Short Sales and Option Listing Decisions

• Published date: 01 September 2014
• Author: Tyler J. Brough,Benjamin M. Blau
• DOI: http://doi.org/10.1111/fima.12039
• Date: 01 September 2014

Short Sales and Option Listing

Decisions

Benjamin M. Blau and Tyler J. Brough∗

We find that stocks with higher levels of prelisting short activity have a greater probability of

option listing. These results are driven by the prelisting short activity of market makers, which

suggests that exchanges believe that stocks with greater short selling will provide option market

makers a better opportunity to hedge with short sales in the spot market. We also confirm that

after options are listed, stockswith more prelisting short activity have more option trading activity.

These results indicate that option exchanges strategicallylist options for stocks they believe with

generate high trading volume therebymaximizing the profits of exchange members.

Unlike the decision to list stocks on an exchange, which is made at the firm level, the decision to

list options is made by exchanges. Mayhew and Mihov (2004) suggest that according to industry

sources, exchanges strategically choose to list options for stocks that they believe will generate

the most option trading volume, thereby maximizing long-term profits for exchange members.

Mayhew and Mihov (2004) examine factors that influence the listing decision and find that

trading volume, volatility, and market capitalization are each positively related to the likelihood

of option listing. Danielsen, Van Ness, and Warr (2007) extend this research by demonstrating

that liquidity improvements also increase the probability of option listing. We examine whether

exchanges choose to list options for stocks with high short selling activity.

The possibility that short selling would influence listing decisions is suggested by two schools

of thought. First, prior research (Figlewski and Webb, 1993; Danielsen and Sorescu, 2001) often

views options and short sales as substitutes insofar as a bearish investor can choose to either short

the stock by borrowing shares or synthetically short the stock by trading a combination of options.

Because of this substitutability between options and short sales, exchanges mayassume that stocks

with greater short activity will have higher levels of option volume as informed investors will

migrate from the shorting market to the options market once options become available.

In addition, liquidity in the options market depends, in part, on the availability of shorting

in the stock market. For instance, option market makers often hedge bearish option activity by

shorting the underlying security (Evans, Geczy, Musto, and Reed, 2009); Battalio and Schultz

(2011) find that in the presence of binding short sale constraints, option market liquidity decreases

dramatically as option market makers are unable to hedge with short sales in the spot market. If

exchanges seek to list options for stocks that will generate the most option trading activity, then

they can choose to list options for stocks with the most shorting activity.In addition to testing the

hypothesis that short activity is an important determinant of the option listing decision, we also

The authors gratefully acknowledge helpful comments from Frank Caliendo, Drew Dahl, Jared Delisle, Chantell Hall,

Kathleen Fuller, Grant McQueen, Mike Pinegar, Jason Smith, Bonnie Van Ness, Robert Van Ness, Mark Van Boening,

Ryan Whitby,and seminar participants at Brigham Young University, the University of Mississippi, Utah State University,

and the 2011 Financial Management Association Meeting in Denver, Colorado. We also recognize and are thankful for

the insightful suggestions from an anonymous referee and Marc Lipson (Editor).

∗Benjamin M. Blau is an Associate Professorand Tyler J.Brough is an Assistant Professor in the Department of Economics

and Finance at the Jon M. Huntsman School of Business, Utah State University,Logan, UT.

Financial Management •Fall 2014 •pages 703 - 724

704 Financial Management rFall 2014

attempt to distinguish which of the motivations described above best influence the likelihood of

option listing.

We use semiparametric methods to identify the determinants of the option listing decision.

For the most part, our tests confirm that volume, volatility, market capitalization, and underlying

stock liquidity are important characteristics that explain the likelihood of option listing. After

controlling for these determinants, we find that stocks with high short activity are more likely to

have options introduced than stocks with low short activity.

We also provide evidence that the decision to list options for stocks with high short activity is

driven by the possibility that these stocks will providebetter hedging oppor tunities for those who

provide liquidity to the options market. In these tests, we partition the short sale data into exempt

short sale activity and nonexempt short sale activity. Exempt short sales are flagged by exchanges

as short sales that are exempt from price tests, such as the NYSE’s uptick rule. Nonexempt

short sales are short sales that are not exempt from price tests. Edwards and Hanley (2010) use

exempt short sales to proxy for short sales executed by market makers, while Engelberg, Reed,

and Ringgenberg (2012) use nonexempt short sales as a proxy for the most informed (or client)

short sales. We find that the direct relation between the probability of option listing and prelisting

short activity is driven by market maker short selling instead of client short selling. These results

suggest that exchanges not only list options for stocks with high prelisting short activity, but

exchanges also choose to list options for stocks that already have high levels of short activity by

market makers.

In our final set of tests, we find a strong direct correlation between prelisting short activity and

option volume during the postlisting period. These results indicate that listed stocks with high

prelisting short activity generate higher levels of option volume. We are only able to provideweak

evidence that the direct relation between prelisting short activity and postlisting option volume

is stronger for market maker short selling than for client short selling. However, in additional

tests, our panel regression analysis indicates that daily put option volume is directly related to the

fraction of short selling that is executed by market makers. The results from these tests support

the conclusions in Battalio and Schultz (2011) that the quality of derivative markets is related to

the strength of the shorting market in the spot asset.

Prior studies (Figlewski and Webb, 1993; Danielsen and Sorescu, 2001) document an increase

in monthly short interest after the introduction of options and argue that option listings mitigate

short sale constraints. Instead, our results seem to suggest that any increase in short activity

around option listings occurs during the month prior to the listing and is endogenous with the

listing decision. Further, stocks with the highest level of prelisting short activity generate the

most postlisting option volume. These results have important implications for the development

of derivative markets. Our findings suggest that the health of these derivatives markets depends,

in part, on the ability to short the underlying spot asset.

I. Prior Research and the Development of Hypotheses

Prior research has examined the determinants of the option listing decision. Jennings and

Starks (1986) posit that option exchanges are likely to list options for stocks based on various

characteristics, such as investor interest, trading activity, and volatility. In the first empirical

study that identifies the determinants of the listing decision, Mayhew and Mihov (2004) estimate

a logistic regression of pooled monthly data consisting of stocks that list options and stocks that

are eligible, but do not list options. They find that stocks with high underlying volume, volatility,

and market capitalization generally have a greater probability of option listing. These results are

consistent with the contention in Jennings and Starks (1986).