0
$\begingroup$

What are the steps involved in options market making?

Does it roughly follow this procedure:

  1. Choose a pricing model, e.g. Black-Scholes.
  2. Calibrate the model, e.g. Volatility.
  3. Quote a bid-ask spread for the option.
  4. Trade (which results in a net long/short position).
  5. Compute the Greeks to hedge the position.
  6. Repeat.

It seems to me that an important remaining factor would be forecasting volatility once the position is delta hedged?

Thank you in advance!

Improve this question
$\endgroup$

1 Answer 1

Reset to default
3
$\begingroup$

It's not clear from the question, but there are two scenarios:

1/ you have an actively traded and well-established options market in which you're going to get involved as a new participant and take on risk;

2/ you're establishing a new options market which is either non-existent or very illiquid.

In 1/ all you're really doing is providing liquidity. The more interesting question is referring to 2/. The steps you describe (with the exception of 3.) are in a sense posteriori. So if your question was only related to 1/ you can stop reading here! Btw model choice is relevant to the product being quoted. Let's stick to vanilla options (so just Black-Scholes).

In 2/ the key objectives are to understand the liquidity/flow of the underlying and analyze its historical realized volatility. This allows one to get a sense of where implied volatility, if it existed, may be. Next, some idea of the term structure of the vols is necessary i.e. how the vols vary with expiry. This is usually followed by price discovery (done in an inter-dealer setting showing out vanilla price runs), establishing demand/supply, open interest from prospective clients... etc. All of this derives from the dynamics of the underlying that you intend to quote the options on, as this is the only data you have to work with. Having a proxy market for options in another underlying which is strongly correlated to the one you intend to establish your new options market on is also very useful. Particularly to get an idea of a realized/implied vol spread (implieds usually trade over).

Market-making (as described above) is not exact science and is driven by necessity (demand/supply). This sometimes has an unfortunate side-effect in many younger options markets being very one sided (e.g. bid only). This is a huge drawback as things never really get off the ground and the market just dies. Having two-way flow is essential to the survival of a market.

Finally, options market-makers don't "forecast volatility": I'm not sure how one would even go about doing that. There is a notion of forward volatility which can be deduced from the term structure of the current spot volatilities. But this is in itself a separate market which has its own dynamics. The dynamics of the spot vols are driven by how much the underlying delivers as a well as demand/supply, and market-makers adjust their quotes primarily in accordance with these factors.

Edit: I should mention that all of the above discussion is only with respect to ATM vols. Establishing skew and off-strike vols is a whole other world of hurt, which I'll leave for another day.

Improve this answer
$\endgroup$
6
  • $\begingroup$ Thanks for your answer! What do you mean by "done in an inter-dealer setting showing out vanilla price runs" ? Also, why would forecasting volatility not be important ? The P&L of a delta hedged position depends on the difference between realised and implied volatility (and also the gamma). And there are plenty of time series models for forecasting/estimating volatility (GARCH family for example). $\endgroup$
    – FISR
    Sep 9 at 8:42
  • $\begingroup$ The inter-dealer market is where the major broker/dealers (your JP Morgans and HSBCs...etc) trade with each other, aka "the street". A price run is where a desk would show a bid/offer list of prices and various interests would counter etc. This arena is one way market-makers can hedge themselves. "Estimating" (realized) vol (using GARCH etc) is different from "forecasting" (where say 1m vol will be in 1y) - perhaps it's my misunderstanding in your terminology. The former is very relevant as stated my post. I don't see the latter's relevance for a spot-vol market maker. $\endgroup$
    – user35980
    Sep 9 at 9:07
  • $\begingroup$ Okay thank you. You also mentioned "some idea of the term structure of the vols is necessary". Would this be done via calibration ? And what is the term structure needed for specifically ? $\endgroup$
    – FISR
    Sep 9 at 9:43
  • $\begingroup$ Term structure just means how your implied vol varies with shorter dates compared to longer dates (gamma vs vega) - a good starting point can come from your realized vol /time series analysis i.e. using a short rolling window vs longer one...etc $\endgroup$
    – user35980
    Sep 9 at 9:51
  • $\begingroup$ I know what a term structure is, but what is a TS of vol used for ? For example, if I get a term structure of volatility from calibrating to vanilla options for the year ahead, and I want to price more exotic instruments with multiple expiries such as Bermudan option or cliquet options, could I use the volatilities from the term structure on the relevant dates to price these ? Or is the TS for something else ? $\endgroup$
    – FISR
    Sep 9 at 10:09

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge that you have read and understand our privacy policy and code of conduct.

Not the answer you're looking for? Browse other questions tagged or ask your own question.