ways; first with the MMs buying and second with the MMs selling. Then I took the differences of those two calculations and smoothed it. More on the smoothing after you see the results:

Here you see SPX shaded to reflect the smoothed balance of MM liquidity. If SPX is blue, the MMs are providing liquidity to put buyers as of the previous day, and if orange, the MMs are providing liquidity to call buyers as of the previous day. These are of course generalizations, as I really do not know who is doing what, but simply following the logic of the positions. Of course it isn't perfect. If you are expecting certainty, you would be better off in another field.

The smoothing period is extremely interesting. These transactions by the MMs usually last 24 hours or less. But their counterparties do take positions and try to hold them. On average options positions by customers and firms last slightly less than a month, although they are frequently rolled-over. Thus the chart represents on-balance accumulated demand by options traders for liquidity. So what would you suspect to be the best smoothing period?

Wrong! It happens that the period used in the chart was a whopping six months. That fooled me also. That is, the data used in the smoothing is as "old" as six months. Of course, that is a static period; the best smoothing period will always turn out to be adaptive. But this works for illustration.

At this point the work is merely anecdotal; there is no statistical significance worth speaking about. But I present it here as something for further study.


Comments

Name

Email

Website

Speak your mind

Archives

Resources & Links

Search