Sunday, February 24, 2013

Paper: "What Does Individual Option Volatility Smirk Tell Us About Future Equity Returns?"

Quantitative Trading posted a link to a paper about using option volatility smirk as a way to pick longs and shorts called "What Does Individual Option Volatility Smirk Tell Us About Future Equity Returns?" [pdf]. Volatility "smirk" is the difference between the implied volatilities of out-of-the money (OTM) put and at-the-money (ATM) call options.

This is in the same vein as the paper we posted last week, "The Information Content of Option Demand," which showed a relationship between stock returns and option demand imbalances caused by investors with information on the underlying.

The smirk paper finds that by ranking according to volatility smirk, forming a long portfolio consisting of stocks in the bottom quintile and a short portfolio with stocks in the top quintile, and updating weekly, there is an annualized excess return of 9.2%. The theory is that informed traders are buying OTM puts when they anticipate bad news, thereby driving up the implied volatilities of the puts relative to the ATM calls.

What if I told you that the implied vol of a Suntech Power March OTM put is 263 and an ATM call is 145, for a skew of 118? By comparison, the skew for Apple is about 6.

Also, related to the other option demand paper, there is an open interest of ~100k Suntech put contracts for March versus only about 20k call contracts - five times as many.


Alpha Vulture said...

I have an idea that all those papers ignore borrowing costs (or lending income) when calculating Alpha.

CP said...

They typically don't consider borrowing costs, but those are immaterial when you are talking about option skew predicting big declines within a few months.

The procedural quirks of this paper actually detract from alpha.

Alpha Vulture said...

Haven't read the paper, but I do have my doubts. Biggest declines are probably found in the stocks with the highest borrow costs, and these can easily be more than 100% annualized. Paying that a few months isn't immaterial.

CP said...

So, Suntech went to zero (eventually)...

"When there are multiple ATM and OTM options for one stock on one particular day, we further select options or weight all available options using different approaches to come up with one SKEW observation for each firm per day. Our main approach is based on the option's moneyness, which is also used in Ofek, Richardson, and Whitelaw (2004). That is, we choose one ATM call option with its moneyness closest to 1, and one OTM put option with its moneyness closest to 0.95."