Investor Sentiment and Return Predictability of the Option to Stock Volume Ratio

• Author: Jun Sik Kim,Sung Won Seo,Da‐Hea Kim
• DOI: http://doi.org/10.1111/fima.12155
• Published date: 01 September 2017
• Date: 01 September 2017

Investor Sentiment and Return

Predictability of the Option to Stock

Volume Ratio

Jun Sik Kim, Da-Hea Kim, and Sung Won Seo∗

We study the effect of investor sentiment on the relation between the option to stock volume

ratio (O/S) and future stock returns.Relative option volume has return predictability under short

sale constraints. For this reason, we expect and find a stronger O/S-return relation during high

sentiment periods than during low sentiment periods. We find that Baker and Wurgler’s Investor

Sentiment Index affects the O/S-return relation after controlling for consumer sentiment indices

and economic environment factors. While prior studies have used consumer sentiment indices as

alternative measures of investor sentiment, our resultssuggest these effects are distinct.

Investor sentiment is one of the most interesting financial issues among researchers and prac-

titioners. In this regard, Baker and Wurgler’s (2006, 2007) Investor Sentiment Index is designed

to capture the level of market-wide mispricing in the stock market. Baker and Wurgler (2006,

2007) examine the direct relation between their sentiment index and future stock returns. They

also emphasize that time-varying investor sentiment has a different impact on each individual

asset due to the various levels of short sale constraints. In addition, many other studies investigate

the indirect effect of investor sentiment on future stock returns (Yu and Yuan, 2011; Stambaugh,

Yu,and Yuan, 2012; Shen, Yu,and Zhao, 2016). Fur thermore, numerous studies adopt consumer

sentiment indices as alternative measures of investorsentiment (Lemmon and Portniaguina, 2006;

Antoniou, Doukas, and Subrahmanyam, 2013). In this study, we investigate the effect of investor

sentiment on the return predictive power of the option to stock volume ratio (O/S). In addition,

we concentrate on the aspects of Baker and Wurgler’s (2006) (BW hereafter) Investor Sentiment

Index that differentiate it from other consumer sentiment indices. Further, our findings demon-

strate that the BW InvestorSentiment Index may also provideunique information regarding future

stock returns when compared to consumer sentiment indices.

We explore the effect of investor sentiment on the forecasting power of O/S with regard to

future stock returns. Baker and Wurgler (2006, 2007) suggest that their investor sentiment index

captures stock market overvaluation whose occurrence requires two important conditions: 1)

short sale constraints and 2) irrational stock market demand. Short sale constraints directly affect

the predictive power of O/S on future stock returns. Johnson and So (2012) suggest a theoretical

model whereby O/S is negatively related to future stock returns when investors face short sale

constraints. The limitation of shorting stocks also has an important impact on efficient investments

Weare grateful for the helpful comments and suggestions from Suk Joon Byun, Inmoo Lee, and Jun-KooKang. We thank

an anonymous referee and an executive editor for helpful comments. This work was supported by the Incheon National

University Research Grantin 2015.

∗Jun Sik Kim is an Assistant Professor in the Division of International Trade at Incheon National University in Incheon,

Republic of Korea. Da-Hea Kim is in the SKK Business school, Sungkyunkwan Universityin Seoul, Republic of Korea.

Sung WonSeo (Corresponding Author) is an Assistant Professor in the School of Business at Ajou University in Suwon,

Republic of Korea.

Financial Management •Fall 2017 •pages 767 – 796

768 Financial Management rFall 2017

in the option market. This limit to arbitrage makes options attractive. Thus, it is predicted that

investors with negativeprivate information frequently trade options rather than stocks during high

sentiment periods when stock short sale constraints are strong.

In addition, we expect that the predictability of O/S increases whenthere is an ir rational demand

for stocks. As Black (1975) suggests, informed rational investors trade options to utilize the

financial leverage effect of options. Additionally, Easley, O’Hara, and Srinivas (1998) develop a

theoretical model in which options contain privateinformation about future stock price movements

due to informed option trading. Options are attractive as they also provide various payoffs based

on the many kinds of strike prices of call and put options. Consequently, we conjecture that

informed trading in the options market is also triggered by rational investors who wish to take

advantage of irrational demand in the stock market making option implied measures, such as O/S,

informative.1Further, our findings consistently support the hypothesis that O/S has a stronger

relation with future stock returns during high sentiment periods than during low sentiment periods.

Many researchers and practitioners are interested in consumer sentiment indices, such as

the consumer confidence index provided by the Michigan Consumer Sentiment Index (MCSI)

and that of the Conference Board (CB). A consumer sentiment index is related to investor

sentiment as consumers can also be potential investors. Consequently, such an index is used in

prior research as a typical alternative measure of investor sentiment (Lemmon and Portniaguina,

2006; Antoniou et al., 2013). However, consumer sentiment differs from investor sentiment.

First, the two consumer sentiment indices mentioned above are obtained through a survey-based

methodology that involves US households. These households may not participate in the stock

market as investors. Thus, not all consumers are investors. In addition, consumer sentiment

surveys do not ask about investments in the stock market. Consequently, consumer sentiment

may not have a direct connection with investor sentiment. Moreover,consumer sentiment indices

focus on consumer confidence about present and future economic states. They do not consider

the limit to arbitrage. Shleifer and Vishny (1997) emphasize the important role of the limit to

arbitrage in explaining short sale constraints. Thus, consumer sentiment is related to sentiment-

based demand for stocks and contains little information on short sale constraints. As previously

mentioned, both the limit to arbitrage and irrational demand shocks on stocks are key factors

that cause mispricing. Accordingly, consumer sentiment may not have a direct connection with

misvaluation or the BW InvestorSentiment Index. Additionally, respondents’ answers to a survey

may differ from their actual behavior (Baker and Wurgler, 2007). In other words, households

questioned by consumer sentiment surveys may not necessarily invest their capital in the stock

market at the same level they cite in the surveys.

While Johnson and So (2012) focus on the effect of short sale constraints on the predictive

power of O/S, this study investigates the effect of investor sentiment that reflects irrational

demand and short sale constraints. We expect that the predictive power of O/S increases with

irrational demand and short sale constraints. Consistent with our conjecture, we also find that

O/S has greater predictive powerduring high consumer sentiment periods as consumer sentiment

primarily measures the demand for stocks. However, the effect of consumer sentiment is weaker

than that of investor sentiment as the BW Investor Sentiment Index also contains information on

short sale constraints in the stock market. Johnson and So (2012) use a firm-specif ic measure

of short sale constraints, while we deal with market-wide sentiment indices. Thus, we can

concurrently analyze the effect of short sale constraints on each individual stock and on the entire

1The literature that focuses on information provided by options on future stock price movements, such as returns and

return volatility,includes Pan and Poteshman (2006), Bali and Hovakimian (2009), Cremers and Weinbaum(2010), Doran

and Krieger (2010), Xing, Zhang, and Zhao (2010), Byun and Kim (2013), and An et al.(2014).